Code:
sysuse auto, clear gen d = (rep78 > 2) & !missing(rep78)
Sunday, October 31, 2021
gen dummy with missing data
Dear All, Suppose that I run the code
i.e., d-=1, if rep78 > 2 (3,4,5) and d=0 if rep78 <= 2 (1,2). My problem is that I want d =. if rep78 = . (missing). Any suggestions (hopefully, in one line)? Thanks.
Tabulate two way by summarizing the sum
Hi Everyone,
I have the following data on year (first dimension of panel) , market (second dimension of panel) and product code/type (third dimension of panel) along with market share where market is defined as year and market dimension interaction. The market share was generated by :
bysort ye ma: egen tsales=total(qu)
gen share=qu/tsales
where qu is quantity sold of product in a market -year interaction.
Now, org implies the origin country of the product and ma is the market (second dimension of panel) in which the good is sold. What I want is the cross table where columns are destination or ma (Second dimension of panel) and rows as the origin country (location) and each sell to represent how much percentage of sales in a market comes from a certain country. The column sum should be 100. I tried to use the
which did not give me the result. In order to confirm that shares are correct, I checked the following command, and it showed that calculation of share was correct.
I think the tab command giving me number of observations and their average across the years. Rather than summing up the market share of all the goods from the same origin in the given year. Any help is much appreciated.
I have the following data on year (first dimension of panel) , market (second dimension of panel) and product code/type (third dimension of panel) along with market share where market is defined as year and market dimension interaction. The market share was generated by :
bysort ye ma: egen tsales=total(qu)
gen share=qu/tsales
where qu is quantity sold of product in a market -year interaction.
Code:
clear input byte(ye ma) int co byte org float(qu share) 70 1 411 2 1100 .004219555 70 1 241 6 544 .0020867616 70 1 499 7 1880 .007211603 70 1 196 1 8600 .03298925 70 1 413 2 1000 .003835959 70 1 488 2 5400 .02071418 70 1 212 1 7150 .027427107 70 1 134 2 6660 .02554749 70 1 64 3 7800 .02992048 70 1 410 2 2100 .008055514 70 1 500 7 690 .002646812 70 1 400 3 800 .003068767 70 1 439 2 2500 .009589897 70 1 435 3 7200 .027618906 70 1 429 3 4300 .016494624 70 1 36 1 6700 .025700925 70 1 458 3 800 .003068767 70 1 481 2 380 .0014576644 70 1 434 3 1100 .004219555 70 1 217 7 7350 .0281943 70 1 530 7 350 .0013425857 70 1 437 2 7600 .02915329 70 1 269 4 9350 .035866216 70 1 408 2 1350 .005178545 70 1 15 2 2700 .01035709 70 1 412 2 2200 .00843911 70 1 172 2 11000 .04219555 70 1 447 4 560 .002148137 70 1 478 2 150 .00057539385 70 1 455 3 220 .000843911 70 1 402 3 1200 .0046031508 70 1 431 3 2300 .008822706 70 1 430 3 3300 .012658665 70 1 417 1 3700 .01419305 70 1 419 1 1300 .004986747 70 1 26 1 3500 .013425857 70 1 497 1 6350 .02435834 70 1 491 1 6000 .023015754 70 1 174 2 1500 .005753939 70 1 418 1 4650 .01783721 70 1 422 8 7475 .028673794 70 1 521 1 14100 .05408702 70 1 503 7 250 .0009589898 70 1 407 2 2200 .00843911 70 1 406 2 3000 .011507877 70 1 214 1 4450 .017070018 70 1 535 6 2000 .007671918 70 1 524 1 3800 .014576645 70 1 213 1 8000 .030687673 70 1 544 2 5200 .019946987 end
Code:
tab org ma, summarize(share)
which did not give me the result. In order to confirm that shares are correct, I checked the following command, and it showed that calculation of share was correct.
Code:
sum share if ma==1 & ye==71
Assigning colors to stacked bar plot with ordinal y-variable.
Dear Forum,
I would like to plot a stacked bar graph of an "outcome" over hiv status (positive vs. negative) and stratified by sex (male vs. female).
The "outcome" is in fact 3 separate variables ("low", "mid" and "hi") generated from an ordinal yvar with 3 levels ("1" "2" and "3"):
I have the below written stacked box plot command:
- graph bar (sum) cvd_low cvd_mid cvd_hi, over(hiv) over(sex) percent legend(label(1 "Low (<10%)") label(2 "Intermediate (10-30%)") label(3 "High (≥30%)") order(1 - "" 2 - "" 3)) stack
The output is also copied below.
Array
The goal is to have, for example, a navy "HIV negative" bar with different levels of outcome (cvd risk) indicated by different opacities (or shades) of the navy. I will very much appreciate any suggestions you may have on how to go about this.
Thanks in advance.
Itai
I would like to plot a stacked bar graph of an "outcome" over hiv status (positive vs. negative) and stratified by sex (male vs. female).
The "outcome" is in fact 3 separate variables ("low", "mid" and "hi") generated from an ordinal yvar with 3 levels ("1" "2" and "3"):
- low = 1 if yvar==1
- mid = 1 if yvar==2
- hi = 1 if yvar==3
I have the below written stacked box plot command:
- graph bar (sum) cvd_low cvd_mid cvd_hi, over(hiv) over(sex) percent legend(label(1 "Low (<10%)") label(2 "Intermediate (10-30%)") label(3 "High (≥30%)") order(1 - "" 2 - "" 3)) stack
The output is also copied below.
Array
The goal is to have, for example, a navy "HIV negative" bar with different levels of outcome (cvd risk) indicated by different opacities (or shades) of the navy. I will very much appreciate any suggestions you may have on how to go about this.
Thanks in advance.
Itai
egen problem
Hi list,
I want to generate a variable that equals the mean of the values of the first two nonmissing variables in a three-variable list (var1, var2, var3). At the same time, I also want to record which two of the three variables in my list have been used to generate the new variable (I do not need the variable name; the order of the variable is enough). How can I do that efficiently? Thank you very much!
I want to generate a variable that equals the mean of the values of the first two nonmissing variables in a three-variable list (var1, var2, var3). At the same time, I also want to record which two of the three variables in my list have been used to generate the new variable (I do not need the variable name; the order of the variable is enough). How can I do that efficiently? Thank you very much!
Code:
webuse auto,clear replace headroom=. if mpg==17 replace weight=. if mpg==22|mpg==17 //I did the above manipulation so that some observations have only one nonmissing value in the varlist to be considered below //so that you cannot simply apply egen rowmean to fulfil my task keep rep78 headroom weight rename rep78 var1 rename headroom var2 rename weight var3 //how to generate a variable that equals the mean of the first two nonmissing variables in the varlist (var1, var2, var3) //and record which two of the three variables in my list have been used to generate the new variable?
"Encode" a String with another String?
Hello:
I am trying to create a long code list where var1 is a string and what it should represent is shown in var2, also a string.
I then would like to store this coding so that I can refer to it later when var1 shows up, var2 will be displayed in the dataset.
This is in a similar fashion to how I would think -encode- would work for strings to a number, except I want to get another string.
Is there any sort of command available for this?
Or if not, I thought to try to make a loop using a global, but couldn't get the syntax right. I have a few thousand unique strings I'd like to "encode", is there a limit to the number of globals you can have?
Thanks for advice...
I am trying to create a long code list where var1 is a string and what it should represent is shown in var2, also a string.
I then would like to store this coding so that I can refer to it later when var1 shows up, var2 will be displayed in the dataset.
This is in a similar fashion to how I would think -encode- would work for strings to a number, except I want to get another string.
Code:
* clear input str5 var1 str27 var2 "one" "the diameter of the earth" "two" "the diameter of venus" "three" "the diameter of jupiter" "four" "the population of paris" "five" "the surface area of england" "six" "the population of phoenix" end
Or if not, I thought to try to make a loop using a global, but couldn't get the syntax right. I have a few thousand unique strings I'd like to "encode", is there a limit to the number of globals you can have?
Thanks for advice...
Diff in diff: Fix data for one-to-one matching propensity score
I have a data set of companies with the date they were acquired by a business group and a control group in which the companies have been in their business group since their creation. Each company has as many observations as the years between the creation date of the company and 2018 and a dummy variable that is 1 between the acquisition date and 2018. For the control group the dummy variable is 0 in all observations.
I also used one-to-one propensity score matching to form groups within the treatment and the control group considering industry and company creation date. Using the pscore, how can I change the dummy variable for control observations to be 1 in the date their pair in the treatment group was acquired?
Thanks!
I also used one-to-one propensity score matching to form groups within the treatment and the control group considering industry and company creation date. Using the pscore, how can I change the dummy variable for control observations to be 1 in the date their pair in the treatment group was acquired?
Thanks!
New to stata, Impossible problem for a summary table?
Hello all- I'm so glad I found this forum! I've just started working with stata for a biostats class. I've been given a problem set to work on and am already stuck on the first question. The professor has asked us to create a descriptive summary table for continuous data that includes both variable name and variable description (as well as sd, mean, min, max, and # of obs). So far I've tried su, fsum, and univar. Adding label as an optino for fsum gets me the "closest" except there's no "variable label" header in the table. Any ideas?
Maximum number of iterations exceeded
Hello,
I hope you guys are doing well.
I am estimating a model using panel ardl techniques, namely the dynamic fixed effect regressor. And Stata sent this message " Maximum number of iterations exceeded".
I would like to know what it means and how can I overcome this situation?
Thank you so much
I hope you guys are doing well.
I am estimating a model using panel ardl techniques, namely the dynamic fixed effect regressor. And Stata sent this message " Maximum number of iterations exceeded".
I would like to know what it means and how can I overcome this situation?
Thank you so much
Reshape Wide Issue
Hi, I have been trying to reshape my data from long to wide format using reshape wide. The data looks like this:
OrganisationName AssetClass CapitalCommitted TotalCapitalCommitted
Maroni Private Equity 50 120
Susani Private Equity 70 120
Mazia Debt 30 75
Mungo Debt 45 75
Total Capital Committed denotes the total amount of capital committed per asset class by all organisations (I obtained this number by using the command: egen TotalCapitalCommitted = total(CapitalCommitted), by(AssetClass).
I now would like to reshape my data from long to wide format and I have been trying to do so through: reshape wide TotalCapitalCommitted, i(OrganisationName) j(AssetClass) string, however whenever I run this code I get the error message: TotalCapitalCommittedInfrastructure invalid variable name. Would anyone be able to help me? The most important piece of information I have to showcase is the total capital committed per asset class.
Many thanks
OrganisationName AssetClass CapitalCommitted TotalCapitalCommitted
Maroni Private Equity 50 120
Susani Private Equity 70 120
Mazia Debt 30 75
Mungo Debt 45 75
Total Capital Committed denotes the total amount of capital committed per asset class by all organisations (I obtained this number by using the command: egen TotalCapitalCommitted = total(CapitalCommitted), by(AssetClass).
I now would like to reshape my data from long to wide format and I have been trying to do so through: reshape wide TotalCapitalCommitted, i(OrganisationName) j(AssetClass) string, however whenever I run this code I get the error message: TotalCapitalCommittedInfrastructure invalid variable name. Would anyone be able to help me? The most important piece of information I have to showcase is the total capital committed per asset class.
Many thanks
Calculating F-statistic after constrained regression
I have run this unconstrained regression
*Model: lcost=ß_1+ß_2loutput+ ß_3plabor+ ß_4lcapital+ß_5lpfuel
and would like to set the restriction : R = (0, 0, 1, 1, 1) to test hypothesis P labor + βP capital + βP fuel = 1
I defined matrices for R and r
I need to test the null hypothesis of homogeneity based on the F ratio formula using SSR for restricted and unrestricted regression (F-stat=(SSRrestr. − SSRunres./#r)/SSRunres /(n-k)) or based on multiplying the matrices of restriction and coefficients Rß to use the other formula to manually calculate F-stat
How to best do this?
I used this code below but apparently there is an error
*Model: lcost=ß_1+ß_2loutput+ ß_3plabor+ ß_4lcapital+ß_5lpfuel
and would like to set the restriction : R = (0, 0, 1, 1, 1) to test hypothesis P labor + βP capital + βP fuel = 1
I defined matrices for R and r
Code:
matrix R_c = (0,0,1,1,1) gen r=1
How to best do this?
I used this code below but apparently there is an error
Code:
reg lcost loutput lplabor lcapital lpfuel constraint 3 _b[lpfuel]=_b[lplabor]=_b[lcapital]=1 cnsreg lcost loutput lplabor lcapital lpfuel, constraints(3)
How to convert SIC codes into Fama French 12
Hello, I am a uni student and I'm terrible with computers. For a project that is due in a week, we are supposed to convert industry SIC codes (which number in the thousands) into Fama French 12 classifications. I've read a few posts on this forum detailing how to do this, but I'm lost. I'm unfortunately not too proficient at stata. Can anyone give me a step by step explanation? Any help is appreciated. Thank you in advance.
putdocx basic summary statistics
I am running Stata 17 SE.
I would like to send the output of basic summary statistics commands, namely
However, I can't do the usual
presumably because the summary statistics commands do not generate tables, even though their output does look a lot like tables.
For individual elements, the manual suggests
etc
But I am not interested in one element here or there to include in a largely text report...I would like to send to the docx file the whole output...
I would like to send the output of basic summary statistics commands, namely
Code:
summarize
Code:
tabulate
Code:
codebook
Code:
describe
Code:
putdocx table table1=etable
For individual elements, the manual suggests
Code:
putdocx text (" ‘r(mean)’ ")But I am not interested in one element here or there to include in a largely text report...I would like to send to the docx file the whole output...
Reshape dataset including variables names as an additional variable
Hello Statalist community,
I don't have yet much experience using Stata and hope you can help me with a problem I am encountering.
I have a dataset with a time variable (years) and industry variables in which the observations represent the respective industry's beta for each year.
I need to reshape the dataset so that I can have all the betas under the same variable (Ind_Beta) and create an additional variable that identifies the industry, repeating of course the years.
Please find below a simplified example of the current dataset and the desired structure, followed by the dataset example generated by stata dataex.
Any help or hint in the right direction would be greatly appreciated.
Thanks a lot in advance!
Current Dataset Structure:
Desired Structure:
Dataset example generated by stata dataex:
I don't have yet much experience using Stata and hope you can help me with a problem I am encountering.
I have a dataset with a time variable (years) and industry variables in which the observations represent the respective industry's beta for each year.
I need to reshape the dataset so that I can have all the betas under the same variable (Ind_Beta) and create an additional variable that identifies the industry, repeating of course the years.
Please find below a simplified example of the current dataset and the desired structure, followed by the dataset example generated by stata dataex.
Any help or hint in the right direction would be greatly appreciated.
Thanks a lot in advance!
Current Dataset Structure:
| year | Agric | Food | Drug | Books |
| 2000 | 1.12 | 1.45 | 0.97 | … |
| 2001 | 1.05 | 1.6 | 0.88 | … |
| 2002 | 1.18 | 1.34 | 0.92 | … |
| year | Industry | Ind_Beta |
| 2000 | Agric | 1.12 |
| 2000 | Food | 1.45 |
| 2000 | Drug | 0.97 |
| 2000 | … | … |
| 2001 | Agric | 1.05 |
| 2001 | Food | 1.6 |
| 2001 | Drug | 0.88 |
| 2001 | … | … |
| 2002 | Agric | 1.18 |
| 2002 | Food | 1.34 |
| 2002 | Drug | 0.92 |
| 2002 | … | … |
Code:
* Example generated by -dataex-. For more info, type help dataex clear input int year float(agric food beer smoke toys fun books) 2000 .3452645 .13540165 .004693561 .10387945 .5594783 .7765604 .672514 2001 .4666844 .19018325 .22386266 .13462715 .7232649 1.2980112 .6057138 2002 .4180116 .4688401 .4606416 .3165837 .8264946 1.0237317 .7255396 2003 .6058208 .5994597 .54949224 .58115196 .8094561 1.174974 .8105838 2004 .645611 .618094 .5450272 .6523659 1.0529572 1.0226463 .6682123 2005 .7191871 .6432664 .6079998 .8309416 1.2295903 1.1020511 .7842661 2006 .7900746 .5786355 .4802569 .4810273 .9801628 .9165996 .7079894 2007 1.3577225 .7155325 .5886528 .634958 .7687722 1.0139668 .8080173 2008 1.1235923 .6186133 .5526155 .6556639 .8520126 1.213102 1.132914 2009 .6970165 .52631867 .4582001 .3303039 1.0007415 1.5126996 1.307686 2010 .7568682 .5700232 .5300202 .6517677 1.0587844 1.4952796 1.1704426 2011 1.0809941 .6031252 .5268776 .5393996 .980441 1.2700168 1.1069671 2012 1.0136213 .572837 .52827054 .5611398 1.0247171 1.2806007 .9967803 2013 .9104237 .8755165 .7196253 .7702268 .9460964 1.228237 1.176336 2014 .7878072 .7916082 .5841157 .6283697 1.0399202 1.299612 1.1335369 2015 .9148453 .7809891 .7545701 .7404549 1.0749298 1.199249 1.0286957 end
xtunitroot error 'too many variables specified'
Good afternoon.
The dataset looks like this:
In the example appear two country, but I have many more, 25.
My problem is the following.
When I execute:
I receive a message:
too many variables specified
r(103);
It only works if I use one variable:
Is It equivalent if I do one test by variable to do one test with all variables?
What's the problem?
Thank you for any answers and help!
The dataset looks like this:
Code:
* Example generated by -dataex-. For more info, type help dataex clear input long pais float(year ln_co2pc_gr ln_gdppc_gr co2pc_gr gdppc_gr res_share_ch ei_ch) 1 1970 . . . . . . 1 1971 .04318523 . .04413044 . -.6838717 . 1 1972 -.003316879 . -.003311346 . -.5843963 . 1 1973 .035467148 . .036103528 . -.19065896 . 1 1974 -.01744461 . -.017293148 . 1.1816183 . 1 1975 -.04150105 . -.04065147 . -1.1635613 . 1 1976 .02756405 . .027947137 . -.4502255 . 1 1977 .027318 . .02769508 . .4519265 . 1 1978 -.01889515 . -.01871785 . .7232267 . 1 1979 .03856087 . .03931371 . -.03159031 . 1 1980 -.033974648 . -.03340414 . 1.6892276 . 1 1981 -.03340912 . -.032857135 . .3562208 . 1 1982 -.029852867 . -.02941179 . .22335 . 1 1983 .010718346 . .010776284 . .55891395 . 1 1984 -.025982857 . -.02564822 . .3025955 . 1 1985 -.09177828 . -.08769248 . .2845615 . 1 1986 .06215334 . .06412572 . -.57471985 . 1 1987 .035816193 . .0364652 . -.04727695 . 1 1988 .01995754 . .020158285 . -2.0891473 . 1 1989 -.051548 . -.05024241 . -.8556552 . 1 1990 -.08002138 . -.07690334 . 2.5997005 . 1 1991 .04825544 .07338333 .04943905 .07614343 -.7621235 -.003763424 1 1992 .014944077 .06285858 .015055898 .064876065 .8977495 -.004395153 1 1993 .020612717 .06581497 .020826785 .06802921 .9866562 -.005773365 1 1994 .02478409 .04406929 .0250941 .04505472 .05130513 .0004600968 1 1995 -.012306213 -.04115677 -.012230685 -.04032156 .1040608 .003823304 1 1996 .04091835 .04185772 .04176683 .04274599 -.889446 -.0021086514 1 1997 .0079956055 .06639385 .008027056 .06864754 .5504242 -.0032310374 1 1998 .014033318 .02645588 .014132484 .02680954 .14474016 -.0015993596 1 1999 .029364586 -.04557419 .02980039 -.04455157 -1.4123623 .004682073 1 2000 -.013619423 -.018927574 -.013527458 -.018749123 -.7077209 .006655703 1 2001 -.04242325 -.05601025 -.04153631 -.05447092 1.7975198 -.0001920025 1 2002 -.064777374 -.12618446 -.06272339 -.11854777 .4461319 .0078657 1 2003 .05957603 .07396126 .0613862 .07676546 -.9268353 -.00195185 1 2004 .07559109 .0758953 .07852178 .07884938 -1.3884256 .00006687366 1 2005 .02943516 .074453354 .029872544 .07729475 .53251714 -.005813921 1 2006 .04256248 .06724262 .04348141 .06955548 .10701164 -.0033557636 1 2007 .08056164 .07625961 .08389506 .07924238 -1.0392638 -.0004027938 1 2008 .02333355 .02984524 .02360851 .03029477 .4650531 -.0015919253 1 2009 -.07792091 -.07100487 -.07496273 -.06854227 1.7649453 -.0008191558 1 2010 .07104397 .0862999 .07362857 .09013356 .3953756 -.003054402 1 2011 .036629677 .04797363 .03730834 .049143 .19075124 -.001405331 1 2012 .005539894 -.02078247 .0055553 -.020567933 -.5361876 .0016440516 1 2013 .017653465 .013266563 .017810239 .01335466 .006848871 .0002165544 1 2014 -.02623844 -.035855293 -.025896933 -.035220277 .9186995 .0021337224 1 2015 .01380539 .016727448 .013901046 .016867941 -1.0696038 -.001246282 1 2016 -.014710426 -.031025887 -.014602168 -.03054932 .5231693 .001930914 1 2017 -.024267197 .01653099 -.02397574 .016668873 1.1013072 -.0032586295 1 2018 -.04940796 -.034734726 -.04820735 -.03413885 -.1671047 .0007113228 1 2019 -.03939629 -.031279564 -.03863 -.03079537 -.0826373 .00057273713 1 2020 . . . . . . 2 1970 . . . . . . 2 1971 .03735685 . .03806343 . -3.583941 . 2 1972 .006715775 . .006738317 . -3.595976 . 2 1973 .0723815 . .07506522 . 1.0047908 . 2 1974 -.013946533 . -.013849764 . -2.443425 . 2 1975 .06486988 . .067020305 . -5.076143 . 2 1976 .06060696 . .06248108 . 1.480327 . 2 1977 .15053034 . .1624507 . -.9558575 . 2 1978 .05724525 . .05891533 . -5.38981 . 2 1979 .2042713 . .226631 . 1.2600955 . 2 1980 .064723015 . .066863455 . 2.294343 . 2 1981 .0004091263 . .0004090958 . -4.1512866 . 2 1982 -.010876656 . -.010817155 . -1.286386 . 2 1983 -.08180618 . -.07854988 . -.2205604 . 2 1984 .02359295 . .02387344 . .52316314 . 2 1985 -.05776501 . -.05612823 . -.6478212 . 2 1986 .281785 . .3254935 . -.4750064 . 2 1987 .062625885 . .064629 . -5.436717 . 2 1988 .008115768 . .008148801 . .7065676 . 2 1989 .05866051 . .06041504 . -1.0042635 . 2 1990 -.02052593 . -.02031715 . .9601043 . 2 1991 -.0022001266 -.0434866 -.002197118 -.0425546 -.832518 .01618729 2 1992 -.12391949 -.06234837 -.11654889 -.06044444 -1.431724 -.023318866 2 1993 -.013767242 .0041484833 -.01367371 .0041567744 -1.408625 .0024125425 2 1994 .1101904 .016054153 .116491 .016183637 1.988639 .008190635 2 1995 .034041405 .015937805 .03462734 .0160664 -1.8642867 .006239843 2 1996 .0448246 .034890175 .04584408 .035505243 .2397784 .0004650287 2 1997 .12469387 .04205894 .1328022 .04295658 -.6353271 .0079945 2 1998 -.02123642 .032331467 -.021012744 .032859545 .59386986 -.00912828 2 1999 .05735397 -.000790596 .05903048 -.0007899858 -2.49613 .004564094 2 2000 .04701233 .03958511 .04813543 .04037887 2.706013 .004879483 2 2001 -.04536247 -.027560234 -.04434931 -.02718382 -1.0512298 -.00504469 2 2002 -.015072823 .0044546127 -.014959444 .004464366 .05328191 -.002887168 2 2003 .0005598068 .018220901 .00055978925 .018387806 -.10914221 -.002539354 2 2004 .012296677 .010541916 .012372726 .010597872 -.2341628 -.00010549345 2 2005 .017718315 .035025597 .017876115 .035645753 .7751106 -.003184567 2 2006 .008773804 .0531826 .008812081 .05462261 -1.2377497 -.00716155 2 2007 .010412216 .01738453 .010466738 .017535998 .3962357 -.0003483627 2 2008 .15820885 .002524376 .1714113 .002528249 -.927561 .01863219 2 2009 -.035902023 -.05624008 -.03526547 -.05468835 .4229553 .003304608 2 2010 -.07738018 -.02677727 -.07446237 -.02642146 -.8142483 -.00818127 2 2011 .0507679 -.009784698 .05207909 -.0097374 -.3567042 .007837007 2 2012 -.06547451 -.007086754 -.06337695 -.00706133 -.08397305 -.0083234655 2 2013 .010436058 -.016260147 .010490787 -.016129091 -1.083137 .0025542805 2 2014 -.09462738 -.003108978 -.09028804 -.003103939 .20679207 -.01407241 2 2015 -.010979652 .022444725 -.010919874 .022698896 -1.1899234 -.004809058 2 2016 -.06432438 .022831917 -.06229942 .023094524 .5234437 -.009212221 2 2017 .026329994 .00321579 .026679393 .00322063 .5948993 .004054799 2 2018 .01234913 -.007247925 .012426493 -.007221923 .7516962 .000632762 end label values pais pais label def pais 1 "Argentina", modify label def pais 2 "Barbados", modify
My problem is the following.
When I execute:
Code:
xtunitroot ips ln_co2pc_gr ln_gdppc_gr if year > 1990 & year <2020
too many variables specified
r(103);
It only works if I use one variable:
Code:
xtunitroot ips ln_co2pc_gr if year > 1990 & year <2020
What's the problem?
Thank you for any answers and help!
Rename Variables using a loop
Good morning
My problem is the following.
I would like to rename my variables using a loop.
My initial situation is this:
I have a set of variables with this name v1 v2 v3 .... v235
I would like to rename these variables for groups, but assigning each group of variables the same dates, from 2013 to 2019.
So as a final result I would like this:
REVENUES 2013
REVENUES 2014
REVENUES 2015
REVENUES 2016
REVENUES 2017
REVENUES 2018
REVENUES 2019
EBITDA 2013
EBITDA 2014
EBITDA2015
EBITDA 2016
EBITDA 2017
EBITDA2018
EBITDA2019
........
as code I had thought of using this:
local y = 2013
foreach v of var * {
rename (v1-v7) REVENUES`y '
rename (v8-v14) EBITDA`y '
.....
......
local ++ y
}
But unfortunately it doesn't work (I'm a newbie in Stata).
Thank you for any answers and help!
My problem is the following.
I would like to rename my variables using a loop.
My initial situation is this:
I have a set of variables with this name v1 v2 v3 .... v235
I would like to rename these variables for groups, but assigning each group of variables the same dates, from 2013 to 2019.
So as a final result I would like this:
REVENUES 2013
REVENUES 2014
REVENUES 2015
REVENUES 2016
REVENUES 2017
REVENUES 2018
REVENUES 2019
EBITDA 2013
EBITDA 2014
EBITDA2015
EBITDA 2016
EBITDA 2017
EBITDA2018
EBITDA2019
........
as code I had thought of using this:
local y = 2013
foreach v of var * {
rename (v1-v7) REVENUES`y '
rename (v8-v14) EBITDA`y '
.....
......
local ++ y
}
But unfortunately it doesn't work (I'm a newbie in Stata).
Thank you for any answers and help!
Losing a value after dropping duplicates
Hi everyone 
I need a general code to fill the remaining cells at "cvcie0" for the same "gvkey" and for the same "fyear" with the same value.
E.g.: Every row with the gvkey: 1001 and the fyear: 2011 should have the value 73 in the column "cvcie0".
I calculated the value 73 but it will drop after dropping duplicates and I believe the value will not drop if every row has the value.
In my data example, I lose the value in the columns "cvcie0" when dropping duplicates and using the code:
__
quietly by gvkey fyear: gen dup = cond(_N==1,0,_n)
drop if dup>1
__
I hope this data example is fine. As I work with credential data, I needed to build a lil dataset.
And I hope I have used dataex the right way.
Thank you so much!
Best,
Jana
----------------------- copy starting from the next line -----------------------
------------------ copy up to and including the previous line ------------------
Listed 17 out of 17 observations
I need a general code to fill the remaining cells at "cvcie0" for the same "gvkey" and for the same "fyear" with the same value.
E.g.: Every row with the gvkey: 1001 and the fyear: 2011 should have the value 73 in the column "cvcie0".
I calculated the value 73 but it will drop after dropping duplicates and I believe the value will not drop if every row has the value.
In my data example, I lose the value in the columns "cvcie0" when dropping duplicates and using the code:
__
quietly by gvkey fyear: gen dup = cond(_N==1,0,_n)
drop if dup>1
__
I hope this data example is fine. As I work with credential data, I needed to build a lil dataset.
And I hope I have used dataex the right way.
Thank you so much!
Best,
Jana
----------------------- copy starting from the next line -----------------------
Code:
* Example generated by -dataex-. For more info, type help dataex clear input int(gvkey fyear) byte(cvcie1 cvcie0) 1001 2011 55 . 1001 2011 . 73 1001 2011 . 73 1001 2011 . 73 1001 2011 . 73 1001 2011 . 73 1001 2011 55 . 1001 2011 55 . 1001 2011 . 73 1001 2012 64 . 1001 2012 64 . 1001 2012 . 23 1001 2012 . 23 1002 2011 12 . 1002 2011 12 . 1002 2011 . 15 1002 2011 . 15 end
Listed 17 out of 17 observations
How to select optimal GMM model
Greetings,
I am estimating several specifications for the macroeconomic determinants of non-performing loans using system GMM with xtabond2. The literature dopesn't have a theorical model for it, so pretty much all specifications are empirical.
Usually with time series models people select optimal specifications with information criterion like AIC or BIC, but from what I read those methods apply only to maximum likelehood estimations (and I coldn't find a way to apply them with xtabond2). I would like to know if there is a standard way to select a specification between several models with GMM.
best wishes,
Tarek Tuma
I am estimating several specifications for the macroeconomic determinants of non-performing loans using system GMM with xtabond2. The literature dopesn't have a theorical model for it, so pretty much all specifications are empirical.
Usually with time series models people select optimal specifications with information criterion like AIC or BIC, but from what I read those methods apply only to maximum likelehood estimations (and I coldn't find a way to apply them with xtabond2). I would like to know if there is a standard way to select a specification between several models with GMM.
best wishes,
Tarek Tuma
Poisson regression
Hi everyone!
this platform has been a great help for me. I would like to ask questions on Poison regression
Background- i want to estimate the elasticity of income on number of children in the household and income elasticity of years of education. log-log model could give me the required answers but, there are 0's in income variable and other two dependent variable which years of schooling and number of children.
So, I thought of using Poisson GLM.
Questions;
1) Does Poisson mean log-linear form of the regression and If I take log of dependent variable ( number of children/years of education) which has zeros in it, Would that be fine?
2) As independent variable I am using income which has huge values. To lower the magnitude I use log of income. Can I take log of income (independent variable) in Poisson?
3) Can I log transform the income knowing that it has 0 values in few observations.
I really look forward for the help.
Thank you in advance
this platform has been a great help for me. I would like to ask questions on Poison regression
Background- i want to estimate the elasticity of income on number of children in the household and income elasticity of years of education. log-log model could give me the required answers but, there are 0's in income variable and other two dependent variable which years of schooling and number of children.
So, I thought of using Poisson GLM.
Questions;
1) Does Poisson mean log-linear form of the regression and If I take log of dependent variable ( number of children/years of education) which has zeros in it, Would that be fine?
2) As independent variable I am using income which has huge values. To lower the magnitude I use log of income. Can I take log of income (independent variable) in Poisson?
3) Can I log transform the income knowing that it has 0 values in few observations.
I really look forward for the help.
Thank you in advance
Renaming date variables
Hello gang
I'm trying to fix such that my date variable, specifically the one showing month, looks a little "cleaner". Initially I had a date variable that showed the date that an insurance customer set up an insurance offer: "date_insurance_offer". I then used this to make two new variables showing the dates in months and quarters by writing the code:
generate insurance_offer_month = mofd(date_insurance_offer)
format insurance_offer_month %tm
generate insurance_offer_quarter = qofd(date_insurance_offer)
format insurance_offer_quarter %tq
Now the variable showing the quarters is fine, but the variable showing the months is showing 2020m1 for january 2020, 2020m2 for february 2020 etc. What i wanted to do is to make the variable "insurance_offer_month" show the months as they are named. For example 2020m1 would preferably be 2020jan, 2020m2 would be 2020feb etc, or something similar.
Is this possible somehow?
Appreciate any help!
I'm trying to fix such that my date variable, specifically the one showing month, looks a little "cleaner". Initially I had a date variable that showed the date that an insurance customer set up an insurance offer: "date_insurance_offer". I then used this to make two new variables showing the dates in months and quarters by writing the code:
generate insurance_offer_month = mofd(date_insurance_offer)
format insurance_offer_month %tm
generate insurance_offer_quarter = qofd(date_insurance_offer)
format insurance_offer_quarter %tq
Now the variable showing the quarters is fine, but the variable showing the months is showing 2020m1 for january 2020, 2020m2 for february 2020 etc. What i wanted to do is to make the variable "insurance_offer_month" show the months as they are named. For example 2020m1 would preferably be 2020jan, 2020m2 would be 2020feb etc, or something similar.
Is this possible somehow?
Appreciate any help!
Collapse and generate average values
Dear,
I want to collapse the years into periods of, say, two years each. For example, if I have a panel from 1990 to 1995, I want period I = 1990-1991, period II = 1992-1993, and period III = 1993 & 1994. Within each period, I will then calculate the average (mean) or identify the mode for each variable.
How can I do that in Stata?
Let me give an example with the following made-up dataset:
input str10 country year x y
"Indonesia" 1990 24 36
"Indonesia" 1991 28 22
"Korea" 1990 38 27
"Korea" 1991 42 73
"China" 1990 124 458
"China" 1991 12 24
end
I want to collapse 1990 and 1991 into just one period, so the x-value for Indonesia would be (24 + 28)/2 and the corresponding y-value (36 + 22)/2.
Now, imagine I also have the row for years 1992-2021. That means I can't just do "collapse varlist, by (country)". I will first need to create a variable called "period" that takes the value = 1 for 1990 & 1991, = 2 for 1992 & 1993, and so on.
I could do that manually one by one, but there must be a smart way to do this in Stata.
Is it possible to get help?
Best,
I want to collapse the years into periods of, say, two years each. For example, if I have a panel from 1990 to 1995, I want period I = 1990-1991, period II = 1992-1993, and period III = 1993 & 1994. Within each period, I will then calculate the average (mean) or identify the mode for each variable.
How can I do that in Stata?
Let me give an example with the following made-up dataset:
input str10 country year x y
"Indonesia" 1990 24 36
"Indonesia" 1991 28 22
"Korea" 1990 38 27
"Korea" 1991 42 73
"China" 1990 124 458
"China" 1991 12 24
end
I want to collapse 1990 and 1991 into just one period, so the x-value for Indonesia would be (24 + 28)/2 and the corresponding y-value (36 + 22)/2.
Now, imagine I also have the row for years 1992-2021. That means I can't just do "collapse varlist, by (country)". I will first need to create a variable called "period" that takes the value = 1 for 1990 & 1991, = 2 for 1992 & 1993, and so on.
I could do that manually one by one, but there must be a smart way to do this in Stata.
Is it possible to get help?
Best,
Robust Standard Errors how to get value for Wald Chi^2?
Hello everyone!
I am running panel data regressions and am using the 'robust' command as I think my model has heteroskedasticity (I am not sure but when I use the robust command, most of my p-values decrease considerably so I think I should be using 'robust', as it makes my coefficients significant).
Unfortunately, by using 'robust' I do not get any values for Wald Chi(2) and prob>chi(2), which I need.
One last question: I am running a fixed effect model (where governance captures country fixed effects) but my tables indicates "random-effects GLD regression" does this matter?
Here is my model without the robust command, followed with the one using the robust command. Just to put into context, I am using dummy variables for my three countries, and time lags.
I really appreciate your help as I am new with Stata and am a bit overwhelmed with econometric textbooks,
thank you very much.
Fred Nolan.
Array
Array
I am running panel data regressions and am using the 'robust' command as I think my model has heteroskedasticity (I am not sure but when I use the robust command, most of my p-values decrease considerably so I think I should be using 'robust', as it makes my coefficients significant).
Unfortunately, by using 'robust' I do not get any values for Wald Chi(2) and prob>chi(2), which I need.
One last question: I am running a fixed effect model (where governance captures country fixed effects) but my tables indicates "random-effects GLD regression" does this matter?
Here is my model without the robust command, followed with the one using the robust command. Just to put into context, I am using dummy variables for my three countries, and time lags.
I really appreciate your help as I am new with Stata and am a bit overwhelmed with econometric textbooks,
thank you very much.
Fred Nolan.
Array
Array
egen, group
Dear All, Suppose that I have this data set (the original question is here),
The raw data have three variables, A, B, and C (names). If I use
I obtain ID1 variable. However, the desired outcome is ID2 (keep the order/names of C unchanged, and the group number is in an increasing pattern ). Any suggestions? Thanks.
Code:
* Example generated by -dataex-. For more info, type help dataex clear input int(A B) str9 C float(ID1 ID2) 1 2010 "艾一" 2 1 1 2011 "张三" 1 2 1 2012 "张三" 1 2 2 2010 "李四" 3 3 2 2011 "李四" 3 3 3 2012 "车八" 6 4 3 2013 "王五" 5 5 3 2014 "李白" 4 6 end
Code:
egen ID1=group(A C)
How to create an aggregate observational unit by adding values by country and year.
Hello,
I have the following dataset for a number of countries.
I would like to generate an aggregate entry (call it africa under variable "country") which is a sum of all the values of the variable "ca" for all the countries by year.
I cannot figure out how to do it except doing it manually in Excel which is obviously not handy.
Can some help me:
Should I need to clarify my question, please let me know.
I have the following dataset for a number of countries.
I would like to generate an aggregate entry (call it africa under variable "country") which is a sum of all the values of the variable "ca" for all the countries by year.
I cannot figure out how to do it except doing it manually in Excel which is obviously not handy.
Can some help me:
Should I need to clarify my question, please let me know.
Code:
* Example generated by -dataex-. To install: ssc install dataex clear input int year float index str87 country float ca 1980 15 "Algeria" 61248.55 1981 15 "Algeria" 63086 1982 15 "Algeria" 67123.51 1983 15 "Algeria" 70748.18 1984 15 "Algeria" 74710.08 1985 15 "Algeria" 77474.35 1986 15 "Algeria" 77784.24 1987 15 "Algeria" 77239.76 1988 15 "Algeria" 76467.36 1989 15 "Algeria" 79831.92 1980 39 "Benin" 1860.74 1981 39 "Benin" 1965.387 1982 39 "Benin" 2177.307 1983 39 "Benin" 2082.683 1984 39 "Benin" 2247.842 1985 39 "Benin" 2417.076 1986 39 "Benin" 2469.006 1987 39 "Benin" 2431.971 1988 39 "Benin" 2505.768 1989 39 "Benin" 2488.038 1980 1 "Botswana" 1516.087 1981 1 "Botswana" 1694.112 1982 1 "Botswana" 1837.036 1983 1 "Botswana" 2055.612 1984 1 "Botswana" 2335.366 1985 1 "Botswana" 2550.895 1986 1 "Botswana" 2738.062 1987 1 "Botswana" 2963.26 1988 1 "Botswana" 3325.464 1989 1 "Botswana" 4431.94 1980 14 "Burkina Faso" 2192.041 1981 14 "Burkina Faso" 2288.216 1982 14 "Burkina Faso" 2339.263 1983 14 "Burkina Faso" 2311.912 1984 14 "Burkina Faso" 2349.373 1985 14 "Burkina Faso" 2656.291 1986 14 "Burkina Faso" 2867.462 1987 14 "Burkina Faso" 2860.642 1988 14 "Burkina Faso" 3026.523 1989 14 "Burkina Faso" 3091.73 1980 30 "Burundi" 1344.64 1981 30 "Burundi" 1491.376 1982 30 "Burundi" 1484.951 1983 30 "Burundi" 1531.115 1984 30 "Burundi" 1529.591 1985 30 "Burundi" 1708.338 1986 30 "Burundi" 1773.586 1987 30 "Burundi" 1846.2 1988 30 "Burundi" 1936.524 1989 30 "Burundi" 1965.25 1980 2 "Cabo Verde" 278.052 1981 2 "Cabo Verde" 301.547 1982 2 "Cabo Verde" 310.067 1983 2 "Cabo Verde" 339.59 1984 2 "Cabo Verde" 352.435 1985 2 "Cabo Verde" 382.893 1986 2 "Cabo Verde" 393.889 1987 2 "Cabo Verde" 410.86 1988 2 "Cabo Verde" 435.498 1989 2 "Cabo Verde" 460.317 1980 3 "Cameroon" 12794.186 1981 3 "Cameroon" 13726.994 1982 3 "Cameroon" 14749.424 1983 3 "Cameroon" 15988.812 1984 3 "Cameroon" 17275.252 1985 3 "Cameroon" 17681.533 1986 3 "Cameroon" 16752.086 1987 3 "Cameroon" 15813.303 1988 3 "Cameroon" 15216.575 1989 3 "Cameroon" 14737.848 1980 35 "Central African Republic" 1490.16 1981 35 "Central African Republic" 1513.225 1982 35 "Central African Republic" 1540.405 1983 35 "Central African Republic" 1446.484 1984 35 "Central African Republic" 1550.996 1985 35 "Central African Republic" 1602.746 1986 35 "Central African Republic" 1660.092 1987 35 "Central African Republic" 1578.233 1988 35 "Central African Republic" 1611.128 1989 35 "Central African Republic" 1651.01 1980 31 "Comoros" 421.845 1981 31 "Comoros" 438.149 1982 31 "Comoros" 466.119 1983 31 "Comoros" 488.591 1984 31 "Comoros" 508.673 1985 31 "Comoros" 518.847 1986 31 "Comoros" 529.224 1987 31 "Comoros" 537.892 1988 31 "Comoros" 552.348 1989 31 "Comoros" 543.308 1980 16 "Congo" 3243.096 1981 16 "Congo" 3931.224 1982 16 "Congo" 4857.743 1983 16 "Congo" 5130.247 1984 16 "Congo" 5500.578 1985 16 "Congo" 5437.178 1986 16 "Congo" 5062.998 1987 16 "Congo" 5072.587 1988 16 "Congo" 5162.182 1989 16 "Congo" 5253.278 end
Finding Market Shares
Hi all,
I have a data in the following format. I need to calculate the market share where market share would be simply percentage of sales of the brand in the given year and market. For example, A in the table should be Sales of Brand 1 in year 2000 and market 1 to Total sales in year 2000 and market 1. This should be 56 / (56 + 25). Similarly, B should be (34 / (34+54+43) ). Finally, C should be (47/ (47+43)). Any coding help for such problem is much appreciated.
I have a data in the following format. I need to calculate the market share where market share would be simply percentage of sales of the brand in the given year and market. For example, A in the table should be Sales of Brand 1 in year 2000 and market 1 to Total sales in year 2000 and market 1. This should be 56 / (56 + 25). Similarly, B should be (34 / (34+54+43) ). Finally, C should be (47/ (47+43)). Any coding help for such problem is much appreciated.
| Year | Market | Brand | Sales | Market Share |
| 2000 | 1 | 1 | 56 | A |
| 2001 | 1 | 1 | 34 | B |
| 2002 | 1 | 1 | 45 | |
| 2001 | 1 | 2 | 54 | |
| 2002 | 1 | 2 | 46 | |
| 2000 | 1 | 3 | 25 | |
| 2001 | 1 | 3 | 43 | |
| 2002 | 1 | 3 | 41 | |
| 2000 | 2 | 1 | 47 | C |
| 2001 | 2 | 1 | 65 | |
| 2002 | 2 | 1 | 62 | |
| 2001 | 2 | 2 | 21 | |
| 2002 | 2 | 2 | 43 | |
| 2000 | 2 | 3 | 43 |
Saturday, October 30, 2021
group string var with random names
Hi,
I have list of investor name which are not same like "investor group" var. I want to create a variable like investor group from var investor_name. Please suggest me how to do it.
I have list of investor name which are not same like "investor group" var. I want to create a variable like investor group from var investor_name. Please suggest me how to do it.
| Investor_name | Investor_group |
| Blue Ocean | Blue Ocean Partners |
| Blue Ocean Partners | Blue Ocean Partners |
| Blue Ocean Partners LLC | Blue Ocean Partners |
| Breakthrough Energy | Breakthrough Energy |
| Deutsche Bank | Deutsche Bank |
| Goldman | Goldman Sachs |
| Goldman Sachs | Goldman Sachs |
| Goldman Sachs, Inc | Goldman Sachs |
| Google Ventures | |
| J.P. Morgan | JP Morgan |
| JP Morgan | JP Morgan |
| JP Morgan Chase | JP Morgan |
| Kleiner Perkins | Kleiner Perkins |
| Kleiner Perkins Caufield & Byers | Kleiner Perkins |
| Biomet Orthopedics, LLC | Biomet |
| Biomet Spine, LLC | Biomet |
| Biomet Trauma, LLC | Biomet |
| Biomet Sports Medicine, LLC | Biomet |
| BIomet 3i, LLC | Biomet |
| Biomet Microfixation, LLC | Biomet |
| Biomet Biologics, LLC | Biomet |
| Davol Inc. | C. R. Bard |
| Bard Peripheral Vascular, Inc. | C. R. Bard |
| C. R. Bard, Inc. & Subsidiaries | C. R. Bard |
| Bard Access Systems, Inc. | C. R. Bard |
| DePuy Synthes Products LLC | DePuy |
| DePuy Mitek LLC | DePuy |
| DePuy Orthopaedics Inc. | DePuy |
| Synthes USA Products LLC | DePuy |
| DePuy Spine, LLC | DePuy |
ttesti equivalent of oneway ANOVA
Hi all,
I need to compare means of a parameter of 8 groups (independent). I do have SD but I do not have individual data.
I was wondering is there any Stata sytax for oneway ANOVA ?
I very much appreciate help.
With warm regards,
Sateesh
I need to compare means of a parameter of 8 groups (independent). I do have SD but I do not have individual data.
I was wondering is there any Stata sytax for oneway ANOVA ?
I very much appreciate help.
With warm regards,
Sateesh
repeated-measure vs. nested anova problems
I think I want to use Repeated-Measure Anova command, "anova, repeated ()" or "wsanova" commands based on the instruction from this site (https://www.stata.com/support/faqs/s...easures-anova/), but I am unsure if I can use this command with the type of data I have. I am using Stata/MP 14.0 for Mac.
To describe the data, participants from two countries responded to items from three domains (i.e., A, B, and C). So the repeated component here is the domain. We created two conditions (high vs. low) nested within each domain so there were a total of six sub-domains (i.e., A high, A low, B high, B low, C high, and C low). Participants were randomized into either high or low condition within each domain. In other words, participants were not divided into a single condition, and their conditions differed depending on each domain. For instance, some participants answered questions in A-high subdomain, B-low subdomain, and C-high subdomain. I am attaching a portion of the dataset I have. ID is the participant number below.
Could I use the following command to run Repeated-Measure Anova to examine the effect of two between-subject variables (country and condition) and one within-subject variable (domain)? The issue that I am currently having is that each participant could have been assigned to multiple conditions depending on the domain (e.g., participant 3, 5, or 6). If I can't use this command, could you give me some suggestions on what command/analysis I could use?
anova Total Country Condition Country#Condition/ ID | Country#Condition Domain Domain#Country Domain#Condition Domain#Condition#Country, rep (Domain)
**Another option I have been thinking of was using nested anova (two condition nested in each domain), but I am not sure if I could run interaction terms (interactions among Domain#Condition#Country) with nested anova. To sum, what I want to do essentially is to run anova with all three variables (Country, Domain, and Condition) included, and I am not sure what command to use.
I have been spending every waking hour from the past two weeks trying to resolve this so any advice would be truly appreciated. Thank you in advance for any advice!
So
To describe the data, participants from two countries responded to items from three domains (i.e., A, B, and C). So the repeated component here is the domain. We created two conditions (high vs. low) nested within each domain so there were a total of six sub-domains (i.e., A high, A low, B high, B low, C high, and C low). Participants were randomized into either high or low condition within each domain. In other words, participants were not divided into a single condition, and their conditions differed depending on each domain. For instance, some participants answered questions in A-high subdomain, B-low subdomain, and C-high subdomain. I am attaching a portion of the dataset I have. ID is the participant number below.
| ID | Country | Domain | Condition | Total |
| 1 | 1 | 1 | 2 | 2 |
| 1 | 1 | 2 | 2 | 2 |
| 1 | 1 | 3 | 2 | 1.833333254 |
| 2 | 1 | 1 | 1 | 2.416666746 |
| 2 | 1 | 2 | 1 | 1.916666627 |
| 2 | 1 | 3 | 1 | 3.416666746 |
| 3 | 1 | 1 | 2 | 2.916666508 |
| 3 | 1 | 2 | 1 | 2.833333254 |
| 3 | 1 | 3 | 2 | 2.75 |
| 4 | 2 | 1 | 2 | 1.5 |
| 4 | 2 | 2 | 2 | 1.666666746 |
| 4 | 2 | 3 | 2 | 1.166666627 |
| 5 | 2 | 1 | 1 | 2.416666746 |
| 5 | 2 | 2 | 2 | 2.75 |
| 5 | 2 | 3 | 2 | 2.583333254 |
| 6 | 2 | 1 | 1 | 2.333333492 |
| 6 | 2 | 2 | 2 | 3.5 |
| 6 | 2 | 3 | 1 | 2 |
anova Total Country Condition Country#Condition/ ID | Country#Condition Domain Domain#Country Domain#Condition Domain#Condition#Country, rep (Domain)
**Another option I have been thinking of was using nested anova (two condition nested in each domain), but I am not sure if I could run interaction terms (interactions among Domain#Condition#Country) with nested anova. To sum, what I want to do essentially is to run anova with all three variables (Country, Domain, and Condition) included, and I am not sure what command to use.
I have been spending every waking hour from the past two weeks trying to resolve this so any advice would be truly appreciated. Thank you in advance for any advice!
So
Help with Recoding with Multiple Loops
Hello, I have a series of variables that looks like job[i]_sched[y]. [i] varies from 1 to 6. [y] varies from 1997 to 2017, but the last three waves of data are collected biannually, so the years are 2010 2011 2013 2015 2017.
I wanted to recode the data using:
However, when I run this code, it recodes only 6 variables:
How do I get the loop to run through every combination of [i] and [y]?
Please let me know if there is anything I should clarify in this post. I looked to see if I could find a similar post, but I did not.
I wanted to recode the data using:
Code:
local yvars 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 ///
2011 2013 2015 2017
foreach i in 1 2 3 4 5 6 {
local y: word `i' of `yvars' //job number
recode job`i'_sched`y' (-6/-1=.) (2/6 8 = 0)(1 7 9 = 1),gen(job`i'_schedx`y')
}However, when I run this code, it recodes only 6 variables:
Code:
(8734 differences between job1_sched1997 and job1_schedx1997) (8742 differences between job2_sched1998 and job2_schedx1998) (8865 differences between job3_sched1999 and job3_schedx1999) (8913 differences between job4_sched2000 and job4_schedx2000) (8960 differences between job5_sched2001 and job5_schedx2001) (8979 differences between job6_sched2002 and job6_schedx2002)
Please let me know if there is anything I should clarify in this post. I looked to see if I could find a similar post, but I did not.
qq Plots
Hello everbody,
I have estimated parameter alpha for each participant with six different methods. I wanted to plot each alpha of one method with the alpha of another method. When I get the graphs they look weird. Have someone had some qqplots like mine? Or what else could be a good solution for the visualiation of the data?
Thanks.
I have estimated parameter alpha for each participant with six different methods. I wanted to plot each alpha of one method with the alpha of another method. When I get the graphs they look weird. Have someone had some qqplots like mine? Or what else could be a good solution for the visualiation of the data?
Thanks.
svy:logit has more observations than simple logit command
I am working with survey data and use the svy command before running my logit regression as follows: svy: logit Y X1 X2. Running this command my number of observations is 2,774. Running the same command without "svy:" gives me 2,152 observations (these are all the observations that have no missings for the variables I am using). How is that possible?
addplot error after marginsplot
I ran a model (dependent variable = gender) and 3 independent variables – year, position_department_n, and interaction of year and position_department_n. I’d like to show a plot with predicted probabilities with panels for each position_department_n, and add the observed data to the same graph. I can get the first part (predicted probabilities by position_department_n), by adding bydimension(position_department_n) to marginsplot (code below).
However, when I use addplot to add in the observed data, I get an error: "Graph family bygraph_g does not support addplot. r(198);". Is there a way around this to achieve what I am trying to do?
However, when I use addplot to add in the observed data, I get an error: "Graph family bygraph_g does not support addplot. r(198);". Is there a way around this to achieve what I am trying to do?
Code:
. quietly mlogit gender_n c.year##i.position_department_n, rrr vce(cluster person)
. quietly margins i.position_department_n, at(year=(1(1)7))
.
. **WITHOUT ADD PLOT
.
. marginsplot, bydimension(position_department_n) legend(pos(3)) recast(line) plot1opts(lcolor(gs8)) ciopt(color(black%20)) recastci(rarea) ytitle("Propor
> tion") plotd(,label( "F" "M" "U"))
Variables that uniquely identify margins: year position_department_n _outcome
Array
.
. **WITH ADDPLOT
.
. marginsplot, bydimension(position_department_n) legend(pos(3)) recast(line) plot1opts(lcolor(gs8)) ciopt(color(black%20)) recastci(rarea) ytitle("Propor
> tion") plotd(,label( "F" "M" "U")) addplot((scatter prop year if gender_n==4, msymbol(circle) mcolor(black) msize(vsmall)) (scatter prop year if gender_
> n==5, msymbol(circle) mcolor(black) msize(vsmall)) (scatter prop year if gender_n==6, msymbol(circle) mcolor(black) msize(vsmall)))
Variables that uniquely identify margins: year position_department_n _outcome
Graph family bygraph_g does not support addplot.
r(198);svy: mean and sorting estimated means
Greetings, Statalisters.
I'm hoping this is something someone is willing and able to explain to me, since I've failed at figuring it out on my own. I need to sort estimated means, into descending order, per year. I'm using Gallup World Poll, and the estimated means are of an index, by country per year, and respondent level sampling weights need to be a part of the computation of the means to account for survey design, etc.. Since I suspect sharing a dataex of that would make my question more complicated, the following mockup shows the situation and desired outcome. Please see here:
I start with this dataset:
I try this first:
Which produces this:
I try this next:
Which produces a very similar, although slimmer, output:
What I need to somehow produce (using this silly demo example):
Anyone? I realize it looks ridiculous with this demo, but what I have is eleven years of 200 national means that I need to sort from highest to lowest each year. If I could make this demo work, I could make that work too. Thanks in advance for your time.
Cheers,
Erika
Editing to add: I also will need to include a measure of precision of the estimates of these means - i.e. standard error of the mean. Haven't looked at that yet, since I'm stuck on this sorting task, but that's next and probably related.
I'm hoping this is something someone is willing and able to explain to me, since I've failed at figuring it out on my own. I need to sort estimated means, into descending order, per year. I'm using Gallup World Poll, and the estimated means are of an index, by country per year, and respondent level sampling weights need to be a part of the computation of the means to account for survey design, etc.. Since I suspect sharing a dataex of that would make my question more complicated, the following mockup shows the situation and desired outcome. Please see here:
I start with this dataset:
Code:
clear all webuse highschool, clear svyset [pweight=sampwgt] * making a fake year variable set seed 12345 generate rannum = uniform() sort rannum generate year = . lab var year "grad year" drop rannum replace year = 2009 in 1/999 replace year = 2010 in 1000/1999 replace year = 2011 in 2000/2999 replace year = 2012 in 3000/4071 * making a fake outcome variable set seed 54321 generate rannum = uniform() sort rannum generate happy = . lab var happy "happiness index" drop rannum replace happy = 1 in 1/700 replace happy = 2 in 701/2200 replace happy = 3 in 2201/4071 label define happy 1 "unhappy" 2 "neutral" 3 "happy" label values happy happy codebook, compact
Code:
*attempt 1 (using svy:mean with subpop for the if statement restricting year)
svy, subpop(if year==2009): mean happy, over(race) coeflegendCode:
(running mean on estimation sample)
Survey: Mean estimation
Number of strata = 1 Number of obs = 4,071
Number of PSUs = 4,071 Population size = 8,000,000
Subpop. no. obs = 999
Subpop. size = 2,016,463
Design df = 4,070
------------------------------------------------------------------------------
| Mean Legend
-------------+----------------------------------------------------------------
c.happy@race |
White | 2.325196 _b[c.happy@1bn.race]
Black | 2.207406 _b[c.happy@2.race]
Other | 2.30351 _b[c.happy@3.race]
------------------------------------------------------------------------------Code:
*attempt 2 (using weights with arithmetic and stock Stata)
sort race year
by race year: gen meanHappy = sum(happy* sampwgt) / sum(sampwgt)
by race year: replace meanHappy=meanHappy[_N]
tabstat meanHappy if year==2009, statistics(mean) by(race) columns(statistics)Code:
Summary for variables: meanHappy
by categories of: race (1=white, 2=black, 3=other)
race | mean
-------+----------
White | 2.325196
Black | 2.207406
Other | 2.303509
-------+----------
Total | 2.312006
------------------What I need to somehow produce (using this silly demo example):
Code:
* a tabulation that sorts these means in descending order
White | 2.325196
Other | 2.303509
Black | 2.207406Cheers,
Erika
Editing to add: I also will need to include a measure of precision of the estimates of these means - i.e. standard error of the mean. Haven't looked at that yet, since I'm stuck on this sorting task, but that's next and probably related.
Graph with Y axis as string
Hi everyone,
I have data like below: 25 different methods that generate 25 different estiamtion on OR and its 95%CI. I want to generate a plot with the axis as the methods' name so that people can see the results related to the name of the methods (This can be easily done In excel). How can I do it in stata?
Thanks
Chang
I have data like below: 25 different methods that generate 25 different estiamtion on OR and its 95%CI. I want to generate a plot with the axis as the methods' name so that people can see the results related to the name of the methods (This can be easily done In excel). How can I do it in stata?
Code:
__method __or_ __lci_or_ __uci_or_ A .6492 .258189 1.63237 B .682846 .133547 3.49148 C .6492 .10206 4.12953 D .586633 .299436 1.14929 E .594977 .303478 1.16647 F .342782 .134567 .873165 G .119268 .012092 1.17643 H .342782 .016632 7.06464 I .660587 .209265 2.08528 J .660587 .209265 2.08528 K .660587 .209265 2.08528 L .778194 .325669 1.85951 M .711595 .193472 2.61727 N .778194 .185898 3.25763 O .790423 .437463 1.42816 P .018483 .001191 .286725 Q .587435 .298901 1.15449 R .783607 .432965 1.41822 S .601916 .306494 1.18209 T .165793 .012218 2.24971 U .280154 .017255 4.54869 V .790423 .437463 1.42816 W .751405 .499381 1.13062 X .757768 .477562 1.20238 Y .880356 .200964 3.85654
Chang
Changing the value of many observations
Hello
So, i'm working on a pretty big dataset for a Norwegian insurance company, and I wanted to change one variable in regards to its observations, or rather what the value Stata "reads" from it. Initially the observations were string and essentially showed "Expired" for an insurance offer that was no longer valid, "rejected" for a rejected insurance offer (by the company) and "accepted" was just missing values.
I started off using "decode" which made "expired" = 1, rejected = 2 and "accepted" = "." (Just a period). I do not really need the "rejected" observations so i essentially want to make the "expired" = 0 and "accepted" = 1. The variable name is "avsluttet_aarsak" (Which loosely translates to "reason for termination/closure")
I used the code:
sort avsluttet_aarsak
replace avsluttet_aarsak = 0 in 1/762621. --> Which worked for the "expired" observations, but when i wrote:
replace avsluttet_aarsak = 1 in 76262/1500403 --> then all the "accepted" observations became "rejected".
I assume this is because the underlying value that the "rejected" observations have is 1, due to the initial decoding, and this is why that happens.
Does anyone know of an easier and better way of doing this? I'm quite new to Stata so any help/input would be much appreciated
So, i'm working on a pretty big dataset for a Norwegian insurance company, and I wanted to change one variable in regards to its observations, or rather what the value Stata "reads" from it. Initially the observations were string and essentially showed "Expired" for an insurance offer that was no longer valid, "rejected" for a rejected insurance offer (by the company) and "accepted" was just missing values.
I started off using "decode" which made "expired" = 1, rejected = 2 and "accepted" = "." (Just a period). I do not really need the "rejected" observations so i essentially want to make the "expired" = 0 and "accepted" = 1. The variable name is "avsluttet_aarsak" (Which loosely translates to "reason for termination/closure")
I used the code:
sort avsluttet_aarsak
replace avsluttet_aarsak = 0 in 1/762621. --> Which worked for the "expired" observations, but when i wrote:
replace avsluttet_aarsak = 1 in 76262/1500403 --> then all the "accepted" observations became "rejected".
I assume this is because the underlying value that the "rejected" observations have is 1, due to the initial decoding, and this is why that happens.
Does anyone know of an easier and better way of doing this? I'm quite new to Stata so any help/input would be much appreciated
Get the standard error from --exlogistic-- command?
Hi all,
I'm running the exlogistic command for meta-analysis, with id as study id:
How can I get the standard error of the log OR? See the following method cannot work.
Thanks
Chang
I'm running the exlogistic command for meta-analysis, with id as study id:
Code:
exlogistic r id group, binomial(n) group(id)
Code:
Exact logistic regression Number of obs = 576
Group variable: __id Number of groups = 8
Binomial variable: __n Obs per group:
min = 26
avg = 72.0
max = 197
Model score = 2.41637
Pr >= score = 0.1258
---------------------------------------------------------------------------
__r | Odds Ratio Suff. 2*Pr(Suff.) [95% Conf. Interval]
-------------+-------------------------------------------------------------
__group | .5874346 23 0.1664 .279409 1.216778
---------------------------------------------------------------------------Code:
gen se = _se[__group]
Code:
no variables defined r(111);
Chang
Friday, October 29, 2021
multilevel multinomial modelling using the gllamm command in stata
Good day everyone,
I am using the gllamm command in stata 15.0 for multilevel (2-leveled) multinomial logistic regression. The outcome/dependent variable is modern contraceptive use among women and it has 4 sub-categories showing the users of the different types of modern contraceptives [(1)permanent method and Long-acting reversible contraceptive users; (2)medium acting reversible contraceptive users; (3)short-acting reversible contraceptive users], each in contrast to the traditional users/those not using any method at all (this is the reference sub-category). For my study, individual women (level-1) are nested within communities (level-2), and my main independent variable; which is the economic empowerment of women, is also introduced as a random slope. In relation to this, I typed in the following command:
gllamm PLMSorNotnew CWEIcentre, i(ClusterNumber) link(mlogit) family(binomial) adapt nrf(2) eqs(inter slope) pweight(pw)
However, only one random intercept variance, one random slope variance, and thus one covariance were estimated instead of three each.
My understanding is that the single random intercept variance shows the between-community difference in the odd of generally using modern contraceptives, irrespective of the particular type, as opposed to traditional users/those not using any method at all. the single random slope variance, on the other hand, shows the between-community difference in the regression effect of women's economic empowerment on the general use of modern contraceptives, irrespective of the particular type. Please, what is the correct command to produce the three different random intercept variances, the three different random slope variances, and the three different covariances.
Thank you.
I am using the gllamm command in stata 15.0 for multilevel (2-leveled) multinomial logistic regression. The outcome/dependent variable is modern contraceptive use among women and it has 4 sub-categories showing the users of the different types of modern contraceptives [(1)permanent method and Long-acting reversible contraceptive users; (2)medium acting reversible contraceptive users; (3)short-acting reversible contraceptive users], each in contrast to the traditional users/those not using any method at all (this is the reference sub-category). For my study, individual women (level-1) are nested within communities (level-2), and my main independent variable; which is the economic empowerment of women, is also introduced as a random slope. In relation to this, I typed in the following command:
gllamm PLMSorNotnew CWEIcentre, i(ClusterNumber) link(mlogit) family(binomial) adapt nrf(2) eqs(inter slope) pweight(pw)
However, only one random intercept variance, one random slope variance, and thus one covariance were estimated instead of three each.
My understanding is that the single random intercept variance shows the between-community difference in the odd of generally using modern contraceptives, irrespective of the particular type, as opposed to traditional users/those not using any method at all. the single random slope variance, on the other hand, shows the between-community difference in the regression effect of women's economic empowerment on the general use of modern contraceptives, irrespective of the particular type. Please, what is the correct command to produce the three different random intercept variances, the three different random slope variances, and the three different covariances.
Thank you.
Manually producing probabilites after logit
Dear All,
I estimate a logit model and then I need to calculate Prob[Y=1|X], where X is my set of regressors. Obviously, I can use:
after the logit estimation. However, for some specific reasons I need to compute those probabilities "manually". Hence, I have:
The above will give me the same results, as if I use the postestimation command predict probabilities. However, rather then writing down the full list of estimated parameters and regressors, I would like to find out a short way. My attempt was:
My doubts: 1) is the way I tried to generate xb correct? I am not sure about it. If not, how I can modify the calculation? 2) When I enter xb in the last line, I get the error message:
[CODE}
matrix operators that return matrices not allowed in this context
[/CODE]
How can I include the generated values xb?
Thanks in advance for your help.
Dario
I estimate a logit model and then I need to calculate Prob[Y=1|X], where X is my set of regressors. Obviously, I can use:
Code:
predict probabilities
Code:
webuse auto.dta, replace gen dummy=0 replace dummy=1 if price>9500 rename logit dummy logit dummy mpg trunk length gen probabilities2=1/(1+exp(-(_b[mpg]*mpg+_b[trunk]*trunk+_b[length]*length+_b[_cons]*_cons)))
Code:
logit dummy mpg trunk length matrix bmatrix=e(b) mat accum m = mpg trunk length matrix xb=bmatrix*m gen probabilities=1/(1+exp(-(xb)))
[CODE}
matrix operators that return matrices not allowed in this context
[/CODE]
How can I include the generated values xb?
Thanks in advance for your help.
Dario
how many observations per year in panel data
Hi there,
I have a panel dataset. How can I see how many observations there are per year? I see in the data viewer that some years have lots of missing observations, but want to know exactly.
Thank you !
I have a panel dataset. How can I see how many observations there are per year? I see in the data viewer that some years have lots of missing observations, but want to know exactly.
Thank you !
Eliminate a part of a string
Hello there,
I would like to eliminate a part of a string starting from a position recorded in another variable for 2000 characters backward. I can't use dataex as the strings are very long.
To give you an example let's say I want to eliminate the word BB (hence appearing at position 6) from the string text and the 3 characters before
text: AAAA BBC D
Bpos: 6
Wanted: AAC D
Thank you,
Ylenia
I would like to eliminate a part of a string starting from a position recorded in another variable for 2000 characters backward. I can't use dataex as the strings are very long.
To give you an example let's say I want to eliminate the word BB (hence appearing at position 6) from the string text and the 3 characters before
text: AAAA BBC D
Bpos: 6
Wanted: AAC D
Thank you,
Ylenia
How do you de-trend an event study plot?
I have an event study graph that I plot using coefficients from my dependent variable regressed on the leads and lags of treatment. There appears to be slight seasonality in the plot. The data I use is micro-data and I suspect that the source of seasonality could be year of birth or quarter of birth effects. But my event study regression already includes all these fixed effects. Year of birth, as well as quarter of birth fixed effects.
Someone saw the graph and suggested I could detrend the plot (for example subtract the yearly mean from the betas)? Is there a formal way I could do this? Do I transform the data before I run the regression (its microdata, and might be a challenge) or do I transform the beta's post-estimation?
Someone saw the graph and suggested I could detrend the plot (for example subtract the yearly mean from the betas)? Is there a formal way I could do this? Do I transform the data before I run the regression (its microdata, and might be a challenge) or do I transform the beta's post-estimation?
Setting the outcome variable in Cox regression.
Hi,
I am trying to estimate the hazard ratios of the age of first drinking, the goal is to measure the risks of underage drinking (<18 years). The values for age at first drinking are: 0 (never), 12,13, 14...38.
I can run the analysis as it is, but certainly not doing it right because I didn't specify the cut-off age anywhere. I am following a similar study that mentioned: "The outcome variable is the age at first drinking. For teens with no drinking experience at survey time, the age at interview was used to include them in the analysis (i.e., censor under the terminology of survival analysis)."
Please free free to share your insights on this, or link to any material that can teach the method.
Thank you.
I am trying to estimate the hazard ratios of the age of first drinking, the goal is to measure the risks of underage drinking (<18 years). The values for age at first drinking are: 0 (never), 12,13, 14...38.
I can run the analysis as it is, but certainly not doing it right because I didn't specify the cut-off age anywhere. I am following a similar study that mentioned: "The outcome variable is the age at first drinking. For teens with no drinking experience at survey time, the age at interview was used to include them in the analysis (i.e., censor under the terminology of survival analysis)."
Please free free to share your insights on this, or link to any material that can teach the method.
Thank you.
labsize in points generating strange result
Per the request of my publisher, I am going through a set of graphs and specifying text in point sizes. This has worked fine except in one case in which the point specified for xlabels is producing a huge size. Here is the specification I'm using:
xlabel(1 `""1945-" "1950""' 2 `""1951-" "1960""' 3 `""1961-" "1970""'4 `""1971-" "1980""'
5 `""1981-" "1990""' 6 `""1991-" "2000""' 7 `""2001-" "2010""' 8 `""2011-" "2020""',labsize(11pt))
This is the only figure in which I have specified 2-line labels. It works fine if I omit the labsize specification and let it default to the standard size or specify something like *1.5 to increase the size.
Any thoughts regarding what's happening and how to fix it would be appreciated.
xlabel(1 `""1945-" "1950""' 2 `""1951-" "1960""' 3 `""1961-" "1970""'4 `""1971-" "1980""'
5 `""1981-" "1990""' 6 `""1991-" "2000""' 7 `""2001-" "2010""' 8 `""2011-" "2020""',labsize(11pt))
This is the only figure in which I have specified 2-line labels. It works fine if I omit the labsize specification and let it default to the standard size or specify something like *1.5 to increase the size.
Any thoughts regarding what's happening and how to fix it would be appreciated.
Where is the dta file for math scores of pupils in the third and fifth years from different schools in Inner London (Mortimore et al. 1988)
I apologize for asking this question because I assume I am missing something obvious. Many Stata examples use data from Mortimore et al. 1988. But I cannot find the dta file on Stata. I located the mjsps5.dta file for the data but with missing values. Can someone point me to the file without missing values?
Error in dictionary file
I have a dataset in .txt, which I have to translate with a dictionary file. I have written the dictionary file according to instructions from the owners of the dataset. However, there seems to be a problem with one variable (that I know of, but there might be other problems too). The variable STDIND presents with missing observations (about 80% of all observations), even though it should not. If I run the code in R (written in R language) it works, without giving me missing values, so there must be a problem with my Stata code (and I have to use Stata). This is how I wrote the dictionary file:
And this is the code I used to apply the dictionary file (2005_Q2_dict.dct is the dictionary file, sta_2005_2.txt is the dataset in txt):
Is there a problem with the code? Am I doing something wrong, or missing something?
Code:
dictionary {
_column(1) int ANNO %4f "ANNO"
_column(5) int TRIM %1f "TRIM"
_column(6) int REG %2f "REG"
_column(8) int numcff %2f "SG4"
(...)
_column(587) int STDFAM %6f "STDFAM"
_column(593) int STDIND %6f "STDIND"
_column(599) int NN2 %1f "NN2"
_column(600) int RPN2 %1f "RPN2"
_column(601) int TF %2f "TF"
_column(603) int TN2 %1f "TN2"
_column(604) int F0_14 %1f "F0_14"
_column(605) int CP0_7 %1f "CP0_7"
_column(606) int CITTAD %1f "CITTAD"
_column(607) int WAVQUA %1f "WAVQUA"
_column(608) int nasita %1f "SG13"
_column(609) int citita %1f "SG16"
_column(610) int annres %3f "SG18"
_column(613) int NASSES %3f "NASSES"
_column(616) int CITSES %3f "CITSES"
_column(619) int RAPSES %3f "RAPSES"
}Code:
clear
infile using "$PathDict/2005_Q2_dict.dct", using("$Path05Q2/sta_2005_2.txt")Elasticity analysis
Hi all,
I am in urgent need to find the solution of a problem. It would be great if you guys could help
background of the problem:
I have dependent variable- number of children and independent variable- log of household income. I need to find out the income elasticity of number of children with cross sectional data.
Problem:
1. Can we take log of "number of children' to estimate the income elasticity.
2. If i regress lin-log regression model, Can I use the predicted values of slope to derive income elasticity by dividing it with avverage number o children?
I request everyone to please help?
I am in urgent need to find the solution of a problem. It would be great if you guys could help
background of the problem:
I have dependent variable- number of children and independent variable- log of household income. I need to find out the income elasticity of number of children with cross sectional data.
Problem:
1. Can we take log of "number of children' to estimate the income elasticity.
2. If i regress lin-log regression model, Can I use the predicted values of slope to derive income elasticity by dividing it with avverage number o children?
I request everyone to please help?
Assigning random number while using pre-defined observation specific probabilities (Result of an LCA) - using rdiscrete
Hello Statalist,
As a result of an LCA I have three variables containing the probability of each observation to belong to a specific group (prob_1, prob_2, prob_3). The values of all three variables are by definition between 0 and 1.
To account for uncertainty in class assignment ( when using LCA as an independent variable) I want to run a simulation with a specific amount of iterations. In this simulation, the assignment to the class should be random while taking the probabilities (prob_1, prob_2, prob_3) into account.
The following code should demonstrate what I want to do (regarding the assignment of the class)
I my case, probabilities vary across observations. Therefore I would like to substitute the general probabilities by individual probabilities, which are saved in the variables prob_1, prob_2 and prob_3
This however gives me the following error
rdiscrete(): 3253 nonreal found where real required
<istmt>: - function returned error
I guess the error message means that only numbers can be put where I wrote "prob_1" etc.
Therefore I thought of saving individual values in a local and than looping across all observations.
Something like this:
This code gives me the error
sum of the probabilities must be 1
rdiscrete(): 3300 argument out of range
<istmt>: - function returned error
The error message is the reason why I added the display function to the code. The displayed values in my case are:
.3423453
.56572497
.09192975
Which almost perfectly add up to 1.
Can anyone help me with my approach? I'm also open to totally different approaches in case my whole approach was wrong.
Thanks
Jay
As a result of an LCA I have three variables containing the probability of each observation to belong to a specific group (prob_1, prob_2, prob_3). The values of all three variables are by definition between 0 and 1.
To account for uncertainty in class assignment ( when using LCA as an independent variable) I want to run a simulation with a specific amount of iterations. In this simulation, the assignment to the class should be random while taking the probabilities (prob_1, prob_2, prob_3) into account.
The following code should demonstrate what I want to do (regarding the assignment of the class)
Code:
generate wanted = .
mata: st_store(., "wanted", rdiscrete(2100, 1, ("0.5", "0.2", "0.3")))Code:
generate wanted = .
mata: st_store(., "wanted", rdiscrete(2100, 1, ("prob_1", "prob_2", "prob_3")))rdiscrete(): 3253 nonreal found where real required
<istmt>: - function returned error
I guess the error message means that only numbers can be put where I wrote "prob_1" etc.
Therefore I thought of saving individual values in a local and than looping across all observations.
Something like this:
Code:
forvalues i = 1(1)2100 {
local prob_1_local = prob_1[`i']
local prob_2_local = prob_2[`i']
local prob_3_local = prob_3[`i']
display `prob_1_local'
display `prob_2_local'
display `prob_3_local'
mata: st_store(., "wanted", rdiscrete(1, 1, (`prob_1_local', `prob_2_local', `prob_3_local')))
}sum of the probabilities must be 1
rdiscrete(): 3300 argument out of range
<istmt>: - function returned error
The error message is the reason why I added the display function to the code. The displayed values in my case are:
.3423453
.56572497
.09192975
Which almost perfectly add up to 1.
Can anyone help me with my approach? I'm also open to totally different approaches in case my whole approach was wrong.
Thanks
Jay
Sum variables taking into account missing data
Hello,
I have the following dataset:
I would like to sum the number of actors (variable actor_id) by year and gid. My final outcome should be a dataset which include a column with the total number of actors by gid and year. In total, my final dataset should be four columns, gid, year, total number of actors, country_code (it does not change by gid and year, it is time invariant).
I am using the following code:
It is works for me if I do not have missing value. Unfortunately, I have it. For example, in the sample I attach, for gid 62356 and year 1997, there is a missing. I would like to have that the total number of actors by gid 62356 and year 1997 be 0. I am not able to do that.
In my dataset, if one pair year-gid has an actor, there is no any missing for that pair of gid-year pair.
I hope you understand my problem. If not, please let me know. Any suggestion/help/comment to solve it, it is more than welcome.
I have the following dataset:
Code:
* Example generated by -dataex-. To install: ssc install dataex clear input long gid int(country_code year) long actor_id 62356 . 1997 . 62357 . 1997 . 79600 710 2012 2664 79601 710 2015 2664 79601 710 2015 2664 79601 710 1999 2794 79601 710 2013 2664 79601 710 2013 2664 79601 710 1999 2794 79601 710 2013 2664 80317 710 2015 2794 80317 710 2012 2664 80317 710 2017 2664 80317 710 2015 2664 80317 710 2015 2664 80317 710 2012 2535 80317 710 2002 2794 80317 710 2008 2664 80317 710 2015 2664 80317 710 2013 2794 80317 710 2009 2794 80317 710 2014 2794 80317 710 2009 2794 80317 710 2008 2664 80317 710 2017 2794 80317 710 2015 2794 80317 710 2015 2794 80317 710 2015 2794 80317 710 2008 2794 80317 710 2012 2664 80317 710 2015 2774 80317 710 2010 2794 80317 710 2017 2664 80317 710 2015 2664 80317 710 2014 2794 80317 710 2016 2664 80317 710 2016 2664 80317 710 2015 2794 80317 710 2014 2664 80317 710 2015 2794 80317 710 2016 2664 80317 710 2009 2794 80317 710 2012 2664 80317 710 2017 2794 80317 710 1999 2664 80317 710 2013 2794 80317 710 2009 2664 80317 710 2015 2794 80317 710 2012 2664 80317 710 2017 2664 80317 710 2017 2794 80317 710 2016 3181 80317 710 2016 2664 80317 710 2009 2794 80317 710 2012 2535 80317 710 2012 2794 80317 710 2010 2664 80317 710 2012 2664 80317 710 2015 2794 80317 710 2010 2664 80317 710 2015 2664 80317 710 2012 2794 80317 710 2017 2794 80317 710 2015 2664 80317 710 2012 2535 80317 710 2017 2794 80317 710 2012 2794 80317 710 2014 2794 80317 710 2015 2794 80317 710 2017 2794 80317 710 2014 2794 80317 710 2016 2664 80317 710 2015 2794 80317 710 2008 2794 80317 710 2017 2664 80317 710 2015 2794 80317 710 2015 2794 80317 710 2006 2664 80317 710 2012 2664 80317 710 2017 2664 80317 710 2012 2664 80317 710 2017 2664 80317 710 2009 2794 80317 710 2017 2794 80317 710 2001 2794 80317 710 2015 2794 80317 710 2017 2794 80317 710 2012 2664 80317 710 2017 2664 80317 710 2017 2794 80317 710 2017 2664 80317 710 2009 2794 80317 710 2013 2188 80317 710 2014 2664 80317 710 2009 2794 80317 710 2015 2794 80317 710 2015 2664 80317 710 2017 2664 80317 710 2013 2664 80317 710 2012 2664 end
I am using the following code:
Code:
bysort gid year actor_id: gen nactors = _n keep if nactors == 1 egen nactors2 = count(nactors), by(year gid)
In my dataset, if one pair year-gid has an actor, there is no any missing for that pair of gid-year pair.
I hope you understand my problem. If not, please let me know. Any suggestion/help/comment to solve it, it is more than welcome.
Error: <istmt>: 3499 ASREGFMB() not found
Good morning,
I try to run the asreg command and I receive this error. I checked that the installation is correct that I have everything updated in Stata (I am using Stata17) but it does not seem to work.
Also a colleague of mine same settings and everything runs it and it seems to work.
Any suggestion on what I may do wrong?
Thank you.
Best,
Tania
I try to run the asreg command and I receive this error. I checked that the installation is correct that I have everything updated in Stata (I am using Stata17) but it does not seem to work.
Also a colleague of mine same settings and everything runs it and it seems to work.
Any suggestion on what I may do wrong?
Thank you.
Best,
Tania
Cmxtmixlogit discrete choice experiment with choice card blocks
Dear all,
I am encountering a convergence problem with the 'cmxtmixlogit' command, using STATA 16.1 for windows.
I performed a discrete choice experiment with 5 attributes (specified random, hereafter regrouped under $att). I also included an ASC (specified random) to represent the opt-out alternative. My dataset contains 1,476 observations: 82 decision makers (Id_farmer), each evaluating 6 choice cards (card) with 3 alternatives - including the opt-out - (alternative), with cards regrouped under 2 blocks --> a total of 12 cards exist). To analyse the data I wrote:
cmset Id_farmer card alternative
cmxtmixlogit chosen, random(ASC $att)
The model doesn't converge. Could it be related to the fact that each respondent only evaluates 6 out of the 12 cards?
Thanks in advance!
I am encountering a convergence problem with the 'cmxtmixlogit' command, using STATA 16.1 for windows.
I performed a discrete choice experiment with 5 attributes (specified random, hereafter regrouped under $att). I also included an ASC (specified random) to represent the opt-out alternative. My dataset contains 1,476 observations: 82 decision makers (Id_farmer), each evaluating 6 choice cards (card) with 3 alternatives - including the opt-out - (alternative), with cards regrouped under 2 blocks --> a total of 12 cards exist). To analyse the data I wrote:
cmset Id_farmer card alternative
cmxtmixlogit chosen, random(ASC $att)
The model doesn't converge. Could it be related to the fact that each respondent only evaluates 6 out of the 12 cards?
Thanks in advance!
zero-inflated and right-censored count data
Dear all statalists,
Thank you for clicking on my post.
What is the proper way to deal with zero-inflated and right-censored count data?
I have written a histogram of count data that takes integer values from 0 to 30, and found excess zero and bunching at 30.
I have used the zinb command to deal with zero inflation, but there does not seem to be a ul(30) option as implemented in cpoisson.
Please let me know how to deal with this in a practical way.
Thank you for taking the time to read my post.
Thank you for clicking on my post.
What is the proper way to deal with zero-inflated and right-censored count data?
I have written a histogram of count data that takes integer values from 0 to 30, and found excess zero and bunching at 30.
I have used the zinb command to deal with zero inflation, but there does not seem to be a ul(30) option as implemented in cpoisson.
Please let me know how to deal with this in a practical way.
Thank you for taking the time to read my post.
Number of splitvoters from two variables
I need to know the number of split voters in Denmark - those who did not vote for the same party at the general and the local election.
An example:
A person votes for Socialdemokratiet at the general and the local election = not a split voter
A person votes for Socialdemokratiet at the general election and for Venstre at the local election = split voter
Variable 1: Which party did you vote for at the generel election
(Danish parties below)
1. Socialdemokratiet
2. Det Radikale Venstre
3. Det Konservative Folkeparti
4. SF - Socialistisk Folkeparti
5. Kristendemokraterne
6. Liberal Alliance
7. Dansk Folkeparti
8. Venstre
9. Enhedslisten
Variable 2: Which party did you vote for at the local election
(Danish parties below)
1. Socialdemokratiet
2. Det Radikale Venstre
3. Det Konservative Folkeparti
4. SF - Socialistisk Folkeparti
5. Kristendemokraterne
6. Liberal Alliance
7. Dansk Folkeparti
8. Venstre
9. Enhedslisten
How can this be done in Stata? Should i gen. a new variable to make this possible?
I hope someone can help me out here.. :-)
Best regards
Mikkel
An example:
A person votes for Socialdemokratiet at the general and the local election = not a split voter
A person votes for Socialdemokratiet at the general election and for Venstre at the local election = split voter
Variable 1: Which party did you vote for at the generel election
(Danish parties below)
1. Socialdemokratiet
2. Det Radikale Venstre
3. Det Konservative Folkeparti
4. SF - Socialistisk Folkeparti
5. Kristendemokraterne
6. Liberal Alliance
7. Dansk Folkeparti
8. Venstre
9. Enhedslisten
Variable 2: Which party did you vote for at the local election
(Danish parties below)
1. Socialdemokratiet
2. Det Radikale Venstre
3. Det Konservative Folkeparti
4. SF - Socialistisk Folkeparti
5. Kristendemokraterne
6. Liberal Alliance
7. Dansk Folkeparti
8. Venstre
9. Enhedslisten
How can this be done in Stata? Should i gen. a new variable to make this possible?
I hope someone can help me out here.. :-)
Best regards
Mikkel
Thursday, October 28, 2021
Panel data estimatioin
Hello. Need help with panel data estimation.
I am working on panel data with T=96 and N=260. I used Hausman test and it gives FE an appropriate model for estimation as shown
But the hetero and auto test show both problems
In accordance to the above data and results I need suggestions. Should I go for FGLS
or use
with
option? Plus if [CODE][cluster/CODE] then cluster with only Panel ID or Panel ID and Time both?
I am working on panel data with T=96 and N=260. I used Hausman test and it gives FE an appropriate model for estimation as shown
Code:
| Table: Hausman Test | |||
| Ho: difference in coefficients not systematic | |||
chi2(6) = (b-B)'[(V_b-V_B) ^(-1)](b-B) = 251.86 Prob>chi2 = 0.0000 (V_b-V_B is not positive definite)
|
Code:
| Table: Wald test for heteroskedasticity |
| H0: sigma(i)^2 = sigma^2 for all i chi2 (252) = 2.8e+38 Prob>chi2 = 0.0000 |
Code:
| Table:Wooldridge test for autocorrelation |
| H0: no first-order autocorrelation F (1, 244) = 873.172 Prob > F = 0.0000 |
Code:
xtgls
Code:
xtreg
Code:
cluster
converting dates (year and month)
Dear All, Is there a more concise way to go from date to newdate below? Thanks.
Code:
* Example generated by -dataex-. For more info, type help dataex clear input str6 date float newdate "202101" 732 "202102" 733 end format %tm newdate
alternative to keep if for a large list of firm id s
Hi all,
I have a list of firm ids that contains 17,000 firm ids. I have a big dataset which contains over 5 million observations of an even bigger set of firm ids (Dataset A). For dataset A, I want to only keep the firm ids that are mentioned in my list of firm ids (or drop the firm ids that aren't in the list). Can you please help me with this?
Thank you.
I have a list of firm ids that contains 17,000 firm ids. I have a big dataset which contains over 5 million observations of an even bigger set of firm ids (Dataset A). For dataset A, I want to only keep the firm ids that are mentioned in my list of firm ids (or drop the firm ids that aren't in the list). Can you please help me with this?
Thank you.
Why are Frames Useful
Perhaps this is a question for another forum, but I had a more general question about dataframes.
I know the data frame feature was added in Stata (16?). Me personally, I'm old school. I started using Stata (semi-seriously) in my (senior?) year of college, with Stata 15, and now as a 1st semester Ph.D student with Stata 17. Thus, when data frames were added to Stata, I didn't know what they were or why I'd need them.
I know Python and R use dataframes, and I get the basic concept behind them... but all the same, why are they useful? Why would I (or anyone) want to use them, ever? I never got the jazz behind storing multiple datasets into n frames at once. Could anyone a little more.... enlightened, than me talk a little about why people should/would want to use dataframes for serious applied work such that it makes life remarkably easier than Stata pre-data frames?
It isn't that I dislike them or think they're bad, I've just never been in a situation where I thought to myself "Gee, another dataset sure would come in handy right about now." Anyone have any thoughts about this?
I know the data frame feature was added in Stata (16?). Me personally, I'm old school. I started using Stata (semi-seriously) in my (senior?) year of college, with Stata 15, and now as a 1st semester Ph.D student with Stata 17. Thus, when data frames were added to Stata, I didn't know what they were or why I'd need them.
I know Python and R use dataframes, and I get the basic concept behind them... but all the same, why are they useful? Why would I (or anyone) want to use them, ever? I never got the jazz behind storing multiple datasets into n frames at once. Could anyone a little more.... enlightened, than me talk a little about why people should/would want to use dataframes for serious applied work such that it makes life remarkably easier than Stata pre-data frames?
It isn't that I dislike them or think they're bad, I've just never been in a situation where I thought to myself "Gee, another dataset sure would come in handy right about now." Anyone have any thoughts about this?
Reshaping Long Data Issue
I am trying to collapse this data but I have continuously gotten errors regarding the i and j not being unique. Here is a snippit of the dataset:
country variable total
AUS CAP 20,525
AUS CAP 1,206
AUS CAP 1,475
AUS CAP 79,601
AUS CAP 9,928
AUS CAP 553
AUS CAP 1,180
AUS CAP 1,080
...
AUS EMPE 57
AUS EMPE 58
AUS EMPE 31
AUS EMPE 17
AUS EMPE 61
AUS EMPE 50
I am trying to get it to read like this with these variables:
Country CAP EMPE
With the totals for each. If anyone can help, I'd appreciate it. I don't know where to start!
country variable total
AUS CAP 20,525
AUS CAP 1,206
AUS CAP 1,475
AUS CAP 79,601
AUS CAP 9,928
AUS CAP 553
AUS CAP 1,180
AUS CAP 1,080
...
AUS EMPE 57
AUS EMPE 58
AUS EMPE 31
AUS EMPE 17
AUS EMPE 61
AUS EMPE 50
I am trying to get it to read like this with these variables:
Country CAP EMPE
With the totals for each. If anyone can help, I'd appreciate it. I don't know where to start!
Using svy command with graph command
Dear all,
I am trying to use graph in Stata 16 to plot bars of the means of categorical variables (5-point scale) while using the svy command so as to have accurate point estimates.
Here's an example of what I tried to do using a variable "nuclear" that excludes the 6th response category ("don't know").
I had successfully used svyset to calibrate the weights with the rake option:
I also saved the raked weights as "rake_wt":
I know graph has a pweight option, however I get different point estimates when I set pweight equal to my calibrated weights. For example, here's the results from svy: mean:
Versus the graph I get when I set pweight equal to my calibrated weights:
[ATTACH=CONFIG]temp_24523_1635466036082_578[/ATTACH]
Without having to individually save each svy: mean result, is there a way I can graph the accurate point estimates?
Best,
Jason
I am trying to use graph in Stata 16 to plot bars of the means of categorical variables (5-point scale) while using the svy command so as to have accurate point estimates.
Here's an example of what I tried to do using a variable "nuclear" that excludes the 6th response category ("don't know").
Code:
. svy: graph bar nuclear if nuclear!=6 graph is not supported by svy with vce(linearized); see help svy estimation for a list of Stata estimation commands that are supported by svy r(322);
Code:
#delimit ;
svyset [pweight=_one], rake(i.gender2 i.age1 i.region1 i.race3 i.latino i.educ1,
totals(_cons=25127
1.gender2=12236 2.gender2=12891
1.age1=7568 2.age1=8305 3.age1=9254
1.region1=5260 2.region1=4435 3.region1=9499 4.region1=5933
1.race3=18242 2.race3=3191 3.race3=3694
1.latino=20604 2.latino=4523
1.educ1=9944 2.educ1=7748 3.educ1=7435
));
#delimit crCode:
#delimit ;
svycal rake i.gender2 i.age1 i.region1 i.race3 i.latino i.educ1 [pw=_one],
gen(rake_wt) totals(_cons=25127
1.gender2=12236 2.gender2=12891
1.age1=7568 2.age1=8305 3.age1=9254
1.region1=5260 2.region1=4435 3.region1=9499 4.region1=5933
1.race3=18242 2.race3=3191 3.race3=3694
1.latino=20604 2.latino=4523
1.educ1=9944 2.educ1=7748 3.educ1=7435
) ;
#delimit crCode:
. svy: mean nuclear if nuclear!=6
(running mean on estimation sample)
Survey: Mean estimation
Number of strata = 1 Number of obs = 2,836
Number of PSUs = 2,836 Population size = 25,127
Calibration : rake Design df = 2,835
--------------------------------------------------------------
| Linearized
| Mean Std. Err. [95% Conf. Interval]
-------------+------------------------------------------------
nuclear | 3.087027 .0283936 3.031353 3.142701
--------------------------------------------------------------Code:
graph bar nuclear if nuclear!=6 [pw = rake_wt]
Without having to individually save each svy: mean result, is there a way I can graph the accurate point estimates?
Best,
Jason