I have a data set over a period 2008-2017. I estimated a gmm model based on this time period to obtain own and cross acreage and price elasticies. The dependent variable is the log of the acreage of the crop. In the next step, I need to predict the log of the acreage for year 2018. To do this I used the average of the value of each variable during 2008-2017 for the year that I need to predict the log of acreage except price. For the price I used the half of the minimum of the observed price during 2008-2017. In this case I am trying to simulate a shock in the economy that drops the price of the crop. I used the following command:
generate logacr=log(acr)
generate logacrl=log(acrl)
generate logpl=log(pl)
generate logpi=log(pi)
xtabond2 logacr l.logacr l.logacrl logpl fzi fui logpi gdd pop ppt year dev if year<11, gmm(logacr ,collapse) iv(gdd logpl fzi fui ppt year dev) small robust
predict logacr_hat if year>10
So, I want to predict logacrl for year 11. Now I think since I am predicting linearly and am using the half of the minimum price (pl), the predicted acreage that I get should be less than any other historically observed acreage duruing 2008-2017. But the results dont's show this.
Am I wrong? the results don't show what I expecetd. Can anyone help me on this please. Hope I am clear enough.
Thank you.
The results are as following:
Dynamic panel-data estimation, one-step system GMM
------------------------------------------------------------------------------
Group variable: coun_num Number of obs = 108
Time variable : year Number of groups = 12
Number of instruments = 31 Obs per group: min = 9
F(9, 11) = 258.29 avg = 9.00
Prob > F = 0.000 max = 9
------------------------------------------------------------------------------
| Robust
logacrl | Coef. Std. Err. t P>|t| [95% Conf. Interval]
-------------+----------------------------------------------------------------
logacrl |
L1. | .230327 .1489105 1.55 0.150 -.0974228 .5580768
|
logpl | .1325983 .4238889 0.31 0.760 -.8003748 1.065571
|
logacr |
L1. | .5866026 .3187713 1.84 0.093 -.1150084 1.288214
|
logpi | -1.409517 .3515745 -4.01 0.002 -2.183327 -.6357066
fzi | -.0019445 .0012559 -1.55 0.150 -.0047088 .0008197
gdd | .025999 .0348142 0.75 0.471 -.0506265 .1026244
ppt | -.0027826 .0082187 -0.34 0.741 -.0208718 .0153066
pop | -.0004244 .000487 -0.87 0.402 -.0014962 .0006474
year | .025035 .0289901 0.86 0.406 -.0387717 .0888417
_cons | 6.875543 2.696529 2.55 0.027 .9405235 12.81056
------------------------------------------------------------------------------
Instruments for first differences equation
Standard
D.(gdd pop gdd fui year)
GMM-type (missing=0, separate instruments for each period unless collapsed)
L(1/10).(logacrl logpl fzi) collapsed
Instruments for levels equation
Standard
gdd pop gdd fui year
_cons
GMM-type (missing=0, separate instruments for each period unless collapsed)
D.(logacrl logpl fzi) collapsed
------------------------------------------------------------------------------
Arellano-Bond test for AR(1) in first differences: z = -1.70 Pr > z = 0.089
Arellano-Bond test for AR(2) in first differences: z = 1.28 Pr > z = 0.201
------------------------------------------------------------------------------
Sargan test of overid. restrictions: chi2(21) = 61.99 Prob > chi2 = 0.000
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(21) = 0.93 Prob > chi2 = 1.000
And the the data (historical and predicted) looks like this:
county | coun_num | year | acrl | acr | pl | pi | ppt | area | pop | tmin | tmean | gdd | tmax | fui | fzi | dev | logacr | logacrl | logpl | logpi | logacr_hat | exp_logacr_hat | logacrl_hat | exp_logacrl_hat |
San-Diego | 8 | 1 | 580 | 4293 | 86.07 | 115.25 | 22.14 | 4,206.63 | 725.3455 | 48.9 | 62.9 | 22.9 | 76.9 | 267.32 | 355.82 | 28 | 8.364741 | 6.363028 | 4.455161 | 4.747104 | ||||
San-Diego | 8 | 2 | 580 | 4650 | 66.83 | 128.52 | 12.76 | 4,206.63 | 731.613 | 48.6 | 62.4 | 22.4 | 76.2 | 177.75 | 250.15 | 27.6 | 8.444622 | 6.363028 | 4.202152 | 4.856084 | ||||
San-Diego | 8 | 3 | 575 | 4184 | 68.27 | 122.8 | 32.79 | 4,206.63 | 737.1859 | 47 | 60.2 | 20.2 | 73.3 | 220.56 | 229.25 | 26.3 | 8.339023 | 6.35437 | 4.22347 | 4.810557 | ||||
San-Diego | 8 | 4 | 575 | 4355 | 55.69 | 130.66 | 17 | 4,206.63 | 745.4806 | 46.9 | 60.4 | 20.4 | 73.8 | 281.69 | 298.51 | 26.9 | 8.37908 | 6.35437 | 4.019801 | 4.872599 | ||||
San-Diego | 8 | 5 | 517 | 3334 | 115.81 | 317.9 | 13.88 | 4,206.63 | 754.5223 | 48.6 | 62.5 | 22.5 | 76.4 | 279.72 | 302.69 | 27.8 | 8.111928 | 6.248043 | 4.751951 | 5.761737 | ||||
San-Diego | 8 | 6 | 84 | 3420 | 85.75 | 309.47 | 9.76 | 4,206.63 | 763.7608 | 48.3 | 61.7 | 21.7 | 75.1 | 277.18 | 288.66 | 26.8 | 8.137396 | 4.430817 | 4.451436 | 5.734861 | ||||
San-Diego | 8 | 7 | 580 | 2860 | 76.45 | 301.03 | 11.96 | 4,206.63 | 772.4171 | 50.3 | 63.7 | 23.7 | 77.2 | 275.77 | 282.09 | 26.9 | 7.958577 | 6.363028 | 4.336637 | 5.70721 | ||||
San-Diego | 8 | 8 | 577.5 | 2050 | 67.55 | 304.54 | 17.74 | 4,206.63 | 778.7462 | 49.9 | 62.5 | 22.5 | 75.1 | 178.87 | 260.3 | 25.2 | 7.625595 | 6.358708 | 4.212868 | 5.718802 | ||||
San-Diego | 8 | 9 | 575 | 2132 | 100.78 | 276.86 | 19.68 | 4,206.63 | 783.8108 | 49.6 | 62.6 | 22.6 | 75.6 | 156.62 | 214.33 | 26 | 7.664816 | 6.35437 | 4.61294 | 5.623512 | ||||
San-Diego | 8 | 10 | 576.25 | 1819 | 81.1 | 323.42 | 19.63 | 4,206.63 | 789.3223 | 50.3 | 63.2 | 23.2 | 76.1 | 178.03 | 197.61 | 25.8 | 7.506042 | 6.356542 | 4.395683 | 5.778952 | ||||
San-Diego | 8 | 11 | 521.975 | 3309.7 | 27.845 | 233.045 | 17.734 | 4206.63 | 758.2204 | 48.84 | 62.21 | 22.21 | 75.57 | 229.351 | 267.941 | 26.73 | 8.104612 | 6.25762 | 3.326653 | 5.451231 | 9.503097 | 13401.16 | 5.460869 | 235.3018 |
San-Luis | 9 | 1 | 7588 | 31260 | 126.12 | 124.63 | 21.18 | 3,298.56 | 80.89681 | 39.8 | 57.7 | 17.7 | 75.5 | 267.32 | 355.82 | 35.7 | 10.35009 | 8.934323 | 4.837234 | 4.825349 | ||||
San-Luis | 9 | 2 | 7475 | 24844 | 169.68 | 147.61 | 17.87 | 3,298.56 | 81.3167 | 40.3 | 57.7 | 17.7 | 75.1 | 177.75 | 250.15 | 34.8 | 10.12037 | 8.919319 | 5.133914 | 4.994574 | ||||
San-Luis | 9 | 3 | 4590 | 22213 | 113.45 | 113.41 | 33.67 | 3,298.56 | 81.56408 | 40.7 | 56.7 | 16.7 | 72.8 | 220.56 | 229.25 | 32.1 | 10.00843 | 8.431635 | 4.731362 | 4.731009 | ||||
San-Luis | 9 | 4 | 5140 | 18795 | 106.8 | 110.76 | 21.61 | 3,298.56 | 81.97547 | 40.5 | 56.8 | 16.8 | 73 | 281.69 | 298.51 | 32.5 | 9.841346 | 8.544808 | 4.670958 | 4.707366 | ||||
San-Luis | 9 | 5 | 3650 | 16501 | 104.42 | 113.63 | 18.11 | 3,298.56 | 82.65516 | 43.3 | 60.5 | 20.5 | 77.7 | 279.72 | 302.69 | 34.4 | 9.711176 | 8.202482 | 4.648421 | 4.732947 | ||||
San-Luis | 9 | 6 | 3970 | 17836 | 112.35 | 141.06 | 4.36 | 3,298.56 | 83.06413 | 42.4 | 60.2 | 20.2 | 77.9 | 277.18 | 288.66 | 35.5 | 9.788974 | 8.286521 | 4.721619 | 4.949185 | ||||
San-Luis | 9 | 7 | 4471 | 17089 | 130.56 | 126.22 | 15.43 | 3,298.56 | 83.75685 | 45.6 | 62.7 | 22.7 | 79.9 | 275.77 | 282.09 | 34.3 | 9.74619 | 8.405368 | 4.871833 | 4.838027 | ||||
San-Luis | 9 | 8 | 5140 | 16256 | 156.95 | 179.98 | 9.64 | 3,298.56 | 83.93329 | 43.2 | 60.5 | 20.5 | 77.8 | 178.87 | 260.3 | 34.6 | 9.696218 | 8.544808 | 5.055927 | 5.192846 | ||||
San-Luis | 9 | 9 | 4090 | 15036 | 133.82 | 133.26 | 20.19 | 3,298.56 | 84.38955 | 41.8 | 59 | 19 | 76.1 | 156.62 | 214.33 | 34.3 | 9.618202 | 8.3163 | 4.896496 | 4.892302 | ||||
San-Luis | 9 | 10 | 3540 | 13194 | 156.15 | 174.58 | 28.54 | 3,298.56 | 84.48535 | 42.4 | 59.4 | 19.4 | 76.4 | 178.03 | 197.61 | 34 | 9.487517 | 8.171882 | 5.050817 | 5.162383 | ||||
San-Luis | 9 | 11 | 4965.4 | 19302.4 | 52.21 | 136.514 | 19.06 | 3,298.56 | 82.80374 | 42 | 59.12 | 19.12 | 76.22 | 229.351 | 267.941 | 34.22 | 9.867985 | 8.510249 | 3.955274 | 4.916427 | 10.63043 | 41375.08 | 8.081109 | 3232.817 |
0 Response to Predict an out-of-sample value in Stata
Post a Comment