As part of my master's progam in Accountancy, I am doing a replication study. I was wondering if anyone could help me with the following question.
Let me explain my situation: my data exists of financial data on S&P 500 firms during multiple years. I divided the observations in groups (group2_sic2) of firms with the same year and the same industry code (sic2), unless the group would exist of less than 8 firms. In that case, the observations are allocated to a year-specific "other" group (i.e. all observations in the "other" group are from the same year, but do not have the same industry code).
Using these groups, I regressed total accruals on some variables and saved the residuals as "discretionary accruals" (dac_obsv)
Next, I need to match each firm-year observation with another firm-year observation within that group and the match should happen with the observation that has the closest return on assets (ROA). Afterwards, I will use the dac_obsvof the matched observation to calculate the final step (|dac_obsv - dac_matched|).
Hence, my question: How can I match my observations in such a way that they are matched with the observation in the same group that has the closest ROA?
Below, you can find part of my code up till now:
Code:
// Group industries with at least 8 firms and create an "other" group for otherwise
bysort fyear sic2: egen freq_sic2 = count(sic2)
egen group_sic2 = group (fyear sic2)
forvalues i = 2013/2015 {
replace group_sic2 = `i' if (freq_sic < 8 & fyear == `i')
}
/// Create new group ids with consecutive numbers, so the for-loop will work.
egen group2_sic2 = group (group_sic2)
drop group_sic2
// Regress
gen dac_obsv = 0
forvalues i = 1/31 {
reg ACC lag_at1 change_sale ppegt if group2_sic2 == `i'
predict temp, residuals
replace dac_obsv = temp if temp ~= .
drop temp
}
// Calculate ROA
gen ROA = ni / at
dataex gvkey2 fyear sic2 group2_sic2 dac_obsv ROAAnd my data:
Code:
* Example generated by -dataex-. For more info, type help dataex clear input long gvkey2 double fyear float(sic2 group2_sic2 dac_obsv ROA) 7881 2013 10 31 -.1113315 -.09941851 14590 2013 10 31 .0006080783 .04187607 29173 2013 13 1 -.095701 .013625148 14934 2013 13 1 -.07037049 -.00046645055 1678 2013 13 1 -.06539218 .036212016 120093 2013 13 1 -.0013243398 .05796571 7620 2013 13 1 -.11223312 .064163685 11923 2013 13 1 -.08850802 .014359728 9465 2013 13 1 .010281556 .10032787 5439 2013 13 1 -.02721703 .0727167 20548 2013 13 1 -.10563926 .05616714 7017 2013 13 1 -.04868451 .04921392 14359 2013 13 1 -.15535808 -.06820373 8068 2013 13 1 -.04339208 .08500497 6788 2013 13 1 -.03905627 .015854314 16478 2013 13 1 -.1132986 .071861446 8549 2013 13 1 -.008366337 .07755575 7912 2013 13 1 -.05129356 .04979127 1380 2013 13 1 .02779204 .11816438 11228 2013 14 31 .007283384 .0029521224 8823 2013 15 31 .2458316 .2999855 4818 2013 16 31 .032945774 .0802166 66446 2013 17 31 .036919776 .06937753 3144 2013 20 2 .03234896 .09531952 3505 2013 20 2 .01196952 .036411833 5597 2013 20 2 -.0247544 .15314454 180833 2013 20 2 -.006113946 .1869549 24316 2013 20 2 .00552455 .2384082 142953 2013 20 2 -.002784396 .05395758 6375 2013 20 2 .05089426 .11677653 8479 2013 20 2 .016736088 .08699244 179621 2013 21 31 .011967474 .22469084 120877 2013 21 31 .07718544 .11154395 8543 2013 21 31 .05686508 .13009553 25119 2013 22 31 .02828753 .04106178 11456 2013 24 31 .020500677 .03883294 6649 2013 25 31 -.034171026 .0634793 6104 2013 26 31 .00071028 .04424638 7435 2013 26 31 .018824205 .13886736 6435 2013 26 31 .004452804 .11321951 7257 2013 28 3 -.014082456 .04168678 24344 2013 28 3 -.1344164 -.19190174 4213 2013 28 3 .02112188 .04928577 8530 2013 28 3 .017168332 .12784934 62263 2013 28 3 -.0380769 .0762261 2086 2013 28 3 .006112917 .0777765 163946 2013 28 3 .04844807 .13715923 4510 2013 28 3 .06639741 .05613921 23812 2013 28 3 -.022438044 .14380215 6078 2013 28 3 .02262517 .1061142 24468 2013 28 3 .008688239 .15698293 9667 2013 28 3 -.00557985 .11790994 6730 2013 28 3 .023087414 .13290703 8247 2013 28 3 .0029861694 .20368153 4060 2013 28 3 .012006408 .068876706 13721 2013 28 3 .01897236 .0768527 3170 2013 28 3 -.01906245 .16150187 16101 2013 28 3 -.021238964 .14137955 1602 2013 28 3 .03494855 .07683932 294524 2013 28 3 .016154755 .14129241 24856 2013 28 3 .050183 .13667767 2403 2013 28 3 .02687978 .066412725 29392 2013 28 3 .0391585 .09835374 13599 2013 28 3 -.008314132 .1083778 15247 2013 29 31 -.006080544 .05755396 2991 2013 29 31 .015581252 .08442461 186989 2013 29 31 .010419308 .0744055 170841 2013 29 31 .005373346 .07482228 10466 2013 29 31 .016248802 .03077153 4503 2013 29 31 .0408916 .09394247 5234 2013 30 31 .033350963 .035887487 9555 2013 30 31 -.01037242 .01359725 8215 2013 32 31 -.008930976 .021855326 8030 2013 33 31 .011647358 .032099973 1988 2013 34 31 -.008292693 .05202179 7085 2013 34 31 -.005523456 .03923266 316056 2013 34 31 -.06778742 .015657356 5878 2013 35 4 .005155102 .08409296 4108 2013 35 4 .04319837 .0963978 4058 2013 35 4 .031799246 .09255518 8463 2013 35 4 .018515082 .04571117 3650 2013 35 4 .008707452 .10069256 2817 2013 35 4 -.021517944 .04463108 121718 2013 35 4 .009722793 .04259152 189491 2013 35 4 .02357459 .04656863 60894 2013 35 4 .04000037 .04907183 8606 2013 35 4 -.0011881363 .02108979 12053 2013 35 4 -.035254132 .06301119 63892 2013 35 4 .02177991 .06684477 3532 2013 36 5 .023636306 .06886017 14282 2013 36 5 .022842906 .1030592 175404 2013 36 5 -.027236136 .05128756 4199 2013 36 5 .04134171 .05243583 10499 2013 36 5 -.01335562 .114162 7585 2013 36 5 .062775776 .0927348 11465 2013 36 5 .02295861 .05320381 6008 2013 36 5 -.06404007 .1041599 10983 2013 37 6 .04096874 .06314988 118122 2013 37 6 .0006286372 .10971305 5073 2013 37 6 .01134167 .03213822 end
For your information, I am using Stata 16.1.
I hope this gives you enough information about my question. Please let me know if you can help me out!
And thank you in advance.
Kind regards,
Lianne
0 Response to Question:
Post a Comment