Dear users of Statalist,
I have 2*5 double sorted portfolios and I would like to test the significance of the constant terms between two portfolios.
What I've done so far is the following:
For now, I have alphas of portfolios formed based on social scores in the sample of low institutional ownership and high institutional ownership.I would like to compute and test the significance of the difference between alphas(the constant terms of regressions) of the portfolio P15 and portfolio P11 and the difference between alphas of the portfolio P25 and portfolio P21.More specifically, in each sample (firms of low institutional ownership or high institutional ownership, I would like to compare and test the significance of alphas of portfolio with highest social scores and alphas of portfolio with lowest social scores.
I would like to also test the significance of the difference between the spreads(the difference between alpha P15-alpha P11 and alpha P25-alpha P21).
In short, I seek to do the same thing as in the red rectangle in the table below (robust Newey-West (1987) t-statistics are reported in parentheses):
I tried my best to state my problem as clearly as possible and I'd like to thank you in advance for any valuable advice and help.
Here is an extract of my current data:
I have 2*5 double sorted portfolios and I would like to test the significance of the constant terms between two portfolios.
What I've done so far is the following:
Code:
// at each quarter: sort stocks into two portfolios based on their institutional ownership
by rdate, sort: egen io_group = xtile(ownership), nq(2)
// at each quarter: sort stocks in low institutional ownership group and high institutional ownership group respectively into 5 groups based on their social scores
by rdate io_group, sort: egen score_quintile = xtile(socialscore), nq(5)
// compute the equally-weighted returns of portfolios formed on the quintiles of social scores at each quarter
sort rdate io_group score_quintile
by rdate io_group score_quintile: egen pret=mean(excessreturn)
// we keep only one observation per quarter-institutional ownership-social score quintile
duplicates drop rdate io_group score_quintile, force
// compute the CAPM alphas for each portfolio: portfolio P11 refers to the portfolio of stocks of low institutional ownership and lowest quintile of social scores, P25 refers to the portfolio of stocks of high institutional ownership and highest quintile of social scores
gen alpha = .
forvalues i = 1/2 {
forvalues j = 1/5 {
reg pret mktrf if (io_group==`i') & (score_quintile==`j') //mktrf is the market excess return
outreg2 using table3, excel bdec(3) stats(coef tstat) tdec(2) nonotes
replace alpha = _b[_cons] if io_group == `i' & score_quintile ==`j'
}
}
For now, I have alphas of portfolios formed based on social scores in the sample of low institutional ownership and high institutional ownership.I would like to compute and test the significance of the difference between alphas(the constant terms of regressions) of the portfolio P15 and portfolio P11 and the difference between alphas of the portfolio P25 and portfolio P21.More specifically, in each sample (firms of low institutional ownership or high institutional ownership, I would like to compare and test the significance of alphas of portfolio with highest social scores and alphas of portfolio with lowest social scores.
I would like to also test the significance of the difference between the spreads(the difference between alpha P15-alpha P11 and alpha P25-alpha P21).
In short, I seek to do the same thing as in the red rectangle in the table below (robust Newey-West (1987) t-statistics are reported in parentheses):
I tried my best to state my problem as clearly as possible and I'd like to thank you in advance for any valuable advice and help.
Here is an extract of my current data:
Code:
* Example generated by -dataex-. For more info, type help dataex clear input double(rdate mktrf) float(io_group score_quintile pret alpha) 15795 -1.65 1 1 . .0505924 15795 -1.65 1 2 . .030939957 15795 -1.65 1 3 . .03088566 15795 -1.65 1 4 . .04419843 15795 -1.65 1 5 . .02381743 15795 -1.65 1 . . . 15795 -1.65 2 1 . .03029218 15795 -1.65 2 2 . .029775864 15795 -1.65 2 3 . .02672335 15795 -1.65 2 4 . .025443764 15795 -1.65 2 5 . .026201123 15795 -1.65 2 . . . 15795 -1.65 . . . . 15886 -.17 1 1 .3435377 .0505924 15886 -.17 1 2 .08502242 .030939957 15886 -.17 1 3 .19393073 .03088566 15886 -.17 1 4 .12327705 .04419843 15886 -.17 1 5 .13493845 .02381743 15886 -.17 1 . .22667165 . 15886 -.17 2 1 .2026721 .03029218 15886 -.17 2 2 .22140093 .029775864 15886 -.17 2 3 .13633959 .02672335 15886 -.17 2 4 .16820045 .025443764 15886 -.17 2 5 .19008897 .026201123 15886 -.17 2 . .21229225 . 15886 -.17 . . .1177743 . 15978 -.96 1 1 .013684333 .0505924 15978 -.96 1 2 .06183974 .030939957 15978 -.96 1 3 -.011653104 .03088566 15978 -.96 1 4 .0311849 .04419843 15978 -.96 1 5 .05043896 .02381743 15978 -.96 1 . .12082233 . 15978 -.96 2 1 .04019295 .03029218 15978 -.96 2 2 .0311509 .029775864 15978 -.96 2 3 .04290957 .02672335 15978 -.96 2 4 .008771131 .025443764 15978 -.96 2 5 .05210209 .026201123 15978 -.96 2 . .08380613 . 15978 -.96 . . .02712074 . 16070 .01 1 1 .19408153 .0505924 16070 .01 1 2 .07767887 .030939957 16070 .01 1 3 .08270334 .03088566 16070 .01 1 4 .10934102 .04419843 16070 .01 1 5 .12344606 .02381743 16070 .01 1 . .15927666 . 16070 .01 2 1 .13614993 .03029218 16070 .01 2 2 .174243 .029775864 16070 .01 2 3 .1499734 .02672335 16070 .01 2 4 .12489418 .025443764 16070 .01 2 5 .12758258 .026201123 16070 .01 2 . .14843191 . 16070 .01 . . .14874525 . 16161 0 1 1 .014269484 .0505924 16161 0 1 2 .034955855 .030939957 16161 0 1 3 .017997812 .03088566 16161 0 1 4 -.029656284 .04419843 16161 0 1 5 .0182078 .02381743 16161 0 1 . .06061478 . 16161 0 2 1 .0026539804 .03029218 16161 0 2 2 .030466346 .029775864 16161 0 2 3 .05469561 .02672335 16161 0 2 4 .04070667 .025443764 16161 0 2 5 .02181926 .026201123 16161 0 2 . .0390017 . 16161 0 . . .008688615 . 16252 .48 1 1 .0045106304 .0505924 16252 .48 1 2 -.01369509 .030939957 16252 .48 1 3 -.01494146 .03088566 16252 .48 1 4 -.01883998 .04419843 16252 .48 1 5 -.032066233 .02381743 16252 .48 1 . -.0031058835 . 16252 .48 2 1 .07111678 .03029218 16252 .48 2 2 .04519347 .029775864 16252 .48 2 3 -.001190891 .02672335 16252 .48 2 4 .028608523 .025443764 16252 .48 2 5 .023474427 .026201123 16252 .48 2 . .016408863 . 16252 .48 . . .067791395 . 16344 .06 1 1 -.008275456 .0505924 16344 .06 1 2 .04520285 .030939957 16344 .06 1 3 -.029116403 .03088566 16344 .06 1 4 -.05764899 .04419843 16344 .06 1 5 -.03761478 .02381743 16344 .06 1 . -.020807166 . 16344 .06 2 1 -.006838386 .03029218 16344 .06 2 2 -.014781592 .029775864 16344 .06 2 3 -.034003496 .02672335 16344 .06 2 4 -.007011445 .025443764 16344 .06 2 5 -.02536957 .026201123 16344 .06 2 . -.018418022 . 16344 .06 . 1 -.11654709 . 16344 .06 . . .02564771 . 16436 -.14 1 1 .11657377 .0505924 16436 -.14 1 2 .05655047 .030939957 16436 -.14 1 3 .08888225 .03088566 16436 -.14 1 4 .035002757 .04419843 16436 -.14 1 5 .10658953 .02381743 16436 -.14 1 . .131198 . 16436 -.14 2 1 .13029574 .03029218 16436 -.14 2 2 .07498024 .029775864 end format %td rdate
Comment