Hello.
I have a dataset that consists of variable information such as (MAX, LMAX, LBETA, LSIZE, LBM) for a list of firms (ID) for monthly time frame (mndate).
I'm attaching an excerpt from Bali, Cakici & Whitelaw (2011) that I wanted to recreate:
"...Each month from July 1962 to December 2005 we run a firm-level cross-sectional regression of the maximum daily return in that month (MAX) on subsets of lagged predictor variables including MAX in the previous month and seven control variables that are defined in the Appendix. The table reports the time-series averages of the cross-sectional regression coefficients, their associated Newey-West (1987) adjusted t-statistics (in parentheses), and the regression R-squareds."
What I understand is this:
1) Carry out cross-sectional regressions each month of the form
2) Store the coefficient estimates and R-squared
3) Average the coefficient estimates and R-squared across all time periods.
However, I can't figure out how to do this in STATA. I thought this was a Fama-Macbeth regression and tried out asreg, however I think that is not how its supposed to work. Please help me out.
My data is as follows:
I have a dataset that consists of variable information such as (MAX, LMAX, LBETA, LSIZE, LBM) for a list of firms (ID) for monthly time frame (mndate).
I'm attaching an excerpt from Bali, Cakici & Whitelaw (2011) that I wanted to recreate:
"...Each month from July 1962 to December 2005 we run a firm-level cross-sectional regression of the maximum daily return in that month (MAX) on subsets of lagged predictor variables including MAX in the previous month and seven control variables that are defined in the Appendix. The table reports the time-series averages of the cross-sectional regression coefficients, their associated Newey-West (1987) adjusted t-statistics (in parentheses), and the regression R-squareds."
What I understand is this:
1) Carry out cross-sectional regressions each month of the form
Code:
reg MAX LMAX LBETA LSIZE LBM
3) Average the coefficient estimates and R-squared across all time periods.
However, I can't figure out how to do this in STATA. I thought this was a Fama-Macbeth regression and tried out asreg, however I think that is not how its supposed to work. Please help me out.
My data is as follows:
Code:
* Example generated by -dataex-. To install: ssc install dataex clear input float mndate int ID double(MAX LMAX LBETA LSIZE LBM) 558 1 .04980730873810124 .0504730081583058 1.148639564559923 10.214115468963447 .2590673575129534 708 1 .04691023675818373 .04462440158645742 1.7003144190375998 11.853398971537617 .7751937984496123 508 1 .21740513881155032 .035915334384147846 1.4686103395566852 7.793792967685271 1.5625 672 1 .04603660487783623 .023008494129308943 1.171682541444171 11.983450833086566 .4587155963302752 680 1 .0208122388811931 .027188787189847297 1.301872873140102 12.32027550220945 .38759689922480617 710 1 .03893648422250595 .04705667454472562 1.721526661367964 11.92458468324172 .7246376811594204 762 1 .07500596018595784 .03543922770674324 1.2463125113159177 12.005240214767351 1.3513513513513513 615 1 .022668709915432128 .04990227024018101 1.448160018583783 9.990643669050488 1.2658227848101264 629 1 .020133458313941865 .04283854692543554 .9782682353172124 10.617393520772247 .8130081300813008 695 1 .04991228651220814 .03300951658833684 1.208794752382793 11.98266280876514 .6060606060606061 607 1 .04754600234299655 .03514413230418234 1.188717742696194 10.02157676353828 1.2345679012345678 546 1 .02059286600560961 .05062926092122422 .8909925223717089 9.391290154985224 .4291845493562232 525 1 .08176285302439812 .07469470994373317 1.0132721012338854 8.528971553372259 .8547008547008548 537 1 .02052248178849525 .03658736914415428 1.6244572311651557 9.130918304998692 .5555555555555556 730 1 .03303639012943771 .06233536455319677 1.2376753101554256 12.290528217564354 .625 572 1 .049834280303030276 .04451978681401991 1.197497251762921 10.275421385235479 .5235602094240838 714 1 .03276353747344585 .03753522082232941 1.7202343666313664 11.720219270049597 .8849557522123894 653 1 .049493804199251645 .03498132184348335 1.0981232030777508 11.696145784422665 .42016806722689076 end format %tm mndate
