Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • log-log model

    i m working on a log log model.
    first run a regression of log wage on education, log experience, female, married, large-firm.
    use the corrected prediction to calculate the marginal effect of an additional year of experience for an unmarried male at a small firm at the average education and experience.

    do i do this right?
    gen lnhrearn=ln(hrearn)
    gen lnexper=ln(exper)
    scalar avexper=r(mean)
    scalar lnavexper=ln(avexper)
    scalar aveducrecode=r(mean)
    scalar sig2b=e(rss)/e(df_r)
    di "corrected factor="exp(sig2b/2)
    scalar predlnhrearn1=_b[_cons]+_b[educrecode]*aveducrecode+_b[married]*0+_b[female]*0+_b[largfirm]*0+_b[lnexper]*lnavexper
    scalar sig2b=e(rss)/e(df_r)
    scalar yhatd=exp(predlnhrearn1)*exp(sig2b/2)
    disp _b[lnexper]*(yhatd/avexper)

    i ask because i got different stata outcome with hand calculation.
Working...
X