Announcement

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

  • Quantile regression with 4Qs

    Dear Statalist Members,

    I have a variables, called haz06 (i.e., height for age z-score). I would like to run a quantile regression with 4 quantiles such that 0-25, 25.1-50, 50.1-75, and 75.1-100. So, I need

    Codes like qreg haz06 z_ihs_refugee_share i.plate i.survey_year if child_birth_year!=2012, quantile(0.25) or sqreg haz06 z_ihs_refugee_share i.plate i.survey_year if child_birth_year!=2012, quantiles(25 50 75) do not give me the the interval I provided above. I need 4Qs.

    If you can help me with the code, I would be really appreciated. Thank you so much.


    sum haz06, detail

    Length/height-for-age Z-score
    -------------------------------------------------------------
    Percentiles Smallest
    1% -4.36 -11.53
    5% -2.66 -10.57
    10% -2.03 -9.74 Obs 5,892
    25% -1.15 -9.5 Sum of Wgt. 5,892

    50% -.3 Mean -.2900204
    Largest Std. Dev. 1.620961
    75% .55 14.25
    90% 1.46 15.4 Variance 2.627515
    95% 2.06 16.18 Skewness .8385876
    99% 3.73 16.78 Kurtosis 14.71797

  • #2
    I think there may be a misunderstanding of what qreg does
    what seems you are seeking is. A way to use a subsample such that y is within say 25th and 50th quintiles
    this kind of subsampling is not correct
    perhaps you may want to revise your question and methodology
    F

    Comment

    Working...
    X