Dear all,
Thanks to Prof. Baum, a new package is available in SSC.
VC_PACK is a set of commands that can be used for the estimation and assessment of varying coefficient models with a single running variable. This is a kind of semiparametric regression model.
You can read about the details here:https://journals.sagepub.com/doi/abs...36867X20953574
It estimates model of the following type:
y = x * b(z) +e
where Z is a single continuous variable. The coefficients b(z) are approximated using local linear regression.
You will find that this command is faster than -npregress kernel- for the estimation of optimal bandwidths when samples are of moderate size.
Please let me know if you find any problems with this.
Best regards
Fernando
0 Response to new to SSC vc_pack: Estimation of smooth varying coefficient models
Post a Comment