This Stata package offers fast estimation and inference procedures for the distribution regression models. First, the new command drprocess implements new algorithms that are much faster than repeatedly calling commands for binary regression, especially when a large number of regressions or bootstrap replications must be estimated. Second, this command provides analytical estimates of the variance-covariance matrix of the coefficients for several regressions allowing for weights, clustering and stratification. Third, in addition to traditional pointwise confidence intervals, drprocess also provide functional confidence bands and tests of functional hypotheses. Fourth, predict – called after drprocess – can generate monotone estimates of the conditional quantile and distribution functions obtained by rearrangement. Fifth, plotprocess conveniently plots the estimated coefficients with their confidence intervals and uniform bands.
The easiest way to install this package is to type:
net install drprocess, from("https://raw.githubusercontent.com/bmelly/Stata/main/")
directly in the Stata command prompt.
Alternatively you can download the files in this zip-file and copy them in your personal ado directory. (Where is my personal ado directory?)
When you have installed the package you can type "help drprocess" and "help plotprocess" to learn more about these commands.
This article in preparation for the Stata Journal provides more information: Quantile and distribution regression in Stata: algorithms, pointwise and functional inference