Replication data and code for the simulations in "Quantile Ratio Regression for Panel Data"

This dataset contains the R code and simulation outputs used to produce the Monte Carlo results in the paper "Quantile Ratio Regression for Panel Data". The scripts generate unbalanced panel data with up to n = 5000 units and T in {4, 6} periods, under two correlation regimes for the covariates (rho in {0, 0.5}) and two conditional distributions (Log-normal and Weibull). We compare four estimators of the quantile ratio regression model for panel data: (i) unregularized QRR with unit fixed effects (QRR1); (ii) ridge-regularized QRR with empirical-Bayes tuning (L2-QRR); (iii) the cross-sectional QRR implemented via the qrr function in the R package qtools (QRR2); and (iv) a naive QRR based on two separate fixed-effects quantile regressions. The deposited files include the R scripts that generate the data and run all B = 500 replications for each design, together with the saved .RData workspaces and a README file explaining how to rerun the simulations and reproduce the tables reported in the paper.