Quantile Regression with Nonadditive Fixed Effects

This paper introduces a quantile regression estimator for panel data (QRPD) with nonadditive fixed effects, maintaining the nonseparable disturbance term commonly associated with quantile estimation. QRPD estimates the impact of exogenous or endogenous treatment variables on the outcome distribution using ``within" variation in the treatment variables or instruments for identification purposes. Most quantile panel data estimators include additive fixed effects which separates the disturbance term and assumes the parameters vary based only on the time-varying components of the disturbance term. QRPD is consistent for small T and straightforward to implement. The nonadditive fixed effects are never estimated or even specified. Simulation results show that QRPD works well even when instrumental variables quantile regression and additive fixed effect quantile estimators are biased. I estimate the effect of the 2008 tax rebates on the distribution of consumption, finding evidence of substantial heterogeneity in the household spending response to transitory income.

Stata code available: "ssc install qregpd"

https://ideas.repec.org/c/boc/bocode/s458157.html