Semiparametric Inference on Trending Panel Data with Spatial Dependence
报告人: Peter M. Robinson, London School of Economics
时间:2016-07-01 10:00 ~ 11:00
地点:Room 217, Guanghua Building 2
Abstract:
Semiparametric panel data modelling and statistical inference with fractional stochastic trends, nonparametrically time-trending individual effects, and general cross-sectional correlation and heteroscedasticity in innovations is developed. The fractional stochastic trends allow for a wide range of nonstationarity, indexed by a memory parameter, nesting the familiar I(1) case and allowing for parametric short-memory. The individual effects can nonparametrically vary simultaneously across time and across units. The cross-sectional covariance matrix allows for a wide range of spatial dependence, covering familiar parametric models but avoiding misspecification by being nonparametric. The main focus is on robust estimation of the time series parameters. We obtain standard asymptotics, with a central limit theorem, whatever the true value of the memory parameter, and unlike the nonstandard asymptotics for autoregressive parameter estimates at a unit root.
About the Speaker: