Uniform nonparametric inference for time series

This paper provides the first result for the uniform inference based on nonparametric series estimators in a general time-series setting. We develop a strong approximation theory for sample averages of mixingales with dimensions growing with the sample size. We use this result to justify the asymptotic validity of a uniform confidence band for series estimators and show that it can also be used to conduct nonparametric specification test for conditional moment restrictions. New results on the validity of heteroskedasticity and autocorrelation consistent (HAC) estimators with increasing dimension are established for making feasible inference. An empirical application on the unemployment volatility puzzle for the search and matching model is provided as an illustration.