Partial Time-Varying Coefficient Regression and Autoregressive Mixed Model

Regression and autoregressive mixed models are classical used to analyze the relationship between time series response variable other covariates. The coefficients in traditional regression constants. However, for complicated data, of covariates may change with time. In this article, we propose a kind partial time-varying coefficient model obtain local weighted least-square estimators functions by polynomial technique. asymptotic normality properties derived under regularity conditions, simulation studies conducted empirically examine finite-sample performances proposed estimators. Finally, use real data about Lake Shasta inflow illustrate application model.