Inference for random coefficient volatility models

Estimating functions have been shown to be convenient to study inference for nonlinear time series models. One such model is the recently proposed Random Coefficient Autoregressive (RCA) model with Generalized Autoregressive Heteroscedasticity (GARCH) errors (Thavaneswaran et al., 2009). We derive the martingale estimating functions for the joint estimation of the conditional mean and variance parameters and we show the information gain relative to conditional least square estimation.