LAUSR

Abstract A class of functional coefficient ARCH-M models are discussed in this paper. Based on empirical likelihood method, a weighted empirical likelihood estimation procedure is proposed for estimating the functional coefficients. Furthermore, the constructed empirical log-likelihood ratio is shown to be asymptotically chi-squared, and then the pointwise confidence intervals for functional coefficients are constructed. The proposed empirical likelihood estimation method can deal with the heteroscedasticity in ARCH-M model, and is more robust and effective. Some simulations are also undertaken to assess the finite sample performance of the proposed estimation procedure.