CONDITIONAL QUANTILE ESTIMATION AND INFERENCE FOR ARCH MODELS

Quantile regression methods are suggested for a class of ARCH models. Because conditional quantiles are readily interpretable in semiparamet ric ARCH models and are inherently easier to estimate robustly than po pulation moments, they offer some advantages over more familiar method s based on Gaussian likelihoods. Related inference methods, including the construction of prediction intervals, are also briefly discussed.