The prediction of conditional losses - an evaluation of univariate Value at Risk models
Engle (1982) explained in his classic paper what is meant by a "good" interval forecast. His main insight was that intervals should be narrow in tranquil times and wide in volatile times so that the times when the interval forecast fails are spread out over the sample and not come in clusters. On the basis of Engle's criterias, Christoffersen (1998) was the first to form a complete t
