Chapter 49 Arch models

摘要: Abstract This chapter evaluates the most important theoretical developments in ARCH type modeling of time-varying conditional variances. The coverage include specification univariate parametric models, general inference procedures, conditions for stationarity and ergodicity, continuous time methods, aggregation forecasting multivariate covariance formulations, use model selection criteria an context. Additionally, contains a discussion empirical regularities pertaining to temporal variation financial market volatility. Motivated part by recent results on optimal filtering, new variance better characterizing stock return volatility is also presented.