Regression estimation and prediction for discrete time processes
The construction and study of a nonparametric predictor are the main purpose of this chapter. In practice such a predictor is in general more efficient and more flexible than the predictors based on BOX and JENKINS method, and nearly equivalent if the underlying model is truly linear. This surprising fact will be clarified at the end of the chapter.
KeywordsMarkov Process Regression Estimation Kernel Estimator Regression Estimator Markov Inequality
Unable to display preview. Download preview PDF.