Adaptive prediction by least squares predictors in stochastic regression models with applications to time series (Q1102060): Difference between revisions

The asymptotic performance of the least squares predictors \(\hat y_ n\) of the stochastic regression model \(y_ n=\beta_ 1x_{n1}+...+\beta_ px_{np}+\epsilon_ n\) is considered. In particular, the accumulated cost function \(\sum^{n}_{k=1}(y_ k-\hat y_ k-\epsilon_ k)^ 2\) is studied. The results are then applied to nonstationary autoregressive time series. A statistic is also constructed to show how many times one should difference a nonstationary time series in order to obtain a stationary series.