An extension of Lai and Wei's law of the iterated logarithm with applications to time series analysis and regression (Q910090)

Extending a result of \textit{T.-L. Lai} and \textit{C.-Z. Wei} [Ann. Probab. 10, 320-335 (1982; Zbl 0485.60031)], a LIL is proved for \(S_ N=\sum_{n}a_{Nn}u_ n\) with \((a_{Nn})_{-\infty <n<\infty}^ a \)double array of non-random constants and \(u_ n=\sum^{\infty}_{j=0}k_ j\epsilon_{n-j}\), where the \(\epsilon_ n\) are independent random variables. This is applied to finite Fourier transforms and least square estimates in regression models in which the random disturbances form a linear series.