Multiple Regression with Stationary Errors
David B. Duncan, Richard H. Jones · Journal of the American Statistical Association · 1966
A practical computing technique is presented for the joint estimation of regression coefficients and the error spectrum in regression problems with stationary errors. In problems where unweighted least squares is not efficient, the procedure gives an estimate approaching the minimum variance linear unbiased estimate. Whether unweighted least squares is nearly efficient or not efficient, the procedure, in general, gives a much closer estimate of the covariance matrix of the estimated regression coefficients. The proposed approach, which involves the finite Fourier transformation of the observations and regression vectors, gives a useful intuitive understanding of the effects of correlated errors on regression.
Cited by 22 works. doi:10.1080/01621459.1966.10482184
Authors
References
- A New Approach to Linear Filtering and Prediction Problems
- An Algorithm for the Machine Calculation of Complex Fourier Series
- Mathematical Methods for Digital Computers.
- An Inverse Matrix Adjustment Arising in Discriminant Analysis
- On empirical spectral analysis of stochastic processes
- SOME THEOREMS IN LEAST SQUARES
- A Reappraisal of the Periodogram in Spectral Analysis
- A Reappraisal of the Period Spectral Analysis
- Estimating relations between time series
- AN INTRODUCTION TO ESTIMATION THEORY FOR DYNAMICAL SYSTEMS