Regression parameter estimation with serially correlated errors

Least-squares regression is a standard statistical technique for relating variables by means of functional relationships. On the assumption that the errors are independent of one another and have equal variance, the regression parameter estimators are approximately efficient and the conventional estimators of the variances of these estimators are approximately unbiased. However, if the errors are correlated, then in general neither property holds true. This thesis is directed towards overcoming these deficiences when errors are serially correlated; arising because a series of measurements is made on a single experimental unit, rather than on a number of separate units. Two types of solution are considered: (1) model the error process, either empirically or mechanistically, and estimate any unknown parameters jointly with the regression parameters (chapters 4 to 8); (2) derive estimators of the variances of least-squares regression parameter estimators which take account of the error correlations (chapter 9). The diversity of models considered in the first solution is facilitated computationally by the specification in chapter 2 of a new class of serially-structured error processes, termed generalized autoregressive-moving average processes. In chapter 3 solutions of linear stochastic difference and differential equations are shown to be in this class. Empirical models of the error processes are based entirely on the corresponding data sets; no other information being available on the forms the correlations should take. Basically, a regression

[1]  A G Hawkes,et al.  Relaxation and fluctuations of membrane currents that flow through drug-operated channels , 1977, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[2]  P. Robinson,et al.  Alternative models for stationary stochastic processes , 1978 .

[3]  G. Box,et al.  Distribution of Residual Autocorrelations in Autoregressive-Integrated Moving Average Time Series Models , 1970 .

[4]  Robin Thompson,et al.  A Note on the Estimation of the Parameters of the Autoregressive-Moving Average Process , 1977 .

[5]  M. Crowder Maximum Likelihood Estimation for Dependent Observations , 1976 .

[6]  M. Dhanoa,et al.  A partial adjustment model to describe the lactation curve of a dairy cow , 1982 .

[7]  Anthony C. Atkinson,et al.  The use of residuals as a concomitant variable , 1969 .

[8]  P. D. P. WOOD,et al.  Algebraic Model of the Lactation Curve in Cattle , 1967, Nature.

[9]  E. Hnizdo,et al.  Multibreed comparisons of British cattle. Variation in body weight, growth rate and food intake , 1984 .

[10]  S. M. Pandit,et al.  Unique estimates of the parameters of a continuous stationary stochastic process , 1975 .

[11]  Walter W Garvin,et al.  Introduction to Linear Programming , 2018, Linear Programming and Resource Allocation Modeling.

[12]  D. Cochrane,et al.  Application of Least Squares Regression to Relationships Containing Auto-Correlated Error Terms , 1949 .

[13]  H. Cramér Mathematical methods of statistics , 1947 .

[14]  Performance of the Durbin-Watson Test and WLS Estimation When the Disturbance Term Includes Serial Dependence in Addition to First-Order Autocorrelation , 1982 .

[15]  D. A. Pierce Least squares estimation in dynamic-disturbance time series models , 1972 .

[16]  J. Griffiths The Theory of Stochastic Processes , 1967 .

[17]  J. R. Parks A theory of animal weight response to controlled feeding. , 1975, Journal of theoretical biology.

[18]  A. M. Walker,et al.  A Survey of statistical work on the Mackenzie River series of annual Canadian lynx trappings for the , 1977 .

[19]  M. H. Quenouille A Large‐Sample Test for the Goodness of Fit of Autoregressive Schemes , 1947 .

[20]  Zvi Griliches,et al.  Small-Sample Properties of Several Two-Stage Regression Methods in the Context of Auto-Correlated Errors , 1969 .

[21]  C. A. Glasbey A Generalization of Partial Autocorrelations Useful in Identifying ARMA Models , 1982 .

[22]  Louis Brand,et al.  Differential and Difference Equations , 1968 .

[23]  John S. White THE LIMITING DISTRIBUTION OF THE SERIAL CORRELATION COEFFICIENT IN THE EXPLOSIVE CASE , 1958 .

[24]  T. W. Anderson,et al.  Statistical analysis of time series , 1972 .

[25]  D. B. Preston Spectral Analysis and Time Series , 1983 .

[26]  Michael Gaster,et al.  The spectral analysis of randomly sampled records by a direct transform , 1977, Proceedings of the Royal Society of London. A. Mathematical and Physical Sciences.

[27]  George Zyskind,et al.  On Canonical Forms, Non-Negative Covariance Matrices and Best and Simple Least Squares Linear Estimators in Linear Models , 1967 .

[28]  D. M. Allen,et al.  Analysis of growth and dose response curves. , 1969, Biometrics.

[29]  S. Chatterjee,et al.  Regression Analysis by Example , 1979 .

[30]  D. Cook,et al.  Multiple regression in geographical mortality studies, with allowance for spatially correlated errors. , 1983, Biometrics.

[31]  J. H. Wilkinson,et al.  Handbook for Automatic Computation. Vol II, Linear Algebra , 1973 .

[32]  C. Glasbey CEVOPE — a Program for Finding Conservative Estimates of the Variances of Regression Parameter Estimators when Errors are Serially Correlated , 1984 .

[33]  P. R. Fisk,et al.  Models of the Second Kind in Regression Analysis , 1967 .

[34]  D. J. Finney,et al.  Growth Curves: their Nature, Uses, and Estimation , 1978 .

[35]  Walter Krämer,et al.  Finite Sample Efficiency of Ordinary Least Squares in the Linear Regression Model With Autocorrelated Errors , 1980 .

[36]  G. Box,et al.  On a measure of lack of fit in time series models , 1978 .

[37]  J. Durbin,et al.  Testing for serial correlation in least squares regression. II. , 1950, Biometrika.

[38]  C. Ansley An algorithm for the exact likelihood of a mixed autoregressive-moving average process , 1979 .

[39]  N E Manos,et al.  Stochastic Models , 1960, Encyclopedia of Social Network Analysis and Mining. 2nd Ed..

[40]  Paul Newbold,et al.  On the finite sample distribution of residual autocorrelations in autoregressive-moving average models , 1979 .

[41]  C. L. Chiang ON REGULAR BEST ASYMPTOTICALLY NORMAL ESTIMATES , 1956 .

[42]  Exponential regression with correlated observations. , 1968, Biometrika.

[43]  R. Anderson,et al.  The Problem of Autocorrelation in Regression Analysis , 1954 .

[44]  O. Mayo,et al.  Nearest Neighbour (Nn) Analysis of Field Experiments , 1983 .

[45]  F. E. Satterthwaite An approximate distribution of estimates of variance components. , 1946, Biometrics.

[46]  J. Johnston,et al.  Econometric Methods, 2nd Ed. , 1976 .

[47]  M. H. Quenouille Discrete autoregressive schemes with varying time-intervals , 1958 .

[48]  A. Wald,et al.  On the Statistical Treatment of Linear Stochastic Difference Equations , 1943 .

[49]  J. Durbin Estimation of Parameters in Time‐Series Regression Models , 1960 .

[50]  Chris A. Glasbey,et al.  The analysis of data from germination tests , 1984, The Journal of Agricultural Science.

[51]  C. L. Chiang,et al.  An introduction to stochastic processes and their applications , 1978 .

[52]  D. A. Pierce Distribution of Residual Autocorrelations in the Regression Model with Autoregressive-Moving Average Errors , 1971 .

[53]  Geoffrey S. Watson,et al.  The inefficiency of least squares , 1975 .

[54]  G. S. Watson,et al.  SERIAL CORRELATION IN REGRESSION ANALYSIS. I , 1955 .

[55]  J. Ord,et al.  Spatial Processes. Models and Applications , 1982 .

[56]  B. Blight,et al.  A Bayesian approach to model inadequacy for polynomial regression , 1975 .

[57]  J. Mills,et al.  Production characteristics of high-yielding dairy cows , 1983 .

[58]  J. H. Wilkinson The calculation of the latent roots and vectors of matrices on the pilot model of the A.C.E. , 1954, Mathematical Proceedings of the Cambridge Philosophical Society.

[59]  C. A. Glasbey,et al.  Correlated Residuals in Non‐Linear Regression Applied to Growth Data , 1979 .

[60]  J. Durbin,et al.  Testing for serial correlation in least squares regression. I. , 1950, Biometrika.

[61]  C. A. Glasbey Nonlinear Regression with Autoregressive Time Series Errors , 1980 .

[62]  G. Ljung,et al.  The likelihood function for a stationary Gaussian autoregressive-moving average process with missing observations , 1982 .

[63]  The Use of Exponentially Damped Polynomials for Biological Recovery Data , 1981 .

[64]  Andrew Harvey,et al.  Maximum likelihood estimation of regression models with autoregressive-moving average disturbances , 1979 .

[65]  D. B. Duncan,et al.  Multiple Regression with Stationary Errors , 1966 .

[66]  R. W. Wedderburn Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method , 1974 .

[67]  A. Ronald Gallant,et al.  Nonlinear Regression with Autocorrelated Errors , 1976 .

[68]  Chris Chatfield,et al.  Some Recent Developments in Time‐Series Analysis , 1977 .

[69]  Richard H. Jones FITTING A CONTINUOUS TIME AUTOREGRESSION TO DISCRETE DATA , 1981 .

[70]  H. D. Patterson,et al.  Discussion of : Statutory and Recommended List Trials of Crop Varieties in the United Kingdom , 1980 .

[71]  P. Robinson,et al.  Estimation of a time series model from unequally spaced data , 1977 .

[72]  John J. Spitzer,et al.  Small-Sample Properties of Nonlinear Least Squares and Maximum Likelihood Estimators in the Context of Autocorrelated Errors , 1979 .

[73]  THE EFFICIENCIES OF ALTERNATIVE ESTIMATORS FOR AN ASYMPTOTIC REGRESSION EQUATION , 1958 .

[74]  S. Wu,et al.  Modeling of Continuous Stochastic Processes from Discrete Observations with Application to Sunspots Data , 1974 .

[75]  J. Cobby,et al.  On fitting curves to lactation data , 1978 .

[76]  Yonathan Bard,et al.  Nonlinear parameter estimation , 1974 .

[77]  C. A. McGilchrist,et al.  Stochastic Growth Curve Analysis , 1979 .

[78]  C. J. Stone,et al.  Introduction to Stochastic Processes , 1972 .

[79]  P. Gill,et al.  Quasi-Newton Methods for Unconstrained Optimization , 1972 .

[80]  J. Chipman EFFICIENCY OF LEAST-SQUARES ESTIMATION OF LINEAR TREND WHEN RESIDUALS ARE AUTOCORRELATED , 1979 .

[81]  J. H. M. Thornley,et al.  Mathematical models in plant physiology : a quantitative approach to problems in plant and crop physiology , 1976 .

[82]  David A. Pierce,et al.  Least squares estimation in the regression model with autoregressive-moving average errors , 1971 .

[83]  R. Jennrich Asymptotic Properties of Non-Linear Least Squares Estimators , 1969 .