An Adjusted Maximum Likelihood Estimator of Autocorrelation in Disturbances

  • Clifford Hildreth
  • Warren T. Dent


Although it has been shown [9] that the maximum likelihood (M.L.) estimator of the autocorrelation coefficient in linear models with autoregressive disturbances is asymptotically unbiased, several Monte Carlo studies [8], [6], [15] suggest that finite sample bias is usually large enough to be of some concern. In the next section an approximation to the bias is developed and used to obtain an adjusted estimator with substantial smaller bias. Section 3 presents the results of applying the adjusted M.L. estimator, the unadjusted M.L., and two other estimators to Monte Carlo data. Some interpretations and conjectures comprise Section 4 and computing procedures are discussed in Section 5. The remainder of this section contains a brief sketch of maximum likelihood estimation.


Maximum Likelihood Estimator Monte Carlo Study Autocorrelation Coefficient Extreme Difference Monte Carlo Data 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    Anderson, R. L. and Anderson, T. W., The Distribution of the Circular Serial Correlation Coefficient for Residuals from a fitted Fourier Series, Annals of Mathematical Statistics, XXI (1950) 50–81.Google Scholar
  2. [2]
    Chipman, John S., The Problem of Testing for Serial Correlation in Regression Analysis, Technical Report 4, Department of Economics, University of Minnesota, 1965.Google Scholar
  3. [3]
    De Leeuw, Frank and Gramlich, Edward, The Channels of Monetary Policy: a Further Report on the Federal Reserve-M.I.T. Model, The Journal of Finance, XXIV (1969) 265–90.Google Scholar
  4. [4]
    Dent, Warren T., The Distribution of the Maximum Likelihood Estimator of the Autocorrelation Coefficient in a Linear model with Autoregressive Disturbances, Unpublished Ph.D. Thesis, University of Minnesota, 1971.Google Scholar
  5. [5]
    Durbin, J., Estimation of Parameters in Time-Series Regression Models, Journal of the Royal Statistical Society, Series B, XXII (1960) 139–53.Google Scholar
  6. [6]
    Griliches, Zvi and Rao, P., Small-Sample Properties of Several Two-Stage Regression Methods in the Context of Autocorrelated Errors, Journal of the American Statistical Association, LXIV (1969) 253–72.Google Scholar
  7. [7]
    Henshaw, R. C., Jr., Testing Single-Equation Least-Squares Regression Models for Autocorrelated Disturbances, Econometrica, XXXIV (1966) 646–60.CrossRefGoogle Scholar
  8. [8]
    Hildreth, Clifford and Lu, John Y., A Monte Carlo Study of the Regression Model with Autoregressive Disturbances, Rand Memorandum RM5728PR, Santa Monica, 1969.Google Scholar
  9. [9]
    Hildreth, Clifford, Asymptotic Distribution of Maximum Likelihood Estimators in a Linear Model with Autoregressive Disturbances, Annals of Mathematical Statistics, XL, (1969) 583–94.CrossRefGoogle Scholar
  10. [10]
    Hildreth, Clifford and Lu, John Y., Demand Relations with Auto-Correlated Disturbances, Technical Bulletin 276 of the Michigan State University Agricultural Experiment Station, East Lansing, 1960.Google Scholar
  11. [11]
    Hoos, S., and Shear, S. W., Relation Between Auction Prices and Supplies of California Fresh Bartlett Pears, Hilgardia, XIV, California Agricultural Experiment Station, 1942.Google Scholar
  12. [12]
    Klein, Lawrence R., Economic Fluctuations in the United States, 1921–41. New York: John Wiley and Sons, 1950.Google Scholar
  13. [13]
    Linstrom, I., and King, Richard A., The Demand for North Carolina Slicing Cucumbers and Green Peppers, North Carolina State College, A. E. Information Series, No. 49, 1956.Google Scholar
  14. [14]
    Prest, A. R., Some Experiments in Demand Analysis, The Review of Economics and Statistics, XXXI (1949) 33–49.CrossRefGoogle Scholar
  15. [15]
    Reilly, D. P., Evaluation of the Small Sample Properties of Five Alternative Estimation Methods when the Errors are Correlated; Discussion Paper No. 87, Department of Economics, University of Pennsylvania, 1968.Google Scholar
  16. [16]
    Theil, Henri and Nagar, A. L., Testing the Independence of Regression Disturbances, Journal of the American Statistical Association, LVI (1961) 793–806.CrossRefGoogle Scholar
  17. [17]
    Zellner, Arnold and Tiao, G. C., Bayesian Analysis of the Regression Model with Autocorrelated Errors, Journal of the American Statistical Association, LIX (1964) 763–78.CrossRefGoogle Scholar

Copyright information

© Clifford Hildreth and Warren T. Dent 1974

Authors and Affiliations

  • Clifford Hildreth
    • 1
    • 2
  • Warren T. Dent
    • 1
    • 2
  1. 1.The University of MinnesotaUSA
  2. 2.The University of IowaUSA

Personalised recommendations