臺灣博碩士論文加值系統

對於廣義線性模式（generalized linear models，GLMs）中迴歸係數之最大概似估計量（maximum likelihood estimates，MLEs）通常可藉由一疊代重新加權最小平方法（iterative reweighted least squares，IRLS）算則獲得。在本研究中，我們發現在IRLS算則收斂時最後一次疊代後的線性迴歸模式可視為等同於原廣義線性模式之一個線性表示式。這樣的發現不僅含有其原本等式之統計意義，更可以進一步提供我們一個十分有用的工具，即將廣義線性模式的線性化用於模式配適之目的。因此，藉由利用IRLS算則的雙重角色，我們對於廣義線性模式提出一個新的判定係數（coefficient of determination，R2）以評估模式配適好壞，及各種常用的迴歸診斷（regression diagnostics）統計量。

The maximum likelihood estimates (MLEs) of the regression coefficients in generalized linear models (GLMs) are usually obtained by an iterative reweighted least squares (IRLS) algorithm (Charnes, Frome, and Yu 1976). In this study, we find that at the end of the last iteration of the IRLS algorithm, an equivalent linear regression representation of a GLM can also be obtained. This finding is not only very interesting in its own rights, but it provides us with a useful tool to linearize GLMs for model fitting purposes. By making use of the dual role of the IRLS algorithm, we propose a new coefficient of determination and various regression diagnostic statistics for GLMs and their quasi-likelihood extensions as an illustration of this tool.

1 INTRODUCTION 1
2 THE ITERATIVE REWEIGHTED LEAST SQUARES (IRLS) ALGORITHM 2
3 APPLICATIONS 3
3.1 Coefficient of Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3.2 Regression Diagnostic Statistics . . . . . . . . . . . . . . . . . . . . . . . . . 6
4 SIMULATIONS 8
4.1 Data Generation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
4.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.2.1 Linear Regression Model . . . . . . . . . . . . . . . . . . . . . . . . . 11
4.2.2 Logistic Regression Model . . . . . . . . . . . . . . . . . . . . . . . . 12
4.2.3 Poisson Regression Model . . . . . . . . . . . . . . . . . . . . . . . . 13
5 DISCUSSION 15
6 ACKNOWLEDGMENTS 16
7 APPENDIX 16
8 REFERENCES 22
9 TABLES 26
10 FIGURES 39

Bowerman, B. L. & O’Connell, R. T. (1990). Linear Statistical Models: An Applied Approach,
2nd ed. Belmont, CA: Duxbury Press.
Charnes, A., Frome, E. L. & Yu, P. L. (1976). The equivalence of generalized least squares
and maximum likelihood estimates in the exponential family. J. Amer. Statist. Assoc.
71, 169-71.
Collett, D. (2003). Modelling Binary Data, 2nd ed. Boca Raton, FL: Chapman & Hall/CRC.
Cox, D. R. & Snell, E. J. (1989). Analysis of Binary Data, 2nd ed. London: Chapman &
Hall.
Dobson, A. J. (2002). An Introduction to Generalized Linear Models, 2nd ed. Boca Raton,
FL: Chapman & Hall/CRC.
Fahrmeir, L. & Tutz, G. (2001). Multivariate Statistical Modelling Based on Generalized
Linear Models, 2nd ed. New York, NY: Springer-Verlag.
Firth, D. (1991). Generalized linear models. In: D. V. Hinkley, N. Reid & E. J. Snell (Eds),
Statistical Theory and Modelling: In Honour of Sir David Cox, FRS, London: Chapman
& Hall, pp. 55-82.
Gail, M. H., Wieand, S. & Piantadosi, S. (1984). Biased estimates of treatment effect in
randomized experiments with nonlinear regressions and omitted covariates. Biometrika
71, 431-44.
Greene, W. H. (2003). Econometric Analysis, 5th ed. Upper Saddle River, NJ: Prentice-Hall.
Hamilton, L. C. (1992). Regression with Graphics: A Second Course in Applied Statistics.
Belmont, CA: Duxbury Press.
Hosmer, D. W. & Lemeshow, S. (2000). Applied Logistic Regression, 2nd ed. New York,
NY: John Wiley & Sons.
Hu, F. -C., Tsai, T. -L., Shau, W. -Y. & Chang, H. -C. (2003). The omitted-variable bias in
linear regression model: A general result. Unpublished manuscript, Division of Biostatistics,
Graduate Institute of Epidemiology, College of Public Health, National Taiwan
University, Taipei, Taiwan, R.O.C. (To be Submitted)
Landwehr, J. M., Pregibon, D. & Shoemaker, A. C. (1984). Graphical methods for assessing
logistic regression models (with discussion). J. Amer. Statist. Assoc. 79, 61-83.
McCullagh, P. (1991). Quasi-likelihood and estimating functions. In: D. V. Hinkley, N. Reid
& E. J. Snell (Eds), Statistical Theory and Modelling: In Honour of Sir David Cox,
FRS, London: Chapman & Hall, pp. 265-86.
McCullagh, P. & Nelder, J. A. (1989). Generalized Linear Models, 2nd ed. London: Chapman
& Hall.
Mittlb¨ock, M. & Schemper, M. (1996). Explained variation for logistic regression. Statist.
Med. 15, 1987-97.
Mittlb¨ock, M. & Schemper, M. (2002). Explained variation for logistic regression – Small
sample adjustments, confidence intervals and predictive precision. Biom. J. 44, 263-72.
Nagelkerke, N. J. D. (1991). A note on a general definition of the coefficient of determination.
Biometrika 78, 691-2.
Nelder, J. A. & Wedderburn, R. W. M. (1972). Generalized linear models. J. Roy. Statist.
Soc., Ser. A 135, 370-84.
Neuhaus, J. M. & Jewell, N. P. (1993). A geometric approach to assess bias due to omitted
covariates in generalized linear models. Biometrika 80, 807-15.
Pregibon, D. (1981). Logistic regression diagnostics. Ann. Statist. 9, 705-24.
Schemper, M. (2003). Predictive accuracy and explained variation. Statist. Med. 22, 2299-
308.
Searle, S. R. (1971). Linear Models. New York, NY: John Wiley & Sons.
Tsai, T. -L., Shau, W. -Y. & Hu, F. -C. (2003). Generalized path analysis and generalized
simultaneous equations model for recursive systems with responses of a mixed type.
Unpublished manuscript, Division of Biostatistics, Graduate Institute of Epidemiology,
College of Public Health, National Taiwan University, Taipei, Taiwan, R.O.C. (Submitted
to Structural Equation Modeling)
Wang, P. -C. (1985). Adding a variable in generalized linear models. Technometrics 27,
273-6.
Wang, P. -C. (1987). Residual plots for detecting nonlinearity in generalized linear models.
Technometrics 29, 435-8.
Wedderburn, R. W. M. (1974). Quasilikelihood functions, generalized linear models and the
Gauss-Newton method. Biometrika 61, 439-47.