Presentation is loading. Please wait.

Presentation is loading. Please wait.

Penalized Maximum Likelihood Logistic Regression

Similar presentations


Presentation on theme: "Penalized Maximum Likelihood Logistic Regression"— Presentation transcript:

1 Penalized Maximum Likelihood Logistic Regression
Joseph Coveney Cobridge Co., Ltd.

2 Topics Separation in Logistic Regression Approaches to Separation
Firth’s Bias-reduced GLMs firthlogit: syntax and examples Caveats and to-do’s

3 Separation in Logistic Regression

4

5 Complete Separation Dataset adapted from D. W. Hosmer and S. Lemeshow, Applied Logistic Regression Second Edition. (New York: John Wiley & Sons, 2000), pp. 138–39.

6 Quasi-complete Separation
Dataset adapted from D. W. Hosmer and S. Lemeshow, Applied Logistic Regression Second Edition. (New York: John Wiley & Sons, 2000), pp. 138–39.

7

8 Approaches to Separation
Remove predictors Pool groups Remove interaction terms Gather more data Use alternatives

9 Exact Logistic Regression

10 But . . . Dataset from D. M. Potter A permutation test for inference in logistic regression with small- and moderate-sized data sets. Statistics in Medicine 24:693–708.

11

12 [19] D. Firth. 1993. Bias reduction in maximum likelihood estimates
[19] D. Firth Bias reduction in maximum likelihood estimates. Biometrika 80:27–38.

13 firthlogit

14

15

16 But redux

17 But redux, continued

18

19

20 Profile Likelihood Ratio CIs

21

22 Caveats Profile Penalized Likelihood CIs Small-sample Behavior

23 G. Heinze and M. Ploner, A SAS macro, S-PLUS library and R package to perform logistic regression without convergence problems. Technical Report 2/2004. Medical University of Vienna. p. 36.

24

25

26 To-do’s Profile Penalized Likelihood CIs Modify ml d0


Download ppt "Penalized Maximum Likelihood Logistic Regression"

Similar presentations


Ads by Google