Skip to main content
. Author manuscript; available in PMC: 2018 Mar 30.
Published in final edited form as: Pharm Stat. 2017 Dec 19;17(2):182–197. doi: 10.1002/pst.1845

Table 3.

GLM estimated through MLE with and without Firth correction for T = 100, J = 5, b = 20 in a case of scenario (i) with D = 0.24. Results for 5000 trials. True values were assumed to be β0 ≈ −0.8473, βt ≈ 0.2719 and βz = β1 = 0.

(I) GLM fitting without correction for CR

E (β̂i) E (MSE) E (pvalue < 0.05)

β̂0 -0.8684 0.1992 0.5174
β̂t 0.2610 0.0243 0.4018
β̂1 0.0070 0.1900 0.0544

(II) GLM fitting with correction for CR

β̂0 -0.8370 0.1838 0.5224
β̂t 0.2509 0.0227 0.4012
β̂1 0.0067 0.1775 0.0534

(III) GLM fitting without correction for FLGI

β̂0 -1.4465 8.9957 0.4110
β̂t 0.1898 0.0307 0.1844
β̂1 0.0038 18.2440 0.0142

(IV) GLM fitting with correction for FLGI

β̂0 -0.9307 0.3947 0.4670
β̂t 0.1825 0.0301 0.1858
β̂1 0.0048 0.7993 0.0456