Skip to main content

Table 4 Univariate Probit Model Results

From: Non-linear mixed models in the analysis of mediated longitudinal data with binary outcomes

Sample Size

Effect Size

WLSMV

Bias ML-IRLS (Splus)

ML-IRLS (SAS)

250

0.3

1.7

1.4

1.5

500

0.3

0.4

0.3

0.3

750

0.3

0.6

0.5

0.5

900

0.3

0.1

0.007

0.007

1000

0.3

0.3

0.2

0.2

5000

0.3

-0.1

-1.6

-0.2

250

2.0

2.7

1.9

1.9

500

2.0

2.0

1.6

1.6

750

2.0

1.2

0.9

0.9

1000

2.0

1.0

0.8

0.8

500

-0.3

0.1

-0.05

-0.05

500

5.0

4.5

3.3

3.3

  1. Simulated univariate probit model with a single predictor and single binary outcome. Simulation results fit using weighted least squares with robust standard errors (WLSMV) in Mplus and maximum likelihood via iteratively re-weighted least squares (ML-IRLS) in Splus, and SAS.