Skip to main content

Table 2 Simulation Results from 200 simulation runs. Fitted model is probit-log-skew-normal.

From: A probit- log- skew-normal mixture model for repeated measures data with excess zeros, with application to a cohort study of paediatric respiratory symptoms

PROBIT component

β < α 1

(negatively skewed)

β = α 2

(symmetric)

 

Bias

MSE 4

Bias

MSE

Intercept (-1)3

0.0303

0.0128

0.0054

0.0116

βViral Infection (2.5)

-0.0186

0.0130

-0.0064

0.0112

s11 (1)

-0.0927

0.0243

-0.0914

0.0235

Log-beta component

    

Intercept (-20)

    

βViral Infection (0.75)

0.0004

0.0050

-0.0090

0.0055

s22 (0.2)

-0.0059

0.0010

-0.0085

0.0012

s12 (0.2)

-0.0167

0.0028

-0.0199

0.0033

  1. 1. β = 70, α = 130
  2. 2. β = α = 100
  3. 3. Numbers in parentheses refer to true parameter values
  4. 4. Mean Square Error (MSE)
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us