Skip to main content

Table 2 Details of the patterns of missingness in the ovarian cancer study 18

From: Comparison of imputation methods for handling missing covariate data when fitting a Cox proportional hazards model: a resampling study

Pattern (R i )

Stage

POST-RT

Age

Frequency Probability (p i )

Cumulative probability

0

1

1

1

  

1

1

1

0

0.08

0.08

2

1

0

1

0.17

0.25

3

1

0

0

0.04

0.29

4

0

1

1

0.25

0.54

5

0

1

0

0.04

0.58

6

0

0

1

0.34

0.92

7

0

0

0

0.08

1.00

% missing out of total incomplete cases

71

64

22

  
  1. Key: 1 = observed and 0 = missing