test/testing notes 2020-12-03.md

In UCL/mimix: Reference-Based Imputation for Longitudinal Clinical Trials with Protocol Deviation

Notes on mimix causal

Successes in monotone imputation

I tested the impact of increasing fev by x units at time 4 in the treatment arm. This increases the treatment effect at time 4 by 1 without impact on the treatment effects at other times. I verified that this increases imputed values for patterns 0100 and 1100 only: by xK0K1^4 at time 8 by xK0K1^8 at time 12

Limitations

I haven't tested non-monotone imputation

Problems

Could allow K1 not specified if K0=0

Mistake at runmimix.R line 53: running mimix with options method="causal",K0=8,K1=0.5 wrongly fails with error K1 Causal constant not in range 0..1

Can idvar be >1 variable? not at present!

How do I carry another variable into the imputed data set?

UCL/mimix documentation built on Sept. 7, 2024, 5:26 p.m.