how to deal with missing values ?


New Member
I have around 80 measurements done for 30 participants over 10 different time points.
There is some missing data:

1- some participants have missing values for some measurements.
2- Some participants have measurements done over X time points where X < 10.

I want to check if there was an improvement in certain measurements over time. But how can I deal with missing values?
I cannot use default values. and I cannot ignore participants who have missing values, otherwise I might end up at the end with a very small number of participants.

help is appreciated.


Less is more. Stay pure. Stay poor.
You have to first devise a case for missingness pattern? At random, monotonic, etc.

Why do you have missing data?