longitudinal multivariate data - multivariate repeated measures


New Member

Am dealing with a multivariate longitudinal data where we collect data from two group of subjects (cases and controls) at baseline , one month , two month, and at three months . At each of these four time points we collect data (which basically consist of 4000 variable of which many of them are highly correlated, similar to NIR or nmr spectra). There is also clinical measurements such as hight, weight, biochemistry data etc which are also collected at the same time)

The idea is to identify the differences in the groups as well as the differences occuring with time

Any suggestion for the best data analysis methods?

Thanks in advance :)
For differences within groups occurring over time, you could try latent growth curve modeling. It will tell you how much change occurs for subjects within each group. Then you could compare the change rates between the groups.

The other way is to calculate the intercept and slope for your dependent variable measured over the four points in time. Then you can use your independent variables to predict the intercept and slope for each subject. And finally, you can compare the results between the two groups. I hope this helps.