General linear mixed model - 5 years

#1
Dear all,
every help is highly appreciated!
I have a data set that looks like the one I attached to this post.
I have five independent variables for each company (for each time period one). Now I need to know, if the values significantly increase over the time of five year. In other words I need to find out, if the time has a significant influence on the independent variable. Can anyone help me in figuring out which test to use or how to solvev my problem?
Thank you very much for your efforts,
Best, Max
 
#2
Hi, indeed, you should use a Mixed Model to investigate that. Fist you can try a Linear Mixed Model with time (year) as the independent continuous variable and Company as the random effect. This accounts for the fact that you have repeated measurements regarding each company. You can use a random intercept and slope model, if the algorithm does not converge, you can use only random intecept. After model fiting, you should look at the residuals. If they are normally distributed (evalated e.g. via QQ-Plot), everything is fine. If not, you could think about a GLM or robust methods, like bootstrapping.