Cross-classified multilevel model


I am updating the estimations made in this paper (, in which they try to analyze who is the best driver in F1 history and the influence of the driver and the car taking as dependent variable the points achieved in each race, using for this purpose a cross classified multilevel model.

To replicate their estimations it is not complicated, applying the following code:

Model <- lmer(zPoints ~ 1 + (1|Team) + (1|Driver) + (1|TeamYear),
              data = Data, REML = FALSE)

The problem lies in what I want to see how the random part of the model varies, i.e. the variance according to driver, team and TeamYear, for each year. For this, I have a variable Year indicating the year in which the points are obtained (1950-2020). They put the variable Year in the fixed part and as a predictor and for each of the variables of the random part, but then to obtain the variance for each of these variables according to the year, they calculate variance functions. How can I calculate this in Rstudio?

Finally, I want to see which driver and team have the highest variance for each year, which would explain most of the points obtained, and, therefore, would be the best driver and team each year. How can I do this?

This topic was automatically closed 54 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.