At the moment mrds doesn't standardise covariates before performing optimisation. This can lead to numerical problems when finding parameter estimates. One can simply standardise an existing variable by subtracting the mean and dividing by the standard deviation, creating a new variable; but it would be nice to have the functionality in mrds by default.
At the moment mrds doesn't standardise covariates before performing optimisation. This can lead to numerical problems when finding parameter estimates. One can simply standardise an existing variable by subtracting the mean and dividing by the standard deviation, creating a new variable; but it would be nice to have the functionality in mrds by default.