# Should I convert age as independent variable in my Quantile Regression?

I'm applying quantile regression to a dataset where the dependent variable is a measurement of load (utilization) of a specific technology. The model includes a number of independent variables including distance between each participant and a shared office, and participant age.

Until now, I've just defined age and distance as a numerical ordinal variable when doing quantile regression using the quantreq package in R. However, a supervisor (who mainly works with logistic regression) told me that age (and probably distance as well), unless they had a linear relationship with the dependent variable should be converted into categorical variables before doing the regression analysis.

Is this transformation really necessary as Quantile Regression papers repeatedly state that QR isn't bound by distributions of normality?

The distribution of age is shown in the following histogram:

And the next figure shows a plot of age and load for the data. This figure also shows the skewed distribution of the dependen variable.

As I've understood Quantile Regression, it should not be necessary to convert the age variable as we look at the regression coefficient for age in a given quartile, and not across the entire distribution?