Can you regress sasdate on a continuous variable

[RACHEL2425]

Is the following regression appropriate?

For example: 

PROC REG DATA=work.wastewater PLOTS = (FITPLOT);

where '30nov2021'd <= sasdate <= '9jan2022'd ;

MODEL LOGMGC = sasdate / clb;

RUN;

[PaigeMiller]

Of course, you CAN do this and get results. Is it appropriate? That depends on a lot of things, including an assumption that LOGMGC increases or decreases linearly as the day changes over this time period. Only you can determine if that is a good assumption for your problem.

More naturally, people use time series models to model the effect of changing time on some response, this does not impose a linear response, but allows for all different types of response of the variable LOGMGC to the change in day, including modeling the effect of auto-correlation, periodic changes, and other time series effects.