I am analyzing the number of donors cancelling commitments as a monthly time series which varies significantly with economic indicators, some according to internal data, and maybe according to season. I want to see the variance after adjusting for the economy and season (assuming the remainder of variance is due to the non-profit's behavior). Is a good way to do this to fit a linear regression model in R pseudo-code:
fit <- lm(cancels ~ economy + season)
And then use the residuals?