I have observed sea surface temperature from 1900-2008 on the x axis of a graph and observed number of storm days for the same period on the Y axis. I have used quadratic regression in excel to produce a regression equation for these two variables.
I then have some future predicted sea surface temperatures (X Values)for 100 years in the future. Can i substitute these future temperatures (X values) into the regression equation to model the (y) Number of future days storms will occur. Or have i made some fundamental error?
Apologies for the delay. Yes, you can indeed extrapolate the number of future storm days. You have to keep in mind though that the quality of your predictions would depend on the correlation between the variables, and usually has to be strong. The confidence interval should also be stated, this makes it possible for someone else to understand the accuracy of the prediction.