I am trying to assess change in an independent variable based on a regression model. Basically I am trying to predict sales from the percentage of calls; at Baseline I predicted sales from percentage of calls and obtained a model. My next goal is to create a trend analysis of what the hypothetical sales would be if the percentage of calls was increased to a given value. Here is where it gets difficult.

If percentage of calls range from 0 - 100%, and I want to know what the number of sales would be if percentage of calls was at a minimum of 30% for everyone, how would I do it? I created a new set of data based on the average of the differences between participants who were below 30%, and added the average to their old value. If their old value was still not 30%, they were automatically bumped to 30%. So essentially every participant had their percentage increased by the average difference; is this the proper way to do it?

From what I can tell this is a forecasting procedure, but I want to know what the number of sales would be if the percentage of calls is at a minimum of 30%, then 40%.

Any feedback is helpful and appreciated! Thanks.