Forecasting

Forecasting can be tricky but it's an important aspect to planning production, fulfilling sales, and managing inventory.

All too often in certain corporate settings projections are done by someone leaning back in their chair and rubbing their tummy, feeling out their gut.

While forecasting can be complex, one of the most basic and useful tools we have is the linear regression.

Doing the math by hand is possible but fortunately computers can do it for us.

A couple important things to know regarding the model’s usefulness: 1. Your P-value should be below your significance level. 2. The R-squared value relates to the strength of the correlation in the data.

If your correlation is very high and the model output doesn’t seem to be working, you may need to evaluate your data set. Real world data rarely has an R-squared above 0.9; it can vary heavily depending on what your data set is attempting to describe.