From the course: Data, Economic Modeling, and Forecasting with Stata
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Overly tight model fit - Stata Tutorial
From the course: Data, Economic Modeling, and Forecasting with Stata
Overly tight model fit
- There are several big risks when you're building forecast models. And one of the biggest risks that's also counterintuitive is that sometimes you can build a model that fits historical data too closely. I know that sounds crazy, and it's true that forecast models leverage historical data in order to help you make future predictions. But, if the models fit too tightly to historical data, that means those models may not be very effective at adapting to new changes in future data. This is something statisticians and econometricians refer to as a problem of fit. And it is a real issue where sometimes that perfect model for predicting the past is not good when data in the future is released. There are a couple of ways you can test for this. First, you could test some of the historical data to build a model and then test it against other historical data that comes after the time period for which you built the model. If it's…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.