CIMR - Creating Improved Macroeconomics Reasoning

Many people, ranging from CEOs of major companies to the average person trying to figure out what they should invest in, benefit from predictions of macroeconomic measures like GDP or unemployment rates. Professional forecasters make these kinds of predictions regularly through the survey of professional forecasters put out by the Philly Fed. However, research has proven that not only are these predictions inaccurate, but that these forecasters are also overconfident in their predictions.Given the broad applications of these predictions, it’s important that they are as accurate as possible. So, our project attempts to make more accurate forecasts of these macroeconomic measures. By taking into account forecasters’ history of overconfidence and inaccuracies, as well as historical economic data, we can produce a more accurate prediction using machine learning techniques.