Notre Dame CSE 60647 Data Science Project (Fall 2018)
Shane Johnson, Catherine Markley, Yuanhao Niu, Nick Rocco
We try to use machine learning methods to predict equity premium based on predictors from the seminal paper of Goyal and Welch (2008).
We first build and clean our dataset. Then we try to predict bull/bear market as a classification job. Portfolio based on the prediction were constructed accordingly. In a parallel task, we use different regression approaches to predict equity premium. Our models outperforme the historical average forecast.