The ERP contest is about prediction of the stock market. The objective is to build a predictive model for weekly S&P 500 returns.

For this purpose Hull Tactical is providing you with this data set of variables that may aid this task. There are numerous ways to build a prediction model, ranging from a simple linear regression to more sophisticated methods from machine learning.
The data file above contains various financial variables over a span of 1952-2018. A brief description of each of these variables may be found here . To get started on building an intuition as to why these may hold predictive power, you may look at the Baltic Dry Index (BDIY), the Volatility Index (VIX), calendar effect (SIM, TOM) and sentiment index (Hulbert). You can find all of these variables in the description files linked above.

** Please do not use the Hull Tactical data set for any purpose other than the ERP contest.