The term market regime is another way to describe marketing additions that broadly affect how stocks behave during a period of time. Market regimes change over time, so it’s helpful to generate features that capture changes in these market conditions. There are two market regime features that we will work with. One is market dispersion and the other is market volatility. Market dispersion looks at the cross-section of stocks and how spread out they are. To think about this conceptually, we can just look at the top-performing and lowest-performing stock on a single day. The difference between them can vary from day-to-day. On days when the best and worst performing stock are not that much far apart, there isn’t much benefit from correctly choosing the correct stocks to go long or short. On the other hand, on days when the best stock is doing very well and the worst stock is doing very poorly, then choosing the correct stocks to go long or short has a more significant impact on a fund’s performance. Market dispersion can be calculated as a standard deviation. To visualize this idea, we can think of looking at the Stock Universe on a single day and seeing how dispersed or spread out they are from the center. As with other features, we will utilize time horizons of one and six months. Market dispersion measures how much investors are making a distinction between stocks. If cross-sectional dispersion is low, it means all stocks are correlated. If a major company stock goes up, others go with it. If cross-sectional dispersion is high, the stocks are likely moving in different directions. So people are paying attention to idiosyncratic aspects of stocks rather than common factors of stocks.