In the formula, rt is the return rate of the trading asset at time t; bi is the regression coefficient of each order. The time points for trading decisions are the beginning of time t and the end of time t-1. At this time, the daily return rate for a total of N days from rt-1 to rt-N is known data. By calculating the above regression formula, we can get the relationship between t and t. The predicted value of the time rate of return. When the expected rate of return at time t is positive, the trading asset at time t is long. When the expected rate of return at time t is negative, the trading asset at time t is short. During the classification process, it is also necessary to ensure that the two classification criteria are combined to cover all possibilities. Therefore, when the predicted value is zero, it can be classified as many or as empty. #ETH #crypto2023 #Binance #BNB #xiaoyiclub