Fundamental Trends Imply Returns
factor.formula
Calculate the moving average of the kth fundamental factor of stock i at the end of month t with a time scale of L quarters:
At the end of each month, a cross-sectional regression analysis is performed on the stock returns for the next period (t+1) using the moving average of the fundamental factors of all stocks in the past L quarters:
At the end of each month, the fundamental trend implied return factor is calculated based on the expected value of the regression coefficient and the current fundamental moving average:
in:
- :
is the kth fundamental factor value of stock i in month t, lagged by j quarters. For example, if j=0, it means the factor value of the current month t, and if j=1, it means the factor value of quarter t-1.
- :
is the moving average of the kth fundamental factor of stock i in the tth month based on the past L quarters, and the value of L is 1, 2, 4 and 8, representing the past 1, 2, 4 and 8 quarters respectively.
- :
is the return of stock i in month t+1, which is used to measure the future performance of the stock.
- :
is the total number of different fundamental factors selected. In this factor study, K=7, and the selected factors include: return on equity (ROE), return on total assets (ROA), earnings per share (EPS), accrual-based profit to equity (APE), cash flow-based profit to total assets (CPA), gross profit to total assets (GPA), and net payout ratio (Net Payout Ratio). These factors describe the fundamentals of listed companies from multiple perspectives, including profitability, operating efficiency, and shareholder returns.
- :
It is the intercept term of the cross-sectional regression model in month t, representing the benchmark return level.
- :
It is the regression coefficient corresponding to the moving average of the fundamental factor k on the time scale L in the tth month in the cross-sectional regression model, indicating the expected impact of each unit change in the moving average of the factor on the next period's stock return.
- :
is the expected value of the regression coefficient for the next period (t+1) in the tth month in the cross-sectional regression model. In practical applications, this value can be approximately replaced by the regression coefficient of the current period (t).
- :
It is the error term of the cross-sectional regression model, which indicates the part of the rate of return that cannot be explained by the model.
factor.explanation
The fundamental trend implied return factor (FIR) takes into account multiple fundamental indicators and their trend changes at different time scales. Through the cross-sectional regression method, the predictive ability of the moving average of the fundamental factor on future stock returns is extracted, and the final factor value is constructed based on the expected value of the regression coefficient and the current factor moving average. The construction concept of this factor is to capture the investment opportunities contained in the trend changes of the company's fundamentals. The higher the factor value, the better the company's fundamental trend and the greater the potential for future returns.