Fundamental trend expected return
factor.formula
First, calculate the moving average of the fundamental factors for each stock over different time spans:
Secondly, at the end of each month, the moving average of the fundamental factors calculated in the previous period is used to perform a cross-sectional regression on the stock returns of the next period, thereby estimating the risk premium of different fundamental factors on future returns over different time spans:
Finally, the expected return of the stock is calculated using the risk premium coefficient estimated by regression and the moving average of the current fundamental factors:
in:
- :
It represents the value of the $k$th fundamental factor of stock $i$ in month $t$ after a lag of $j$ quarters. For example, if $j=0$, it is the current value; $j=1$ is a lag of one quarter.
- :
represents the moving average of the $k$th fundamental factor of stock $i$ in month $t$, with a time span of $L$ quarters. $L$ takes values of 1, 2, 4, and 8, representing a time span of 1 quarter, half year, year, and two years, respectively.
- :
represents the return of stock $i$ in month $t$, usually the return after taking into account dividend reinvestment.
- :
It represents the intercept term of the $t$ monthly cross-sectional regression and represents the overall market return level.
- :
It represents the regression coefficient of the $k$th fundamental factor in the $t$th monthly cross-sectional regression, with a time span of $L$th quarterly moving average, which represents the risk premium or return prediction ability of the fundamental factor in a specific time span. $E_t[\beta_{t+1,L}^{k}]$ represents the expectation of the regression coefficient in the future $t+1$th period. In practice, the current regression coefficient is usually used as the estimated value of the future coefficient.
- :
represents the residual term of the $t$-month cross-sectional regression, representing the specific return of stock $i$ that cannot be explained by the model.
- :
Indicates the number of fundamental factors selected for calculation. 7 fundamental factors are selected here: return on equity (ROE), return on total assets (ROA), earnings per share (EPS), accrual-based profit to equity ratio (APE), cash-based profit to total asset ratio (CPA), gross profit to total asset ratio (GPA), net payout ratio. These factors cover multiple dimensions such as profitability, operating efficiency, cash flow and shareholder returns, aiming to capture the value and growth of enterprises from different angles.
factor.explanation
This factor (FIR) combines the multi-quarter moving averages of multiple fundamental factors through a cross-sectional regression method and correlates them with future stock returns. The core idea is that fundamental trends have a certain degree of persistence and can predict future returns. This factor takes into account the absolute level of fundamental information and its trend over time, aiming to discover stocks with continued improvement in fundamentals and estimate their expected returns based on their risk premium. The higher the expected return, the higher the investment value of the stock. This factor combines the concepts of value investing and trend tracking.