Changes in operating efficiency
factor.formula
The calculation formula of operating efficiency change factor is:
in:
- :
The operating income of the i-th quarter represents the total income obtained by the enterprise through its main business activities in that quarter.
- :
The operating costs in the i-th quarter represent the direct costs incurred by the enterprise to achieve operating income in that quarter.
- :
The intercept term of the regression model represents the expected operating income level when the operating cost is zero. In actual business scenarios, it can usually be regarded as the impact of fixed costs.
- :
The slope term of the regression model represents the expected change in operating income for every unit change in operating cost. It can reflect the efficiency of income output that can be brought about by the unit cost input of the enterprise.
- :
The residual of the regression model in the i-th quarter represents the difference between the actual operating income and the operating income predicted by the model. A positive residual means that the actual income is higher than the model expectation, indicating that the operating efficiency has improved compared with the historical trend; a negative residual means that the actual income is lower than expected, indicating that the operating efficiency has declined compared with the historical trend. This residual is used as the core value of the operating efficiency change factor.
- :
i ∈ {0, 1, 2, ..., N-1}, represents the index of the time series, where 0 represents the most recent quarter and N represents the length of the historical quarters to be traced back. The default value is N = 8, which means that the data of the most recent 8 quarters are traced back.
factor.explanation
The calculation steps of this factor are as follows:
-
Data preparation: Take the company's operating income (Revenue) and operating cost (Cost) data for the most recent N quarters (default N=8).
-
Data preprocessing: Perform Z-Score standardization on the operating income and operating cost data respectively. Z-Score standardization converts the data into a standard normal distribution with a mean of 0 and a standard deviation of 1, eliminating the impact of different dimensions and orders of magnitude, making the data between different companies comparable.
-
Linear regression: Take the standardized operating income as the dependent variable, and perform ordinary least squares (OLS) linear regression on the standardized operating cost. Establish the model: $Revenue_i = \alpha_i + \beta_i Cost_i + \epsilon_i$. The purpose of this regression model is to fit the linear relationship between historical operating income and operating costs.
-
Residual extraction: Get the residual value $\epsilon_0$ of the regression model in the most recent quarter (i.e., quarter 0, i=0). This residual value is the operating efficiency change factor value of the day. A positive residual indicates that the operating efficiency of the quarter is higher than the historical level, and a negative residual indicates that the operating efficiency of the quarter is lower than the historical level.
-
Factor explanation: The size of the factor value represents the degree to which the operating efficiency of the quarter deviates from the historical trend. A positive factor value indicates that the operating efficiency of the quarter has improved. The larger the factor value, the greater the improvement. On the contrary, a negative factor value indicates that the operating efficiency of the quarter has decreased. The smaller the factor value, the greater the decrease.
This factor value can help investors judge the short-term change trend of the company's operating efficiency, thereby assisting investment decisions.