Dependent and independent variables are statistical concepts that come into play when trying to make numerical predictions. Since stock investors often rely on statistics, you may hear such terms frequently in research reports. What a particular analyst will define as the dependent variable varies based on the model used. Therefore, a basic understanding of statistical modeling is necessary for any stock investor.
Statisticians often build models to predict values of such things as weather temperature, inflation rate or the price of a specific stock. To predict the numeric values of these phenomena, values of other items are often used. In such a model, the item whose value you are trying to predict is the dependent variable. The input variables, which you will use to predict the dependent variable, are referred to as independent variables. When trying to predict inflation rates in a country, you may use its population, imports, exports, birth rates and so on, all of which are independent variables; inflation is the dependent variable.
The dependent variable in a statistical model for stock investors is usually the stock's price. Since investors are primarily concerned with the value of shares, most models try to predict what the share will be worth in the future. There are exceptions, however. An analyst may desire to know how many shares will change hands, in which case the dependent variable becomes the trading volume. In other cases, analysts try to predict a value called volatility. Volatility measures how wildly the stock's price changes and is therefore a good predictor of risks associated with holding that stock.
The input that the analyst uses to predict the stock price is the independent variable. The level of a popular stock market index, such as the S&P 500, or the profit per share of the stock's issuing corporation are some of the commonly used independent variables. In some instances, the analyst uses the price of one stock to predict that of another. This is usually done if two stocks are issued by competing corporations, whose stock prices have exhibited similar movements in the past. In such cases, the independent and dependent variables are both stock prices.
While there is always one dependent variable in a model, there may be multiple independent variables. Such models are referred to as multiple regression analysis. The analyst may, for example, attempt to predict the price of a stock by using the debt-to-asset ratio, profit per share and dividend per share of the issuing corporation. In theory, there is no limit to the number of independent variables that can be used. However, the results of excessively complex models with too many variables can be misleading. Most analysts therefore limit the number of independent variables to anywhere from three to five.
- Comstock Images/Stockbyte/Getty Images