Regression analysis

MySales uses both simple and multiple regression analysis to break down sales into influencing factors such as seasonality, trend, price, discount, etc. However, the algorithms used in MySales are not limited to applications of simple and multiple, linear and nonlinear regressions. Gradient descent is also employed to build forecasts based on a larger number of factors than classical multiple regression allows. Additionally, neural networks are used both for sales forecasting and as a set of neural networks for promo forecasting. There is also an algorithm for automatically assessing the similarity of items in promotions, as well as other machine learning algorithms. The key distinction and know-how of all MySales algorithms is their predictable behavior under various conditions and low tendency to overfitting, also known as the "over-fit" effect. This is achieved because machine learning in MySales is performed within the framework of logical rules and constraints, set both by system parameters and by individual customer settings.

Regression analysis is a statistical method used to evaluate relationships between variables. It encompasses a variety of modeling methodologies and analyses of the behavior of multiple variables to determine the relationship between a dependent variable and one or more independent variables. Regression analysis helps to understand how the value of a dependent variable (or "target variable") changes when one of the independent variables changes, while other independent variables remain fixed. In most cases, regression analysis estimates the conditional expected value of the dependent variable given the values of the independent variables—that is, the average value of the dependent variable at fixed values of the independent variables. Less commonly, the focus is on a quantile or another parameter of the conditional distribution of the dependent variable, given the independent variables. In all cases, the goal of regression analysis is to reconstruct the functional dependence between the target variable and the independents, i.e., to determine the regression function.

Regression analysis is used for forecasting; its usage significantly overlaps with the field of machine learning. Regression analysis is also applied to find and study relationships between the dependent variable and independent variables. Under limited conditions, regression analysis can be used to determine causal relationships between independent and dependent variables. However, this may lead to errors or spurious relationships, so caution is necessary. For example, correlation does not imply causation.

Regression analysis is widely used in many fields, such as aerospace and defense, aviation, and astronomy, for forecasting or understanding the impact of external or internal factors. The effectiveness of regression methods in practice depends on the dataset being analyzed and the regression approach chosen for that dataset. Since the true form of the data-generating process is generally unknown, regression analysis often relies to some extent on assumptions about this process. These assumptions can sometimes be tested if there is sufficient data. Regression models for forecasting are often useful even if the assumptions are somewhat violated, although they may not be optimal.