use the term multiple regression analysis began Pearson (Pearson) in the works, dated 1908 year yet.He described it as an example of the agent conducting the sale of real estate.In his notes specialist trade houses led the account a wide range of source data of each individual building.By results of trades determines which factor had the greatest impact on the price of the transaction.
analysis of a large number of transactions gave interesting results.On the final cost was influenced by many factors, sometimes leading to a paradoxical conclusion and even explicitly "emissions" when the original house with high potential was sold at a reduced price index.
second example of the use of such an analysis, see the work of a specialist on the staff, which was entrusted with the definition of employee benefits.The challenge lay in the fact that not require a fixed amount for each distribution, and strict adherence to its values of specific work performed.The emergence of a variety of tasks that are almost similar variant solutions, require a more detailed review at a mathematical level.
in mathematical statistics was given a significant place under the section "regression analysis" in it united practical techniques used to study addictions fall under the concept of regression.These relationships are observed between the data obtained in the course of statistical research.
regression analysis among the many major tasks set itself three objectives: to define the regression equation of the general form;construction estimates of the parameters that are unknown, that are part of the regression equation;statistical regression test hypotheses.In the course of studying the relationship that arises between a pair of values derived from experimental observations and a number of components (set) type (x1, y1), ..., (xn, yn), based on the theory of regression and assume that for one valueY there is a certain probability distribution, despite the fact that another X remains fixed.
result Y depends on the value of X, this dependence can be determined by various laws, and the accuracy of the results influence the nature and purpose of the analysis of observations.The experimental model is based on certain assumptions, which are simplified but plausible.The main condition is that the value of the parameter X is controlled.Its values are set to the beginning of the experiment.
If in the course of the experiment, a pair of uncontrolled variables XY, the regression analysis carried out by the same method, but for the interpretation of the results, in which we study the connection study random variables, correlation analysis methods are used.Methods of mathematical statistics are not an abstract theme.They find application in my life in various spheres of human activity.
in the scientific literature to determine the above mentioned method has found widespread use of the term linear regression analysis.To use the term of X regressor or predictor and dependent variables Y-also called criterial.This terminology reflects only the mathematical relationship variables, but not investigative causal relationship.
Regression analysis is the most common method used in the course of processing the results of a wide variety of observations.Physical and biological study according to their means of this method, it is implemented in the economy and in technology.Weight of other areas using regression analysis models.Analysis of variance, design of experiments, statistical analysis of multivariate closely with this method of learning.
