You get a bonus - 1 coin for daily activity. Now you have 1 coin

Regression analysis

Lecture



Regression analysis is a statistical method for studying the influence of one or several independent variables.   Regression analysis on dependent variable   Regression analysis . Independent variables are otherwise called regressors or predictors, and dependent variables are criterion variables. The terminology of the dependent and independent variables reflects only the mathematical dependence of the variables ( see False Correlation ), and not cause-effect relationships.

Content

  • 1 Regression analysis goals
  • 2 Mathematical definition of regression
  • 3 Least squares method (calculation of coefficients)
  • 4 Interpretation of regression parameters
  • 5 See also
  • 6 Literature

Regression analysis goals [edit]

  1. Determination of the degree of determinism of criterion (dependent) variable variation by predictors (independent variables)
  2. Predict the value of the dependent variable using independent (s)
  3. Determination of the contribution of individual independent variables to the variation of the dependent

Regression analysis cannot be used to determine whether there is a relationship between variables, since the existence of such a relationship is a prerequisite for applying the analysis.

Mathematical definition of regression [edit]

Strictly regression dependence can be defined as follows. Let be   Regression analysis - random variables with a given joint probability distribution. If for each set of values   Regression analysis conditional expectation defined

  Regression analysis (regression equation in general),

that function   Regression analysis called magnitude regression   Regression analysis by magnitude   Regression analysis and its graph is a regression line   Regression analysis by   Regression analysis or regression equation .

Addiction   Regression analysis from   Regression analysis manifested in a change in average values   Regression analysis when it changes   Regression analysis . Although at each fixed set of values   Regression analysis magnitude   Regression analysis Remains a random variable with a certain distribution.

To clarify the question, how accurately is the regression analysis evaluating the change   Regression analysis when it changes   Regression analysis , average variance is used   Regression analysis with different sets of values   Regression analysis (in fact, we are talking about the scattering measure of the dependent variable around the regression line).

In matrix form, the regression equation (SD) is written in the form:   Regression analysis where   Regression analysis - error matrix. With the reversible matrix X◤X, we obtain the column vector of the coefficients B, taking into account U◤U = min (B). In the particular case for X = (± 1), the X◤X matrix is ​​rotable, and the SD can be used in the analysis of time series and the processing of technical data.

Least squares method (calculation of coefficients) [edit]

In practice, the regression line is most often sought as a linear function.   Regression analysis (linear regression) that best approximates the desired curve. This is done using the method of least squares, when the sum of squared deviations of the actually observed   Regression analysis from their ratings   Regression analysis (referring to estimates using a straight line that pretends to represent the desired regression dependence):

  Regression analysis

(   Regression analysis - sample size). This approach is based on the well-known fact that the sum appearing in the above expression takes the minimum value for the case when   Regression analysis .

To solve the regression analysis problem using the least squares method, the notion of the residual function is introduced:

  Regression analysis

The condition of the minimum of the residual function:

  Regression analysis

The resulting system is a system   Regression analysis linear equations with   Regression analysis unknown   Regression analysis .

If we represent the free terms of the left side of the equations by the matrix

  Regression analysis

and the coefficients for the unknowns on the right side are matrices

  Regression analysis

then we get the matrix equation:   Regression analysis which is easily solved by the Gauss method. The resulting matrix will be a matrix containing the coefficients of the equation of the regression line:

  Regression analysis

To obtain the best estimates, it is necessary to fulfill the prerequisites of the OLS (Gauss – Markov conditions). In the English literature, such estimates are called BLUE ( Best Linear Unbiased Estimators - “the best linear unbiased estimates”). Most of the dependencies studied can be represented using MNC-nonlinear mathematical functions.

Interpretation of regression parameters [edit]

Options   Regression analysis are private correlation coefficients;   Regression analysis interpreted as a fraction of the variance Y, explained   Regression analysis , while fixing the influence of other predictors, that is, it measures the individual contribution   Regression analysis in the explanation of Y. In the case of correlating predictors, there is a problem of uncertainty in the estimates, which become dependent on the order of inclusion of predictors in the model. In such cases, it is necessary to use methods of analysis of correlation and step-by-step regression analysis.

Speaking about non-linear models of regression analysis, it is important to pay attention to whether it is non-linear with independent variables (from a formal point of view easily reduced to linear regression), or non-linear with estimated parameters (causing serious computational difficulties). With the nonlinearity of the first type from the substantive point of view, it is important to distinguish the appearance in the model of members of the form   Regression analysis ,   Regression analysis indicating the presence of interactions between signs   Regression analysis ,   Regression analysis etc. (see. Multicollinearity).

See also [edit]

created: 2014-11-06
updated: 2024-11-13
352



Rating 9 of 10. count vote: 2
Are you satisfied?:



Comments


To leave a comment
If you have any suggestion, idea, thanks or comment, feel free to write. We really value feedback and are glad to hear your opinion.
To reply

Probability theory. Mathematical Statistics and Stochastic Analysis

Terms: Probability theory. Mathematical Statistics and Stochastic Analysis