SPSS data Analysis-Multiple linear regression

Only one independent variable and the linear regression of the dependent variable are called simple linear regression, but in fact, such a simple relationship in the real world almost does not exist, all things are interconnected, a problem must be produced by a number of factors combined effect of the results.

For linear regression with multiple independent variables and a dependent variable called multiple linear regression, some data are called multivariate linear regression, but I think the meaning of the plurality should be the true dependent variable rather than the independent variable, and the multi-collinearity is also generated for multiple independent variables, so I still agree that it is called multiple linear regression.

The multiple linear regression is similar to the applicable condition and simple linear regression, and is also the linear relationship between the independent variable and the dependent variable, the residual variance of the residuals, the residuals are normally distributed, but because the independent variables are more than 1, it is also necessary to require that there is no correlation between the independent variables, that is, there is no multiple collinearity. However, there is no relevant two variables that are not present, so the conditions are relaxed to be acceptable as long as they are not strongly correlated.

Multiple linear regression in the process of SPSS and simple linear regression, just the content of a few more, and because of the more information, it is recommended to set the analysis steps, the commonly used steps for
1. Plot scatter plots to determine if there is a linear trend
2. Preliminary modeling, including setting variable selection methods
3. Residual analysis, analysis of the residual error after modeling the normality, independence, variance homogeneity and other issues
4. Strong impact point and multi-collinearity judgment
5. Revise the model based on the above analysis and repeat 3-4 until the model achieves optimal results.

ANALYSIS-Regression-linear

SPSS data Analysis-Multiple linear regression