Instructions for Use

The summary of generalized linear regression results refers to the establishment of a generalized linear regression model after the training process is completed, which can obtain corresponding analysis results, evaluate the effectiveness of the trained model, and select the output parameters as needed. The return parameters include: *Variable: An array of field names in a generalized linear regression model, referring to the fields of the independent variables in the training model. *Coefficent: Regression coefficient. *CoefficientStandardErrors: The standard error between the regression coefficient and the intercept. *TStatistic: The T-statistic of the regression coefficient and intercept. *Probability: The probability of regression coefficient and intercept. *AIC: The AIC criterion of the model (minimum information criterion). Can be used to test model performance and compare regression models. Considering the complexity of the model, models with lower AIC values will better fit the data. AIC is not an absolute measure of fit, but it is very useful for models that are more applicable to the same dependent variable and have different explanatory variables. *Dispersion: The discretization of a generalized linear regression model. *DegreesOfFreedom: Degrees of freedom. *ResidualDegreeOfFreedomnull: The residual degree of freedom of the zero model. *ResidualDegreeOfFreedom: Residual Degrees of Freedom.

Parameter Description