nessopk.blogg.se - Simple linear regression equation from stata

: This shows the 95% confidence interval for the coefficient.When investigating the relationship between two or more numeric variables, it is important to know the difference between correlation and regression.

Take 0.104 for example, it means the coefficient in front of variable depression is not significantly different from zero.

P>|t|: This gives the p-value for testing the significance of each Coef.

These are the t test statistic regarding whether Coef is significantly different from zero or not. Err., same as what we did in our t tests.

t: The values in this column are derived from taking the ratio of Coef.

Same as the t tests we’ve learned, standard error is used to compute the t test statistic. Err.: These are the standard errors associated with the coefficients.

Coef: This column gives regression coefficients.

_cons represents the constant (or the y-intercept). The variable on the top alcohol is the dependent variable, and the variables down below are all the independent variables.

alcohol: This column shows all the DVs and IVs.

In this source table df (degree of freedom) and MS (same as \(S^2\) in ANOVA test) are also given. Where SS Model and SS Residual are just equivalent to SS Between and SS Within in ANOVA test. These different sources of SS are not different from what we’ve seen in ANOVA test. SS Total gives the total variability in variable \(y\), the dependent variable (DV) SS Model gives the variability of the DV that can be explained by all the independent variables (IVs) using this linear regression model SS Residual represents the variability of DV that cannot be explained by the model.

Source Table: This gives results regarding whether the linear model as a whole is significant or not, or in other words whether all the coefficients in the linear regression model are zero.

The output needs to be further explained: The output can be found in the picture below: