Computing coefficient significance – t-statistics and p-value
In this section, we will see four statistical probes to verify whether the variables have a strong enough relationship to build a predictive model. First, we have to understand a chart with data variables, a linear regression model, and the separation between the data points and the straight line.
The statistical tests are as follows:
- Coefficient of determination
- Coefficient of correlation
- t-statistics
- P-value
Now, we are going to explain the basic concepts to test whether the variables have a relationship to build a predictive model. Figure 9.8 shows the following:
- The distance between the expected values and the linear regression model. These are the errors of the model or the unexplained variations.
- The distance between the expected values' average line and the regression model. This is the explained variation.
- The total variation, which is the sum of the...
