# Calculating t-statistics and p-values

In this section, we will apply our knowledge of how to ascertain whether the variables have a strong relationship to predict new values with a regression model. The complexity, in this case, is that we are not working with just one predictive variable. Now, we are working with two or more variables in a multiple linear regression model.

The statistical probes to determine the significance of the relationship are as follows:

- Determination coefficient
- Correlation coefficient
- t-statistics
- p-value

To find the coefficients, we have to review the distances between the average of the expected values and the linear model.

*Figure 10.8* is the 3D chart of the variations we saw in the previous chapter. The errors or unexplained variations (**sum of squares errors** (**SSEs**)) are the distance between the expected values and the linear...