# Calculating the intercept and slope with formulas

The main components of a linear model are as follows:

- Intercept
- Slope

The **regression slope** defines the difference between the expected values and those of the model. From here, we calculate the first check to determine whether the variables have a relationship and are useful to build a predictive model. We have accepted the hypothesis that the variables are related, and we can use them to build a predictive model. The first check includes the coefficients of determination and correlation.

The **slope** indicates whether the data has a direct or an inverse relationship. It is probable that the predictor value, *X*, grows, while the result variable, *Y*, decreases. In this case, we have an inverse relationship with a negative slope. A slope with a value equal to zero (flat) means there is no relationship between the model variables, predictors, and effects. We use the *t*-statistics test to probe the hypothesis that the slope...