From the course: Machine Learning with Data Reduction in Excel, R, and Power BI

Unlock the full course today

Join today to access over 24,800 courses taught by industry experts.

Correlation and covariance

Correlation and covariance

From the course: Machine Learning with Data Reduction in Excel, R, and Power BI

Start my 1-month free trial Buy for my team

Correlation and covariance

“

- [Instructor] Statistical correlations tell us both the strength of our relationship between two model variables and the direction of this relationship. Statistical variance tells us about the spread of the data points for a single model variable. Covariance, like correlation, tells us the direction of the relationship between two model variables, But unlike correlation, covariance doesn't calculate the strength of the relationship, but instead indicates the spread between two variables. Variance, covariance, and correlation are going to all have positive and negative values, but the range of values for variance and covariance can range from negative infinity to positive infinity, while the range of values for correlation can only go from negative one to positive one inclusive. We can calculate variance and covariance either for the sample or the entire population. The sample population will always be smaller than the…

Contents