Covariance and correlation

Practicing Statistics Interview Questions in R

Zuzanna Chmielewska

Actuary

a factory

a linear relationship

Covariance

Formula for a sample: $$ cov(X, Y) = \frac{\sum_{i = 1}^{n} (x_i - \overline{x}) \cdot (y_i - \overline{y})}{n-1} $$

Formula for a population: $$ cov(X, Y) = \frac{\sum_{i = 1}^{n} (x_i - \overline{x}) \cdot (y_i - \overline{y})}{n} $$

Formula for a sample: $$ cov(X, Y) = \frac{\sum_{i = 1}^{n} (x_i - \overline{x}) \cdot (y_i - \overline{y})}{n-1} $$

Formula for a population: $$ cov(X, Y) = \frac{\sum_{i = 1}^{n} (x_i - \overline{x}) \cdot (y_i - \overline{y})}{n} $$

$ x_1 = 3, x_2 = 5, x_3 = 7 $

$ y_1 = 6, y_2 = 11, y_3 = 13 $

$ \overline{x} = 5$

$ \overline{y} = 10$

$(x_1 - \overline{x}) \cdot (y_1 - \overline{y})= 8 $

$(x_2 - \overline{x}) \cdot (y_2 - \overline{y})= 0 $

$(x_3 - \overline{x}) \cdot (y_3 - \overline{y})= 6 $

$ \sum_{i=1}^{n} (x_i - \overline{x}) \cdot (y_i - \overline{y}) = 14$

$ \frac{\sum_{i=1}^{n} (x_i - \overline{x}) \cdot (y_i - \overline{y})}{n-1} = 7$

$$ corr(X, Y) = \frac{cov(X, Y)}{\sigma_x \cdot \sigma_y} $$

perfectly positively correlated data points

perfectly negatively correlated data points

examples of data points with some positive, some negative and almost no linear relationship

examples of data points with some positive, some negative and almost no linear relationship

examples of data points with some positive, some negative and almost no linear relationship

correlation coefficient

nonlinear relationships

domino

Practicing Statistics Interview Questions in R