|
In statistics, correlation is an assessment of the extent of linear relationship between two variables. A correlation coefficient is the index of the extent of the linear relationship. The most common correlation coefficient is the Pearson product-moment correlation, defined as the covariance of the two variables divided by the product of their standard deviations. The Pearson correlation coefficient is always between –1 and 1. The weaker the relationship, the closer the correlation coefficient is to 0. A positive correlation coefficient indicates a higher value for one variable tends to correspond to a higher value for the other, and a negative correlation coefficient indicates a higher value for one variable tends to correspond to a lower value for the other. Correlation is not an appropriate statistical technique for categorical data such as gender or color. In contrast, correlation is an extremely useful technique for prices. Correlation of stock prices is a simple yet informative analysis. If two stocks have a high correlation, their prices tend to rise and fall in sync. |
