So far, we have only analyzed population parameters for one random variable. However, in the real world, we have many random variables that can be related to each other. Therefore, it is useful to look at some additional population statistics that describe the relationship between two or more random variables. If the profitability of one stock increases, does the profitability of another stock increase, decrease or not affect? The true degree of association (linear) between the yields of two populations (across the population) is known as population covariance or correlation coefficient.

**Yields of two bitcoins**

If the true correlation between the yields of two **bitcoin** is positive in **gdax**, then if the return of X is positive, say, in 1000 observations, the return of Y will also be positive more than 500 times. The strength and frequency of this association is measured by covariance or correlation coefficient, but the latter has the advantage that its value is always between -1 and +1 (while covariance depends on the units used). to measure X and Y, for example, X can be a weight in kilograms or pounds, and Y can represent growth in inches or centimeters, and the selected units will determine the size of the covariance).

In the previous section, we calculated the average return (sample) for FTSE and S & P. The formula for the average R sample weighs each observation equally. If stock returns do have a constant average population value / JL, then the sample average R is an objective estimate of this true average value / x. However, sometimes we want to consider the expected results of various possible scenarios.