In: Statistics and Probability
What impact does regression have on correlation analysis? Explain the two ways correlation can be displayed mathematically. Does one way have an advantage over the other? Please explain your thinking.
Regression analysis has an advantage over correlation analysis that it provides a predictor equation which can be used to predit values of 'y' at given values of 'x'.
On the other hand, correlation analysis is superior to regression in that it does not depend on the choice of a dependent and independent variable. Which means that if the variables are interchanged, the correlation will also be inversed, but the regression analysis may not be reversed to get the new equation.
Correlation can be represented using a graphical method by making a scatter plot of the data. If the points show a linear trend, then there is correlation, if the points form a blob, then there is no correlation.
Correlation can also be represented by computing the correlation coefficient, r.
If r = 0, then there is no correlation.
If r > 0, then there is positive correlation.
If r < 0, then there is negative correlation.
The greater the value of r is towards the extreme value, the stronger is the correlation.
Hope this helps !