Home  >  113 Math  >  113.032 Statistics - Linear Regression Least Squares method

Least Squares Method

Least Squares is the most common method to calculate a linear regression.

To calculate the equation of the least squares line - Start with the usual slope equation y = mx + b - Calculate the mean and standard dev of both x and y - Calculate the r coefficient (get a number between -1 and 1) - Replace the mean of x and y in the slope equation - Calculate m equals m = (StdDev(y) * r) / StdDev(x) - Calculate b equals b = Mean(y) - (m * Mean(x)). You now have the value of m so just swap that in. - Now you can complete the initial equation with the new values for m and b


Source:
    • https://en.wikipedia.org/wiki/Linear_regression
Graph: