Salam Academy
WST03 · Module

Regression and correlation

2 subtopics≈ 65 min in total
  1. Summary statistics and the PMCC

    The summary statistics SxxS_{xx}, SyyS_{yy} and SxyS_{xy} measure spread and how xx and yy vary together. The product moment correlation coefficient r=SxySxxSyyr=\dfrac{S_{xy}}{\sqrt{S_{xx}S_{yy}}} lies between −1-1 and 11 and measures how close the points are to a straight line.

    ≈ 30 min
  2. The least squares regression line

    The regression line of yy on xx, y=a+bxy=a+bx, is the straight line that makes the squared vertical distances from the points as small as possible. Its gradient is b=SxySxxb=\dfrac{S_{xy}}{S_{xx}} and it always passes through (xˉ,yˉ)(\bar x,\bar y), so a=yˉ−bxˉa=\bar y-b\bar x.

    ≈ 35 min