You are in a statistics course minding your business. The teacher presents you with an equation, y=4+2*age+4*height. You immeditiately recognize the equation as a linear regression model. Yay I know something! Then you see statistical output like this,
Label Coefficient standard deviation T statistic Pvalue
Constant 4 2.32 3.33 0.0001
Age 2 1.21 4.44 0.50
height 4 1.03 3.24 0.02
And you are asked to indicate which variables are useful for predicting Y. Excuse me? Useful? You might think, it would be useful if I wasn’t required to take this class in the first place. Ha! Okay seriously, the p-value interpretation is the main focus. You are being required to compare the p-value column. If your P value is less than your level of confidence, it is useful for predicting Y. Statistical significance is another term people use and it means you got something here.
- Moore to follow -Amy