STAT 301 Lecture Notes - Mean Squared Error, Bias Of An Estimator, Standard Deviation
Document Summary
Our model is yi = 0 + 1 x1 + 1 where 1 iid. We need to estimate 0, 1, and 2. = b0 + b1 xi b0 is estimate for 0 b1 is estimate for 1. 2 is the estimate for 2, where se. So mse is the estimated standard deviation of the residuals, called the residual standard error. Note: we use n 2 degrees of freedom so that mse is an unbiased estimator of 2, as we estimate 2 parameters ( 0, 1) beforehand. In regression, we have a certain amount of variability in the y values. We want to understand what portion of this error can be accounted for by our model. Ssm is the variation explained by our regression model. Sse is the variation left unexplained by our model. We want to account for as much variation as possible with our model.
