What's the formula for calculating a residual? Find the residuals for the players with 149 and 86 hits, respectively
@jim_thompson5910
http://stattrek.com/regression/residual-analysis.aspx on that page they say \[\LARGE e = y-\hat{y}\] or in english Residual = Observed value - Predicted value
so for x = 149, you got the predicted value of 85.329 roughly but the actual observed value is 78. Look at the row that has 149 hits it corresponds to 78 runs so... \[\LARGE e = y-\hat{y}\] \[\LARGE e = 78-85.329\] \[\LARGE e = -7.329\]
the negative residual of e = -7.329 means that we made an overestimate
oh ok
do the same for x = 86
ok cool ty
residuals are the errors and the goal of any regression model is to minimize the errors (so minimize the residual values). The smaller the residuals, the better the fit You can use a residual plot to see how well the model fits or not
ok thnxs,
np
can u help with another ?
Join our real-time social learning platform and learn together with your friends!