PRESS (prediction sum of squares) tells you how well the model will predict new data.
Details
The prediction sum of squares (PRESS) is the sum of squares of the prediction error. Each fitted to obtain the predicted value for the ith observation. Use PRESS to assess your model's predictive ability. Usually, the smaller the PRESS value, the better the model's predictive ability.
For models fitted with prior weights, PRESS is the weighted sum of squared deleted residuals, \(\sum w_i (e_i / (1 - h_i))^2\). It therefore scales with the weights: multiplying every weight by a constant multiplies PRESS by the same constant. Unweighted models are the special case where every weight equals one.
References
Kutner, MH, Nachtscheim CJ, Neter J and Li W., 2004, Applied Linear Statistical Models (5th edition). Chicago, IL., McGraw Hill/Irwin.
See also
Other influence measures:
ols_hadi(),
ols_leverage(),
ols_pred_rsq()
Examples
model <- lm(mpg ~ disp + hp + wt + qsec, data = mtcars)
ols_press(model)
#> [1] 257.8314