For Exercises 13.46 to 13.49, see pages 693694; for 13.50, see page 699; and for 13.51 and 13.52, see pages 705706.

Question 13.48

13.48 Comparing models for hires rate with MSE.

In Exercise 13.47, the mean absolute deviation was introduced as a measure of forecasting accuracy. Another measure is mean squared error (MSE), which is a measure of the average size of the prediction errors in squared units:

where is the residual for period and is the number of available residuals. (Note: Some software refers to this measure as mean squared deviation, or MSD.)

  1. Compute MSE for the residuals from the 1-month, 2-month, 3-month, and 4-month moving-average models determined in Exercise 13.46, part (c).
  2. What moving-average span value corresponds with the smallest MSE?

mean squared error (MSE)

hires