The general manager of a large engineering firm wants to know whether the experience of technical artists influences their work quality. A random sample of 50 artists is selected. Using years of work experience (EXPER) and a performance rating (RATING, on a 100-point scale), two models are estimated by least squares. The estimates and standard errors are as follows:

a. Sketch the fitted values from Model 1 for \(E X P E R=0\) to 30 years.

b. Sketch the fitted values from Model 2 against \(E X P E R=1\) to 30 years. Explain why the four artists with no experience are not used in the estimation of Model 2.

c. Using Model 1, compute the marginal effect on RATING of another year of experience for (i) an artist with 10 years of experience and (ii) an artist with 20 years of experience.

d. Using Model 2, compute the marginal effect on RATING of another year of experience for (i) an artist with 10 years of experience and (ii) an artist with 20 years of experience.

e. Which of the two models fits the data better? Estimation of Model 1 using just the technical artists with some experience yields \(R^{2}=0.4858\).

f. Do you find Model 1 or Model 2 more reasonable, or plausible, based on economic reasoning? Explain.

Transcribed Image Text:

Model 1: RATING = 64.289 +0.990EXPER N=50 R = 0.3793 (2.422) (0.183) (se) Model 2: RATING (se) = 39.464+ 15.312 In(EXPER) N=46 R2 = 0.6414 (4.198) (1.727)