Question: A computer repair service centre is examining the time taken on service calls to repair computers. Data for 30 service calls are obtained (asm2Q2.xls). The information obtained includes:

1) x₁, the number of machines to be repaired (NUMBER)

2) x₂, the number of years of experience of the service person (EXPER).

3) y, the time (in minutes) taken to provide service (TIME).

The model
\[y={{\beta }_{0}}+{{\beta }_{1}}{{x}_{1}}+{{\beta }_{2}}{{x}_{2}}+{{\beta }_{3}}x_{1}^{2}+e\]

has been suggested.

a) Use a graphical means to explain why this model may represent an improvement over the linear model . Confirm your conclusion with a statistical test at 5% level.

b) Identify the observations that have x₁ and x₂ not consistent in pattern with the remaining observations?

c) Based on the suggested model, which observations are unusual in time taken to provide services after taking into account the number of machines serviced and the experience of the service person? Use 2% level in the t table.

d) Which observations have ­­­large influences on the regression coefficients of the suggested model?