# Chapter 13

# Generalized Linear Models

**13.1** **a.**
**b.** Deviance = 17.59 with *p* = 0.483 indicating that the model is adequate.

**c.** Ô_{R} = *e*^{−.0177} = .9825 indicating that for every additional knot in speed the odds of hitting the target decrease by 1.75%.

**d.** The difference in the deviances is basically zero indicating that there is no need for the quadratic term.

**13.3** **a.**
**b.** Deviance = .372 with *p* = 1.000 indicating that the model is adequate.

**c.** The difference in the deviances is Dev(*x*) − Dev(*x*, *x*^{2}) = .372 − .284 = .088 indicating that there is no need for the quadratic term.

**d.** For *H*_{0} : β_{1} = 0, the Wald statistic is *Z* = −.42 which is not significant. For *H*_{0} : β_{2} = 0, the Wald statistic is *Z* = −.30 which is not significant.

**e.** An approximate 95% confidence interval for β_{1} is (−.0018, .0033) and an approximate 95% confidence interval for β_{2} is (7.15 × 10^{−7}, 5.27 × 10^{−7}).

**13.5** **a.**
**b.** Deviance = 14.76 indicating that the model is adequate.

**c.** For

_{1}, we get

*Ô*_{R} ≈ 1 indicating that the odds are basically even. For

_{1}, we get

*Ô*_{R} = 3.52 indicating that every one year increase in the age of the current car ...