b Explain why we can interpret fit1 as a model for the expected rate of

# B explain why we can interpret fit1 as a model for

• Test Prep
• 11
• 100% (21) 21 out of 21 people found this document helpful

This preview shows page 6 - 8 out of 11 pages.

(b) Explain why we can interpretfit1as a model for the expected rate of defective products per time unit, con-ditional on the temperature and pressure settings. 6
(c) If you are fitting a GLM withYiindepPoisson(λi), do you think you could use the inverse of some CumulativeDistribution Function for the link function? Explain carefully why or why not. 5. [30 points](a) Suppose that{x1, . . . , xn}are regularly spaced points on the interval [0,1], in ascending order(x1< x2< . . . < xn), and thatYi=f(xi) +iwithiiidN(0, σ2)withσ= 0.01fori= 1,2, . . . , nwheren= 256.Suppose that you don’t know the form of functionf(x), and you are asked to estimatefbased on observations{y1, . . . , yn}, corresponding to the{x1, . . . , xn}values above. When you plot theyvalues versus thexvalues,you obtain the following: ●● ●● ●● ●● ●● ●● ●● ●● ●●● ●● ●● ●● ●●● ●● ●● ●● ●● ●● ●● ●● ●● ●●● ●● ●● ●● ●● ●● ●●● ●● 0.0 0.2 0.4 0.6 0.8 1.0 -0.6 0.0 0.4 x y