You are on page 1of 2

STA3002: Generalized Linear Models Spring 2023

Homework #3
This is the homework for week 4. The due date is midnight, February 28th , 2023. Late submission
will automatically result in “0” grade.

1 Computation Questions
(Question1): (25 points) Consider the single response variable Y with Y ∼ Bin(n, π).
(a) Find the Wald statistic (π̂ − π)T ℑ(π̂ − π), where π̂ is the maximum likelihood estimator
of π and ℑ is the information.
(b) Verify that the Wald statistic is the same as the score statistic U T ℑ−1 U in this case.
(c) Find the deviance
2[l(π̂; y) − l(π; y)]
.
(d) For large samples, both the Wald/score statistic and the deviance approximately have
the χ2 (1) distribution. For n = 10 and y = 3, use both statistics to assess the adequacy
of the models:
i) π = 0.1; ii) π = 0.3; iii) π = 0.5.
Do the two statistics lead to the same conclusions?
(Question2): (25 points) Suppose Y1 , . . . , YN are independent identically distributed random vari-
ables with the Pareto distribution with parameter θ.
(a) Find the maximum likelihood estimator θ̂of θ.
(b) Find the Wald statistic for making inferences aboutθ.
(c) Use the Wald statistic to obtain an expression for an approximate 95% confidence
interval θ .

2 Programming Questions
(50 points) Children were asked to build towers as high as they could out of cubical and cylindrical
blocks [3, 7]. The number of blocks used and the time taken were recorded (data set: “blocks"). In
this problem, only consider the number of blocks used y and the age of the child x. Fit a GLM for
these data.
1. Fit this GLM using r, and write down the fitted model.
2. Use a Wald test to determine if age seems necessary in the model.
3. Use a score test to determine if age seems necessary in the model.
4. Use a likelihood ratio test to determine if age seems necessary in the model.
5. Compare the results from the Wald, score and likelihood ratio tests. Comment.
6. Is the saddlepoint approximation expected to be accurate? Explain.

1
7. Find the 95% Wald confidence intervals for the regression coefficients.
8. Plot the number of blocks used against age, and show the relationship described by the fitted
model. Also plot the lines indicating the lower and upper 95% confidence intervals for these
fitted values.

You might also like