STU22005: Applied Probability II 2022
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
STU22005: Applied Probability II
Trinity Term 2022
SECTION A
Instructions to students: Section A is completed on Blackboard. Go to the ’FINAL EXAM’ tab on Blackboard and click on ’Final Exam Section A’ to see the questions and submit your answers.
SECTION B
Instructions to students: Section B is submitted on Blackboard under ’Final Exam Section B’ . Write out your answers, scan to a single pdf file, and upload.
1. In a study of 2246 randomly selected drivers, 964 said that they use their mobile phone regularly while driving. It was believed that more than 40% of drivers regularly use their mobile phone while driving. Carry out a hypothesis test to test this claim.
(a) State the null and alternative hypotheses of interest.
(b) Compute the test statistic.
[4 marks]
[4 marks]
(c) Find the p-value, use it to evaluate the test statistic, and give the conclusion. Use a = 0.05. [5 marks]
(d) Briefly (in 2-3 sentences) explain the role of the Central Limit Theorem in this hypothesis test. [7 marks]
(20 marks)
2. A pot experiment was conducted to examine the relationship between the potassium found in plants (y) and two soil characteristics, soil potassium (z1 ) and soil acidity (z2 ; coded 1 for high and 0 for low). Values were recorded for 14 pots where soil potassium and soil acidity were manipulated at establishment.
The y and zi values are for high acidity:
i 1 2 3 4 5 6 7
yi
x1i
and for low acidity:
i
yi
x1i
Page 2 of 4
Oc Trinity College Dublin, The University of Dublin 2022
A multiple regression model was fitted with soil potassium and soil acidity as predictors, with the following output from R:
Call:
lm(formula = y ~ x1 + x2)
Residuals:
Min 1Q Median 3Q Max
-4 .5179 -3 .0321 -0 .1821 2 .4688 5 .0357
Coefficients:
Estimate Std . Error t value Pr(>|t |)
(Intercept) 27 .0571 4 .0274 6 .718 3 .29e-05 *** x1 1 .0982 0 .2373 4 .628 0 .000731 *** x2 7 .2286 1 .8985 3 .807 0 .002906 **
---
Signif . codes: 0 *** 0 .001 ** 0 .01 * 0 .05 . 0 .1 1
Residual standard error: 3 .552 on 11 degrees of freedom Multiple R-squared: 0 .7655,Adjusted R-squared: 0 .7229 F-statistic: 17 .96 on 2 and 11 DF, p-value: 0 .0003433
(a) Express the multiple regression model in matrix notation, using numeric
values in the y and x matrices. [6 marks]
(b) Interpret the estimate for the coefficient of z1 and the associated
hypothesis test. [6 marks]
(c) Show how the standard error for the parameter estimate associated with z2
is calculated.
Hint:
_ 1.28571429 (xT x)− 1 = '(') -0.07142857
'-0.14285714
-0.071428571 0.004464286
0.000000000
[4 marks]
-0.1428571_
0.0000000 '(')
0.2857143
(d) Give a brief (2-3 sentences) explanation of the F statistic in the final line of the R output, written for a non-statistician to understand. [4 marks]
3. In an experiment, independent Bernoulli trials with success probability p were conducted until a success was observed. In five independent repetitions of the experiment, the number of trials until the first success was recorded with the following results: 3, 2, 6, 4, 3.
(a) Show that the log likelihood function can be written as
l(p) = 13log((1 - p)) + 5log(p)
Use this expression to derive the maximum likelihood estimator of the
parameter p, the probability of success in an individual trial. [13 marks]
(b) Construct an appropriate graph to illustrate how the maximum likelihood
estimator is found from the likelihood function. Include labels on both of the axes. [7 marks]
(20 marks)
2022-08-26