MTH5120 Exam 2017

Main Examination period 2017
MTH5120: Statistical Modelling I

Duration: 2 hours
Apart from this page, you are not permitted to read the contents of this
question paper until instructed to do so by an invigilator.
You should attempt ALL questions. Marks available are shown

next to the questions.
Only non-programmable calculators that have been approved from the

college list of non-programmable calculators are permitted in this
examination. Please state on your answer book the name and type of
machine used.
Statistical functions provided by the calculator may be used provided
that you state clearly where you have used them.
The New Cambridge Statistical Tables are provided.
Complete all rough work in the answer book and cross through any work that is
not to be assessed.
Possession of unauthorised material at any time when under examination
conditions is an assessment offence and can lead to expulsion from QMUL. Check
now to ensure you do not have any notes, mobile phones, smartwatches or
unauthorised electronic devices on your person. If you do, raise your hand and give
them to an invigilator immediately.
It is also an offence to have any writing of any kind on your person, including on
your body. If you are found to have hidden unauthorised material elsewhere,
including toilets and cloakrooms, it shall be treated as being found in your
possession. Unauthorised material found on your mobile phone or other electronic
device will be considered the same as being in possession of paper notes. A mobile
phone that causes a disruption in the exam is also an assessment offence.
Exam papers must not be removed from the examination room.
Examiners: L I Pettit, A Gnedin

c Queen Mary, University of London (2017) Turn Over
Page 2 MTH5120 (2017)
Question 1. [23 marks]

The following table gives the age in years (x) and total cholesterol level in mg/ml
(y) for 19 patients suffering from hyperlipoproteinaemia.
x y x y x y x y
46 3.5 20 1.9 52 4.0 30 2.7
57 4.5 25 3.0 28 2.9 36 3.8
22 2.1 43 3.8 57 4.1 33 3.0
22 2.5 63 4.6 40 3.2 48 4.2
28 2.3 49 4.0 52 4.3
P P
Summary statistics for these data are xi = 751, yi = 64.4, Sxx = 3306.74,
Sxy = 192.705, Syy = 12.698.
(a) The data are expected to be linearly related and the simple linear regression
model
yi = α + β(xi − x̄) + εi i = 1, 2, . . . , n
is to be fitted. What assumptions are usually made about the errors (εi )? [4]
(b) Derive the least squares estimators of α and β by minimising a suitable

function. Check that your solution does give a minimum. [10]
(c) Hence find the equation of the fitted model for the cholesterol data. [4]
(d) State the form of the 95% confidence interval for β. Find the numerical
estimate of this interval. [5]

To investigate the effect of dose of a drug on a response, twenty patients were
allocated at random to one of four doses (1,5,9,13mg) so that five patients received
each dose. A simple linear regression model of response was fitted.
(a) Copy and complete the following Analysis of Variance table. [11]
(b) What two hypotheses can be tested? [2]
(c) Carry out these tests using a 1% significance level and make clear your
conclusions. [6]
Analysis of Variance
Source DF SS MS VR
Regression 1 1387.6
Residual Error
Lack of Fit 33.2
Pure Error
Total 19 1454.8

c Queen Mary, University of London (2017)
MTH5120 (2017) Page 3

Data were collected from 17 US Navy hospitals. The variables measured were x1,
average daily patient load, x2, X-rays taken per month, x3, occupied bed days per
month, x4, eligible population in thousands, x5, average length of stay in days, and
Y the staff hours per month.
(a) The data were entered into Minitab and the best subsets regression procedure
carried out. The output is shown below.
Best Subsets Regression: y versus x1, x2, x3, x4, x5
Response is y
Mallows x x x x x
Vars R-Sq R-Sq(adj) Cp S 1 2 3 4 5
1 97.2 97.0 20.4 957.86 X
1 97.1 97.0 21.2 969.53 X
2 98.7 98.5 4.9 685.17 X X
2 98.6 98.4 5.7 700.42 X X
3 99.0 98.8 2.9 614.78 X X X
3 98.9 98.7 3.7 634.99 X X X
4 99.1 98.8 4.0 615.49 X X X X
4 99.1 98.7 4.3 622.09 X X X X
5 99.1 98.7 6.0 642.09 X X X X X
(i) Define the four statistics given in the table: R2 , R2 (adj), Mallows Cp
and S. [4]
(ii) Based on these statistics say which model you would choose for these
data and justify your choice. [8]
(iii) In what way is R2 (adj) an improvement on R2 ? [3]
(b) The Minitab session output for the model with regressors x2, x3 and x5 is
given below.
The regression equation is

y = 1523 + 0.0530 x2 + 0.978 x3 - 321 x5
Predictor Coef SE Coef T P VIF

Constant 1523.4 786.9 1.94 0.075
x2 0.05299 0.02009 2.64 0.021 7.737
x3 0.9785 0.1052 9.31 0.000 11.269
x5 -321.0 153.2 -2.10 0.056 2.493
S = 614.779 R-Sq = 99.0% R-Sq(adj) = 98.8%

c Queen Mary, University of London (2017) Turn Over
Page 4 MTH5120 (2017)
(i) What is meant by multicollinearity? What problems can it cause. [6]

(ii) Define the variance inflation factor (VIF). [3]
(iii) Why do we calculate variance inflation factors? [2]
(iv) Comment on the sizes of the variance inflation factors in this example. [3]
(c) The following normal plot and test of the standardised residuals was
produced.
(i) Comment on what this tells us about the assumption of normally

distributed errors. [3]
(ii) Name one other plot you would like to see to assess if the model is
fitting well. Explain what it would tell you. [3]

MTH5120 (2017) Page 5
(a) For the general linear model Y = Xβ + ε where ε is a vector of errors

assumed to be uncorrelated with zero mean and constant variance σ 2 , state
the formula for the least squares estimator β̂. [1]
(b) Prove that the expectation of β̂ is β. [4]
(c) Derive a formula for the variance-covariance matrix of β̂, quoting any
necessary results. [6]
(d) (i) Define the hat matrix H. [1]

(ii) Show that the vector of fitted values is given by HY . [2]
(e) Show that HH = H. [3]
(f) Express the model
Yi = β0 + β1 xi + β2 x2i + εi i = 1, 2, . . . , n
where the εi have mean zero, variance σ 2 and are uncorrelated, as a general
linear model by writing down the vectors Y and β and the matrix X. [6]
End of Paper.


MTH5120 Exam 2017

Uploaded by

Copyright:

Available Formats

MTH5120 Exam 2017

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

MTH5120 Exam 2017

Uploaded by

Copyright:

Available Formats

Main Examination period 2017

MTH5120: Statistical Modelling I

You should attempt ALL questions. Marks available are shown

Only non-programmable calculators that have been approved from the

Exam papers must not be removed from the examination room.

Examiners: L I Pettit, A Gnedin

Question 1. [23 marks]

(b) Derive the least squares estimators of α and β by minimising a suitable

Question 2. [19 marks]

(b) What two hypotheses can be tested? [2]

Question 3. [35 marks]

Best Subsets Regression: y versus x1, x2, x3, x4, x5

The regression equation is

Predictor Coef SE Coef T P VIF

S = 614.779 R-Sq = 99.0% R-Sq(adj) = 98.8%

(i) What is meant by multicollinearity? What problems can it cause. [6]

(i) Comment on what this tells us about the assumption of normally

Question 4. [23 marks]

(a) For the general linear model Y = Xβ + ε where ε is a vector of errors

(b) Prove that the expectation of β̂ is β. [4]

(d) (i) Define the hat matrix H. [1]

(e) Show that HH = H. [3]

(f) Express the model

You might also like