1

Copying and plagiarism
Any external sources utilised should be correctly referenced using a common referencing
technique (e.g., the Harvard technique). For more details on referencing please visit
http://www.westminster.ac.uk/study/current-students/support-and-facilities/library-itservices/referencing.
Copying and plagiarism carry severe penalties. Please note that the University offers an online
learning tutorial designed to help students understand and avoid plagiarism. This can be
2
accessed by any student under My Organisation on Blackboard. The tab is labelled ‘Avoiding
Plagiarism’.
Penalty for Late Submission
If you submit your coursework late but within 24 hours or one working day of the specified
deadline, 10 marks will be deducted from the final mark, to minimum of the pass mark (50%),
as a penalty for late submission. If you submit your coursework more than 24 hours after the
specified deadline you will be given a mark of zero for the work in question unless a claim of
Mitigating Circumstances has been submitted and accepted as valid.
It is recognised that on occasion, illness or a personal crisis can mean that you fail to submit
a piece of work on time. In such cases you must inform the Faculty Registry Office in writing
on a mitigating circumstances form, giving the reason for your late or non-submission. You
must provide relevant documentary evidence with the form. This information will be reported
to the relevant Assessment Board that will decide whether the mark of zero shall stand. For
more detailed information regarding University Assessment Regulations, please refer to the
following website: http://www.westminster.ac.uk/study/currentstudents/resources/academic-regulations
Submission Details
Please strive to answer all question. The coursework must be returned by the above due date
for full credit. All datasets, together with their descriptions, can be found in the ‘Assessment’
section of the BlackBoard module site. Poor presentation will be penalised. Remember that
this is meant to be an individual coursework and that any indication that it is otherwise may
result in severe penalties.
Electronic submission of problems 1, 2, 3 and 4 is required via BB by 15/05/23, 1 PM. The
main report including appendices should be only in ONE file (MS word or pdf: named
STUDENTID_7BUIS024W_CW_Main, where SURNAME must be replaced by your own
surname). This document should be self-contained– this is a separate single ZIP file (named
STUDENTID_7BUIS024W_CW_Models) to be submitted at the same time in a separate
submission link. To obtain brief initial (formative) feedback, please submit a solution plan of
your work via BB by 14/03/2023, 6PM (Optional).
Your report should contain a short description of the model solutions, answers to specific
questions and your interpretations of the results. The main text of the report should not
exceed 3,000 words with A4 pages (11 point, 1.5 spacing). Also, it may well be useful to
include appendices with tables, graphs, etc. These must be explained in the main text. The
report should be clearly structured, communicate the outcomes of your analysis in an
effective way and avoid too much technical jargon. Marks for each problem are displayed in
parenthesis.
3
Coursework Description
Problem 1 [35 Marks]
Aware of your expertise in data analysis, the London Assembly Housing Committee (LAHC)
has contacted you to undertake the analysis of data related to house price in London and
including deprivation and demographic variables [file London House Prices and Deprivation
per LSOA.xlsx]. The data contain variables such as median house price, population size for
different age groups, surface area, and various deprivation score rank recorded per Lower
Super Output Area (LSOA) in London (4,422 records). You are also provided with another
spreadsheet (IoD2019_Domains_of_Deprivation.xlsx) containing the 2019 Index of Multiple
Deprivation score, its components, and their definition for all Lower Super Output Area (LSOA)
in England (including London). They would like you to answer the questions below. For each
question, indicate what assumptions you have made and justify why you choose a particular
approach (graph, hypothesis test, both, etc.). These can be described either in your Excel
workbook (using notes) or in your report(provided you don’t exceed word count. No more
than 2-3 lines.
a. Median house prices [12]
i. Use suitable summary statistics and chart to analyse the distribution of

1. Camden and neighbouring Westminster
2. Westminster and Kensington and Chelsea
3. Lewisham and Camden
b. Age groups [2]
4
i. Calculate the percentage of the population in each age group per local
authority district.
ii. Use suitable charts to compare the split between age groups per local
authority district.
c. Deprivation variables (note that the values are ranks, 1 being the lowest, e.g.,
most deprived – these are across all of England, not just London) [2]
i. Use suitable charts to explore the distribution of each variable,
excluding the house price.
ii. Use suitable charts to compare each deprivation variable per local
authority district.
d. Relationships between variables. [2]
i. Calculate all correlation coefficients where appropriate.
ii. Explore the relationship between each of the variable and the house
price.
e. Model(s) to forecast house price in terms of the other variables. [7]

f. In the report, describe your findings. [10]
Problem 2 [17 Marks]
London Coaches (LC) is considering running a coach service between London and Brussels.
They may operate the service independently, doing their own sales and marketing. They will
also have to compete against the existing service on this route. Alternatively, they could form
a partnership with the incumbent operator, leaving the sales and marketing to them, and
providing coaches and drivers on demand. The profit per month will depend on the number
of customers on the route as shown in the table below. The probabilities for customer
demand levels, as judged by LC, are also given in the table.
5

Of course, if they choose not to run the service, they will have neither profits nor losses.
a. Using a suitable package such as Precision Tree, represent this problem. Clearly
indicate what the final recommendation and expected payoff are. [3]
b. In order to make a more informed decision, LC could hire a transport consultant from
the University of Westminster to give them a forecast on the number of passengers.
The forecast costs £6,000 and yield two possible results: a “fairly high” or a “fairly low”
passenger number prediction. LC believes that these predictions are equally likely. The
consultant is good at his job and would not predict a fairly low number of passengers
if there will be a high number of passengers or a fairly high number of passengers if
there will be a low number of passengers. If he predicts a fairly low number of
passengers, then there is a 60% chance that there will be a medium number of
passengers and if he predicts a fairly high number of passengers then there is a 40%
chance that there will be a medium number of passengers. Find the company’s
optimal course of action using a decision tree. [5]
c. Perform sensitivity analysis on the probability that the consultant’s prediction about
the number of passengers is “fairly low” or “fairly high” to analyse the robustness of
your recommendation to possible changes of this parameter. [4]
d. In the report, describe your findings. [5]
Problem 3 [30 Marks]
The Lewisham City Council is faced with a severe budget shortage due to unexpected social
costs from the Covid-19 pandemic. The council is now reconsidering an earlier proposal
submitted by a property developer. This proposal seeks to demolish existing council buildings
in the Catford area and replace them with modern residential developments that can attract
6
higher council tax rates. The developer can build one-bedroom, two-bedroom, threebedroom or four-bedroom houses. The council must, therefore, determine how many existing
buildings to demolish and how many units of the new buildings to put up. The following
information is available about the project:
– The Catford area has about 250 units of council buildings that can be demolished. The
cost of demolishing a building is budgeted to be £2500. Each building occupies a 0.30-
acre lot.
– The table below shows the construction cost and the budgeted lot size required for
each type of new house. It also shows the expected annual council tax that could be
generated from each type of property.

– Construction policies require that 10% of the available acreage after demolishing be
reserved for streets, utility instalments and other emergency facilities. Additionally, at
least 10% of the new units should be four-bedroom houses. At least 5% of the new
units should be two-bedroom houses. Finally, at most 45% of the new units should be
two- and three-bedroom houses.
– The council can access a maximum of £12 million financing facility with a High Street
bank.
The council will want to determine the maximum annual council tax that could be generated
after this project. Answer the following questions:
a. Formulate a linear programming model to determine the total number of each type
of new housing unit to be built and the number of old buildings to be demolished to
maximise council tax generated. [7]
b. Formulate and solve the LP model using Excel solver. How much tax will the current
plan generate? [8]
7
c. Using the sensitivity report, answer the following questions: [8]

tax for a four-bedroom unit is set equal to that for a three-bedroom
house?

Problem 4 [18 Marks]
A company sells air conditioners to four regions labelled as East, South, North, and West.
The company has gathered historical data of the monthly demand for air conditioners over
the past 9 years (since 2013)- see file Data_Pb4.xlsx.
a. Explore the time series for each region. [6]
i. Plot the data for each region as time series.
ii. Determine whether these exhibit trend and/or seasonality?
iii. If/Where relevant, calculate seasonal indexes for each region.
iv. State with justification for each region what would be appropriate
forecasting methods to forecast monthly demand for the remainder of
2023.
b. Apply two appropriate forecasting methods for each region to forecast
monthly demand remainder of 2023. [7]
i. For each region, compare performance of the two methods using MAE,
RMSE, and MAPE.
ii. Check and comment on forecasting errors for randomness and
autocorrelation.
iii. Using the best method, produce monthly forecast for each region
remainder of 2023.
c. In the report, describe your findings. [5]