MT2300 Examination project for 2019-20
Your project should be completed using R and submitted to the School (EPMS) Office,
Bedford Buillding 1-29, by 4.00pm on 26th March, 2020. Only one copy is required for
submission, but you should also keep a copy for yourself. Project reports must be word
processed, but hand-written annotations, paricularly on computer output (including formulas
on graphs), are perfectly acceptable.
Your project report should be marked clearly at the top with your RHUL candidate number,
and should not include your name or student number anywhere. Your project must be
your own work. Investigations of academic offences divert significant resources, which
could be better spent on, for example, providing more detailed feedback. If still in doubt
what constitutes an academic offence, plagiarism in particular, please consult the Regulations
on Academic Misconduct. If approached by a classmate requiring help with their project,
please encourage them to see the lecturer or other academic staff instead of discussing your
own work with them as this puts both of you at risk of committing an academic offence,
however unintentionally.
If you do not hand your project in by the deadline given, but do hand it in by 4.00pm on 27th
March, 2020, your mark (out of 100) on the project will be reduced by 10 marks, which is
10% of the total marks for the project. If you hand it in later without a pre-approved
extension or a documented extenuating circumstance, the submission will be recorded but
you will score 0 for the project (in accordance with the Undergraduate Regulations).
You should attempt to write as if you are an analyst making a report to a manager in a
relevant organisation, who has some idea about basic statistics. Make clear what you have
done and why. Do not go into technicalities such as mathematical proofs, but do give the
names of the techniques used with brief explanations. Graphs, tables, and diagrams often
help. Make sure those are sufficiently annotated. If you need to make some assumptions, you
should state them. At least half of the marks are given on the clarity and readability of the
project. Your report should be no more than ten (one-sided) pages in total (including the
cover sheet), longer projects will be penalized. A good project would normally be 5-8 pages
long. Choose an easily readable typeface (e.g. Times New Roman, Serif) in which the
regular style contrasts clearly with the italic, and set it to a standard size (10-12 points).
The data set in the file Data for the Examination Project (on Moodle) was
collected as part of an investigation by a hospital surgical unit interested
in predicting survival in patients undergoing a particular type of liver
operation. A random sample of 54 patients was available for analysis. The
following information was exctracted at the preoperational evaluation:
blood clotting score, prognostic index, enzyme function test score, liver
function test score, age (in years), sex (0=male, 1=female), indicators of
moderate and severe alcohol use (0=no, 1=yes). A follow-up study
provided survival times for the same patients.

Analyse the data. Firstly, investigate all the predictors individually,
discussing their suitability for predicting survival. Secondly, determine
the best regression model, or models, for predicting survival, interpreting
your findings.