Module Applied Statistics
ASSIGNMENTTITLE Course Work 2
DATEANDTIMETOHANDIN Wednesday 7th March 2018
ASSIGNMENTVALUE 25 %
SETBY Dr Nikolaos Korfiatis
Aims
To help assimilate material presented in the lectures by providing appropriate exercises.
Learningoutcomes:
To improve understanding of the material by working individually on problems based on the
material.
Assessmentcriteria:
1. Accuracy of answers.
2. Understanding of subject displayed.
3. Clarity of explanations of working.
4. Quality of reporting.
As always credit is given for a persuasive argument.
Descriptionofassignment:
Solution of problems related to material in lectures.
1. Present solutions to all problems.
2. Giving full working that demonstrates the steps required to obtain the correct solution
and the important principles used.
3. The logic of the solution is important
1
Handinginprocedure:
Please submit your piece of coursework in the following way:
1. Attach this coursework assignment sheet to your piece of coursework. Thisshouldbea
word document reporting your results as per the lecture notes. Do not exceed the 10
pageswithminimumfontsize11pt.
2. In addition to your report submit a printed R file with all the regression commands.
Plagiarism
Plagiarism is the copying or close paraphrasing of published or unpublished work, including
the work of another student; without due acknowledgement. Plagiarism is regarded a serious
offence by the University, and all cases will be investigated. Possible consequences of plagia-
rism include deduction of marks and disciplinary action, as detailed by UEA’s Policy on Plagia-
rism and Collusion.
If you have any questions drop by my office (Thomas Paine Study Centre 0.15) or contact me at
2
AppliedStatistics Coursework2 2017/18
To explore which patient demographic characteristics and organisational factors influence the
length of stay (LOS) among medical admissions, we record data for several hospitals. The data
obtained are as follows (SPSS data set length.sav must be read in R):
Variable description
id id
age Patients’ average age
risk Risk of infection in the hospital
rcr Ratio of the number of infected to the number of uninfected (£100)
xray Ratio of the number of xrays to the number of patients (£100)
beds Average number of beds
aff University Affiliation (0: No, 1: Yes)
avdaily Average number of admissions per day
nurses Average number of nurses
fac Percentage of surgeries
slength LOS
1. Determine the best regression model with the backward elimination and with the for-
ward selection. Explain what happens in every step. Compare the models and select the
best regression model. [marks 20]
2. Check the adequacy of this model. [marks 10]
3. If the assumptions are violated transform. the dependent variable to correct the model
and determine the best regression model. [marks 20]
4. Report any outlier or influential value in this model. [marks 10]
5. In this model decide whether the addition of the interaction term between age and aff
contributes substantially to the prediction of the length of stay. Useallthetaughttests
andmethods. [marks 10]
6. Adding the above interaction term which are the predicted regression models for
(a) University affiliation
(b) Non-University affiliation. [marks 10]
7. Interpret the regression coefficients in the best regression model with the interaction
term. [marks 15]
8. Interpret the multiple R2 in the best regression model with the interaction term. [marks 5]