Statistics 4P03/6P03 Assignments 1998-99
All assignments must be
submitted no later than 5:00 PM on Wednesday April 28 1999, the final
day of the undergraduate examination period.
A01 1999-02-04
This is to be a first report on the
Niagara
River Pollution Data.
- Write two or three pages on the history of pollution in the
Niagara river.
- Carry out Exploratory Data Analysis, with an emphasis on
graphical methods. The work should primarily be done in Splus but
you can also try using a spreadsheet or another statistics
package.
- Write a summary - in words - of what you have learned from
EDA. Emphasize the important and meaningful features of the
graphs, but don't just describe in words what the graphs look
like.
- The EDA should leave you with some conjectures and unanswered
questions. Try to (a) formulate some questions for the data
owners, (b) give some suggestions for further studies, and (c)
propose some modeling ideas for Assignment 2.
A02 1999-03-10
This is to be a second report on the
Niagara
River Pollution Data.
- Repeat the Exploratory Data Analysis of Assignment 1, but this
time using SAS.
- Try some model fitting using SAS and/or Splus; trying both on
the same model is a useful exercise, it will help ensure that you
know how each package works.
- Write a report summarizing what you have learned from your
analysis to date. Has pollution from sources in the Niagara River
watershed decreased over the period of the study?
A03 1999-03-18
Learn a new statistical method. Here are four possibilities; you
can choose one of these or see me to choose something else.
- Use time series analysis to study the
Niagara
River Pollution Data.
- Some pollutants measured in the
Niagara
River Pollution Data were frequently at their upper or lower
detection limits. Modify the modeling techniques you used in
Assignment 2 to allow for detection limits.
- Discretize the pollutant measures in the
Niagara
River Pollution Data; this can avoid problems of detection
limits and eliminate the influence of outliers. Analyze the
categorized data as a multi-way contingency table using a
log-linear model.
- Find another data set suitable for log-linear analysis.
A04 1999-03-18
Today you heard a presentation by Mohit Bhandari and Michael
Boucher on the pilot study "Malalignment following intramedullary
nailing of tibial shaft fractures: Magnitude and clinical
implications," and we had a discussion about the study, its
objectives, its methods, and other ways it could have been done. You
will receive the data files electronically.
Write a consultant's report summarizing what has been learned from
the pilot study and making recommendations for the design of the main
study. Where possible, make specific recommendations about
stratification and sample sizes.
I will give Dr Bhandari copies of your reports in case he wants to
follow up with any of your recommendations.