Project Ideas for 3rd and 4th Year Students:

 

1: Statistics with R

R is a language and environment for statistical computing and graphics.  It provides a wide variety of statistical and graphical techniques. R is available as Free Software under the terms of the Free Software Foundation's GNU General Public License in source code form. It compiles and runs on a wide variety of UNIX platforms and Windows.

The aim of this project is to develop a package using the graphical techniques in R to illustrate statistical techniques.

 

2: Solutions Interactive:

 

Probability with R: An Introduction with Computer Applications  by Jane M. Horgan (2008)  It is based on the CA219 course in Probability.

 

It is intended to develop interactive solutions to the end of chapter exercises as an addendum to the book, along the lines of Tutorials Interactive currently on the CA219 web page.  

 

More than one set of students may work on this project.   You will need to learn R, read some chapters in the book, and choose the set of exercise that you want to activate.  

3. Statistics Interactive:

Develop a teaching tool for parts of the 2nd year Probability Course

The project will concentrate on

It will be an add on to Statistics Interactive already available on the web.

More than one set of students may choose this project. The only requirement is that you are comfortable with statistics

4:  Stratification in Survey Sampling.

 

Stratification is a procedure breaks a population into mutually exclusive group from which sampling is done independently from each group.  It is an important technique is survey design.

 

There is an algorithm called Lavallee Hidiroglou procedure for stratification-skewed populations.   The algorithm is an iterative procedure which starts with initial solutions, and attempts to obtain optimum boundaries to minimise the variance of the stratified mean.  The algorithm is written in SAS.  I want it translated to R.  

 

Details of the algorithm is given in Gunning Horgan and Keogh (2007) Implementation Strategy for Efficient Convergence of the Lavallée and Hidiroglou Stratification Algorithm, Journal of Official Statistics (available from J. Horgan)

 

The SAS version is available at http://www.mat.ulaval.ca/fileadmin/SCS/Programmes/strates_neymann2.sas  and http://www.computing.dcu.ie/~pgunning/LaunchGM.html

 

R is a free downloadable statistical and graphical system.  A description of some of the main features of R are given in Chapters 1, 2 and 3 of the book Probability with R: An Introduction with Computing Applications and summarised in Notes on R,

 

The project is suitable a   4th year student or an 4th year Mathematical Science student who has taken the CA325 module in Survey Design.