Hints and tips:
  • You may need to look up some unfamiliar terms in the title or description to get more of a sense of what is involved.
  • If the project has "No" in the "Current Availability:" field, it is already taken or not being offered this academic year but may be available again in future years.
  • The supervisor name links to a contact details webpage so, if you are interested, you can arrange to discuss this project or even propose a related topic of your own.
Level: BSc, MSci, MSc
Title: Lasso
Supervisor:

Dr Hugo Maruri-Aguilar
Research Area: Probability and Applications [Including Statistics]
Description:

Lasso is a methodology which simultaneously estimates parameters of a regression model and performs selection of terms by shrinking small estimates to zero. It is now a widely available modelling alternative to regression.

The project will consist of a revision of classic and recent literature and analysis of data sets. The software R includes packages for analysing data with this methodology and will be used as part of the project.
Further Reading:
  • J. Bien, J. Taylor, R. Tibshirani, A lasso for hierarchical interactions, Annals of Statistics 41 (2013) 1111–1141.
  • R. Tibshirani, Regression shrinkage and selection via the lasso, JRSSB 58 (1996) 267–288.
  • T. Hastie, R. Tibshirani, J. Friedman, The elements of statistical learning. Springer, New York, 2011.
Key Modules:

MTH5120 Statistical Modelling I (desirable)
Other Information:

Desirable prerequisites for this project are regression (MTH5120 Statistical Modelling I) and either programming experience or willingness to acquire it. Report to be written in LaTeX.
Current Availability: Yes