Project 1

  • Due No due date
  • Points 1
  • Submitting a file upload
  • File types pdf
  • Available 25 Jan 2023 at 0:00 - 31 Mar 2023 at 23:59
This assignment was locked 31 Mar 2023 at 23:59.

General project instructions can be found here

Complete project instructions: Download project1.pdf

Students will work on the projects in groups of at most two members

A computer written, self-containing, report of the subjects presented in the instructuions should be handed in. The name of the attached file should be: SF2930 Project1-FullName1-FullName2.pdf

Introduction

This project aims for illustrating two typical scenarios in modern regression modeling. When performing the project, you have to choose only one of the two following alternative scenarios. Your choice must be clearly specified in the project report.

  • Scenario I: Large-Sample Regression, p < n. You are expected to work with classical methods of statistical inference which are applicable in this case.
  • Scenario II: High-Dimensional Regression, p > n or p>> n. In this situation, scientist is often looking for a few informative predictor variables hidden in an ocean of uninformative ones. There is a variety of approaches, to tackle this "curse of dimensionality" problem in regression analysis, you are expected to focus on a subset of such.

All the R-function needed to perform this project will be presented during the
exercise sessions and are available on the exercises page.

Data for Scenario I:

Choose one of the two datasets to work with

Download bodyfatmen.csv

Download bodyfatwomen.csv

Data for Scenario II:

The data is available in the hdi R package, see instructions in the pdf.

For questions about the projects contact the teaching assistant, contact information is at the start page for the course.

0