Course Overview

This course covers nonparametric analogs of the one- and two-sample t-tests and analysis of variance; the sign test, median test, Wilcoxon’s tests, and the Kruskal-Wallis and Friedman tests, tests of independence; nonparametric regression and nonparametric density estimation; modern nonparametric techniques; nonparametric confidence interval estimates.

Expected outcomes

By the end of the course, the student should be able to

• understand the assumptions underlying the nonparametric methods
• apply nonparametric methods to modern data analysis problems
• get hands-on experience in implementing methods and using existing R packages.

Course Information

• Term: Spring 2019
• Units: 3

Prerequisites

• STATS 60 or STATS 110 or STATS 160; Students should familiar with summary statistics, hypothesis testing, point estimation, interval estimation (confidence intervals), basics of statistical inference, and R.

Texts

• Required:
  • (HWC) Nonparametric Statistical Methods.
    • Authors: Myles Hollander, Douglas A. Wolfe, Eric Chicken
    • Edition: \(3^{rd}\) Edition
    • Print ISBN:9780470387375
    • Online ISBN:9781119196037
• Recommended:
  • (DH): Davison and Hinkley (1997). Boostrap Method and Their Application.
  • (ET): Efron and Tibshirani (1994). An Introduction to the Bootstrap.
  • (KM): Kloke and McKean (2015). Nonparametric Statistical Methods Using R.
  • (L): Lehmann (2006). Nonparametrics: Statistical Methods Based on Ranks.
  • (RHG): Ramsay, Hooker, Graves (2009). Functional Data Analysis with R and MATLAB.
  • (W): Wasserman (2006). All of Nonparametric Statistics.

Other sources

• Recommended Readings: a reading list will be posted on Canvas.
  • Re:BHLLSW2009: Buja, Cook, Hofmann, Lawrence, Lee, Swayne, and Wickham (2009). Statistical Inference for Exploratory Data Analysis and Model Diagnostics.
  • Re:D1983: Diaconis (1983). Theories of Data Analysis: From Magical Thinking Through Classical Statistics.
  • Re:DH1994: Diaconis and Holmes (1994). Gray Codes for Randomization Procedures.
  • Re:DH1995: Diaconis and Holmes (1995). Discrete Probability and Algorithms: Three Examples of Monte-Carlo Markov Chains: At the Interface Between Statistical Computing, Computer Science, and Statistical Mechanics, pg. 43-56.
  • Re:JWH2014: Josse, Wager, and Husson (2014). Confidence Areas for Fixed-Effects PCA.
• Useful links
  • Li:H1997: Holmes (1997). Lecture Notes on Computer Intensive Methods in Statistics.
  • Li:H2004: Holmes (2004). Lecture Notes on Complete Enumeration.
  • Li:C2016: Seiler (2016). Lecture Notes on Nonparametric Statistics.
  • Li:W2016: Wasserman (2016). Lecture Notes on Nonparametric Bayesian Methods.

Software

• In this course, we will use R for computing and R Markdown for producing lecture slides, solutions for homework assignments. R Markdown/Latex is highly recommended to write the midterm project proposal report and final project report. Install the following software:
  • R (required): https://www.r-project.org/.
  • R Studio is highly recommended for syntax highlighting, package management, document generation, and more: https://www.rstudio.com/.
    • The newest version of R Studio is highly recommended (v1.1.463).
  • Latex, which will enable you to create PDFs directly from the R Markdown in RStudio.
    • Latex, which will enable you to create PDFs directly from the R Markdown in RStudio.
    • Install Tinytex
      • install.packages(‘tinytex’)
      • tinytex::install_tinytex()

Evaluation

The final letter grade for this course will be determined by each method of assessment weighted as follows:

• Class participation (5%)
• Weekly homework assignments (50%)
• Midterm project proposal (10%, due on 05/03/2019)
• Final project (35%, due on 06/05/2019)

Course Schedule

R Markdown files

R Markdown files to create the lecture slides and PDFs are available in https://github.com/PratheepaJ/STATS205.