Nancy Reid
 
Information
....Curriculum Vitae
....Contact Details
 
Research
Research overview
....Recent papers
....Recent talks
 
Teaching
....Current
....Previous
 
Miscellaneous
....doing a project
....FAQ
 
 

STA442/2101F: Applied Statistics I

Final Exam: Friday, Dec. 17, 2-5 pm STVLAD
You are permitted 1 sheet (double-sided) of handwritten notes 8 1/2 x 11

Homework 3

December 14 Solutions

Lectures

  • December 8
  • November 30, Dec 1
    • Slide from November 30 (admin only)
    • Slides from December 1: survival data
    • The R library survival is very handy for fitting survival data. There are some examples in the practicals for Davison's book, a pdf file available here.
    • This pdf from John Fox discusses coxph in detail, showing how to incorporate time-dependent covariates, and discussing model checking and residuals
    • This note discusses various types of bias that often come up in medical studies involving survival data
  • November 23, 24
  • November 16, 17
  • November 9, 10
    • Slides on generalized linear models. (Nov.10)
    • HW 3 due December 8.
    • R code illustrating sensitivity of coefficient estimates to constraints imposed.
  • November 2, 3
  • October 26, 27
  • October 19
  • October 12,13
  • October 5,6
  • September 28
    • Slides
    • Example G, annotated
    • Homework 1 due October 13 (corrected Sep 30)
    • Notes on prediction error
    • Ascii file of UN Data for homework 1, for those of you not using R. You can read this file into R by first putting into a file of your choosing, and then using read.table("filename", head=T, row.names=1). However you can just as easily load the data as indicated on the HW sheet.
  • September 23
    • FIX!! to load the datasets for the book, you should install "SMPracticals", not the "statmod" package that I told you. When I tried this, it asked me to first install another package, called "ellipse". After installing the package, I typed library(ellipse), then I installed the SMPracticals package. (A student who did this under windows had ellipse installed automatically.) Then type (within R) library(SMPracticals) and you should have all the text data sets available. For example, try data(venice) and data(nuclear)
    • Handout on linear regression using the Venice data
    • Also, check the list of R resources above: Jeff's is good for novices, and many further resources are available from the "Another list"
  • September 21
  • September 14

    Course information: pdf (recommended), html

    Text

    Statistical Models by A.C. Davison. We will emphasize Chapters 5, 8, 9 and 10.

    Computing

    You are welcome to use the statistical computing package of your choice, but I will refer exclusively to the R computing package. Statistics Dept graduate students can access R on the Statistics Dept computers; undergraduate students can access R on CQUEST. Alternatively, students can install R on the computer(s) of their choice, by downloading its "base" package (for free) from probability.ca/cran or www.r-project.org.

    Sep 30: I just stumbled across a very nice introduction to R, that has lots of Windows screenshots. It can be downloaded from this web site; it's the file "introduction to R".

    There are many helpful introductions to R, including:

    Course information

    STA 442/2101F: Methods of Applied Statistics, I
    Tuesdays, 2-4 pm, Wednesdays 3-4 pm (Note Time Change)

    Fall, 2010

    Course description: This course teaches methods of applied statistics, with the applications studied motivating the sets of methods taught. The undergraduate calendar description is:

    Advanced topics in statistics and data analysis with emphasis on applications. Diagnostics and residuals in linear models, introduction to generalized linear models, graphical methods. Additional topics such as random effects models, split plot designs, smoothing and density estimation, analysis of censored data, introduced as needed in the context of case studies.

    (Prerequisite: ECO327Y/357Y/STA302H)

    We will stay reasonably close to this set of topics, with less emphasis on generalized linear models, smoothing, and density estimation, as these topics are covered in Applied Statistics II.

    Grading: The grade in the course will be based on regular homework (60%) and a final exam (40%). Homework sets will include additional questions for graduate students in statistics.

    Dates for Homework

    available due worth
    HW 1 Sept 28 Oct 13 15%
    HW 2 Oct 13 Nov 3 20%
    HW 3 Nov 3 Dec 1 25%
    Final Exam 40%

    Note re grading: The three homework sets are of equal difficulty, but I have found that students improve quite a bit during the course, as they get more experience with the requirements and the background.

    Text: The course text is Statistical Models by A. C. Davison (Cambridge University Press). Two very good adjuncts are Data Analysis and Graphics using R by Maindonald and Braun and Linear Models with R by Faraway. Your STA 302 text, or any linear regression text, should also be helpful, and I will provide additional references during the course.

    Course web page(s): I am using Blackboard to manage the course lists and grades, but the course information is all on the web page http://www.utstat.utoronto.ca/reid/442F10.html. The Blackboard pages for both STA442F and STA2101F will lead you to this page via the first announcement.

    Computing: You are welcome to use the statistical computing package of your choice, but I will refer exclusively to the R computing package. Statistics Dept graduate students can access R on the Statistics Dept computers; undergraduate students can access R on CQUEST. Alternatively, students can install R on the computer(s) of their choice, by downloading its ”base” package (for free) from probability.ca/cran or www.r-project.org. There are many helpful introductions to R listed on the course webpage.

    Contact: Nancy Reid: SS 6002A, reid@utstat.utoronto.ca, 978-5046.
    Tuesday 4 to 5, Wednesday 4 to 5, or by appointment.