Lawrence Joseph
Bayesian statistics
Professor
E-mail: lawrence.joseph@mcgill.ca
Telephone: +1 (514) 934-1934 ext: 44713
FAX: (514) 934-8293
Complete CV
Division of Clinical Epidemiology
McGill University Health Centre
Royal Victoria Hospital
687 Pine Avenue West
V Building, Room V2.10
Montreal, Quebec
Canada, H3A 1A1
Basic information
Course Outline
Introduction
[1] Sept 5
  • course description and evaluation
  • introduction to statistical analysis in medicine
  • math background
  • Colton
    • Chapter 1 pp 1--7
  • Moore and McCabe
    • Not covered.
  • Armitage and Berry
    • Chapter 1 pp 1--4
Data Summaries and Descriptive Statistics
[2] Sept 7 - Sept 12
  • types of data
  • histograms
  • stemplots
  • boxplots
  • means
  • medians
  • variance
  • relocating/rescaling
  • Colton
    • Chapter 2 pp 11--44
    • Boxplots and stemplots not covered.
  • Moore and McCabe
    • Chapter 1 pp 1--58
    • Chapter 2 pp 106--107
  • Armitage and Berry
    • Chapter 1 pp 4--40
    • Boxplots, stemplots not covered
Probability and Probability Distributions
[3] Sept 14 - Sept 21
  • laws of probability
  • discrete and continuous random variables
  • expectation and variance of r.v.'s
  • diagnostic tests and conditional probabilities
  • Bayes Theorem
  • Normal distribution
  • area under Normal curve
  • binomial distribution
  • Normal approximation to the binomial
  • Poisson distribution
  • Colton
    • Chapter 3 pp 63--92
  • Moore and McCabe
    • Chapter 1 pp 64--79
    • Chapter 3 pp 267--275
    • Chapter 4 pp 287--357
    • Chapter 5 pp 374--390
    • Diagnostic tests not covered
  • Armitage and Berry
    • Chapter 2 pp 41--77
    • Chapter 16 pp 522--525
Inference Concerning Means
[4] Sept 26 - Oct 19
  • random sampling
  • hypothesis testing for means
  • type I and type II errors
  • p-values
  • confidence intervals for means
  • t distribution
  • paired and unpaired samples
  • Bayesian inference
  • sample size calculations and power
  • Colton
    • Chapter 4 pp 99--146
    • Bayes not covered
  • Moore and McCabe
    • Chapter 5 pp 397--405
    • Chapter 6 pp 432--493
    • Chapter 7 pp 502--555
    • Bayes not covered
  • Armitage and Berry
    • Chapter 3 pp 78--84
    • Chapter 4 pp 93--114, 146--149
Midterm Exam
Tuesday October 24, 2000, 9:00 am - 11:00 am
  • Room N2/D2, Stewart Biology Building
Inference concerning proportions and counts
[5] Oct 26 - Nov 9
  • hypothesis testing for proportions
  • sample size calculations and power
  • paired and unpaired samples
  • chi_2-test to compare 2 or more proportions
  • Fishers exact test
  • Bayesian inference
  • Mantel-Haenzel to combine 2 x 2 tables
  • relative risk and odds ratios
  • inference for count data
  • Colton
    • Chapter 5 pp 151--183
    • Bayes, Mantel- Haenzel, relative risk and odds ratio not covered
  • Moore and McCabe
    • Chapter 8 pp 584--609
    • Fishers exact test, Bayes, Mantel-Haenzel, relative risk, odds ratios and counts not covered
  • Armitage and Berry
    • Chapter 3 pp 84--85
    • Chapter 4 pp 118--152
    • Chapter 16 pp 508-519
    • Bayes not covered
Nonparametric Statistics
[6] Nov 14 - Nov 16
  • sign test
  • Rank sum test
  • Wilcoxon signed rank test
  • CI for median
  • Colton
    • Chapter 7 pp 219--226
    • Sign test and CI not covered
  • Moore and McCabe
    • Chapter 14 (0n CDROM only)
  • Armitage and Berry
    • Chapter 13 pp 448--460
    • CI not covered
Regression and Correlation
[7] Nov 21 - Dec 5
  • difference between regression and correlation
  • scatter plots
  • linear regression
  • least squares method
  • estimation of parameters in regression
  • Bayesian inference in regression
  • basic design in regression
  • other types of regression
  • Pearson's correlation
  • Spearman's correlation
  • Colton
    • Chapter 6 pp 189--214
    • Bayes not covered
  • Moore and McCabe
    • Chapter 2 pp 126--168
    • Chapter 10 pp 660--694
    • Bayes not covered
  • Armitage and Berry
    • Chapter 5 pp 154--171
    • Bayes not covered
Final Exam
Thursday December 7, 2000, 9:00 am - 12:00 pm
  • Room 129, Education Building