This preview shows pages 1–4. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: MATH 423/533 Course: MATH 423: Regression and Analysis of Variance MATH 533: Honours Regression and Analysis of Variance Day: Tuesdays/Thursdays Time: 11:35  12:55 pm Room: Burnside 1B24 Instructor: Prof. Russ Steele, Dept. Math and Statistics Office hours: TBA Lecture 1 p. 1/36 Outline Course description Course syllabus Grading Office Hours WebCT Lect u Course description Regression is a statisticians lifeblood Almost every reallife problem begins with a regression approach (if I know X , what does that tell me about Y ) Many standard techniques that you know are special cases of regression (ttests, ANOVA) Many advanced techniques can be reduced to variations on the standard regression problem (functional curve estimation, time series, survival analysis) Everything is regression What you NEED to know Basic probabiity (particularly conditional probability) and distributions (particularly the normal) Maximum likelihood inference (ttests, likelihood ratio tests, Ftests) Modest (more than basic, less than advanced) linear algebra tricks and theory (multiplication, symmetric, inverses, SVD, idempotency, etc.) Basic computing (keyboard, mouse, basic wordprocessing) What you WILL know Understand the three (well, 2.5) ways to understand the regression problem: Basic maximum likelihood Least squares Geometry of the expectation surface Modest experimental design Advanced applied data analysis Lecture 1 p. 5/36 Required prereqs You should have taken one of the following: MATH 323 AND 324 MATH 356 AND 357 Another mathematical probability and statistics sequence (preferably a full year) at another university And a course in linear algebra Lect u he Textbooks: MATH 423 and 533 Classicial and Modern Regression with Applications, by Myers, 2nd Edition Buy this book, Love this book Will be followed very closely At least half of assignments are likely to come directly from the textbook The Textbooks: MATH 423 and 533 Practial Regression and Anova using R by Faraway Available through link on WebCT (can also buy the text , but very little additional material in the nonPDF version ) Very good introduction to using R statistical package with a focus on linear regression and applications Additional Textbook Methods and Applications of Linear Models: Regression and the Analysis of Variance by Ronald Hocking Book is available as ebook through the McGill library system More mathematically rigourous text than Myers, although with less focus on applications Lecture 1 p. 9/36 The Grading Scheme 30% Total Mark will be assignments 20% Total Mark will be midterm 50% Total Mark will be Final Exam Lectu r The Exams The midterm and final exam will be split into two halves, half inclass, half takehome Each half is worth 50% of the grade for that exam (so 10% of overall mark for each half of the midterm, 25% of overall mark for the final) The inclass portions of the midterm and final exam will consist of theoretical and methodological questions...
View
Full
Document
This note was uploaded on 01/15/2010 for the course MATH 423 taught by Professor Steele during the Spring '06 term at McGill.
 Spring '06
 STEELE
 Statistics, Variance

Click to edit the document details