Biostatistics 653 OUTLINE

Home \| Schedule \| Classes \| Problem Sets \| Labs \| e-Mail Biostatistics 140.653 Methods in Biostatistics III Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health
Date	Class	Work Due	Lecture Topic	Reading Assignment

Tues Jan 22	1		Introduction/overview Statistical modeling Regression and correlation Parameter interpretation: slopes; means Analytic purposes	FEH: Ch 1 SW: Ch. 1
Thurs Jan 24	2		Model and estimation: Simple linear regression Statement of model, assumptions Estimation: Least Squares "Quality" of estimation: Accuracy, precision	SW Ch. 2.1-2.4
Tues Jan 29	3		Simple linear regression: Sample characteristics and random component estimation Isolated points; influence ("sensitivity") Decomposition of variance: ANOVA table Residual variance estimation Brief inference introduction	SW: 2.5-2.9
Thurs Jan 31	4	Problem Set 1 due 5 PM FEB 1	Multiple linear regression: uses Multiple predictors Direct versus total effects Nonlinear relationships: Polynomials/splines Categorical predictors: Dummy variables	FEH: Ch 2 SW Ch. 6.1-2
Tues Feb 5	5		Model and estimation: Multiple linear regression Statement of model, assumptions Matrix specification Least squares in the multiple covariate setting Gauss-Markov theorem Introduction to inference: variance components	SW: 3.1-3.4
Thurs Feb 7	6		Inference in multiple linear regression t-based inference for individual parameters Global/F-tests, regions for multiple parameters Confidence intervals for contrasts, model	SW Ch. 3.5; scan Ch. 4
Tues Feb 12	7		More on models with multiple covariates Adjustment Effect modification / interaction Mediation Multiple comparisons	Revisit FEH Ch. 2
Thurs Feb 14	8	Problem Set 2 due 5 PM FEB 13	Case study / review
Tues Feb 19	9		Midterm Examination
Thurs Feb 21	10		Model checking Residual versus predicted plots Standardized, studentized residuals Partial residual plots Outliers, influential points	SW Ch. 8-9
Tues Feb 26	11		Model checking: Two-stage regression Partial correlation / Adjusted variable plots Inference in the face of assumption violations Nonlinearity: transformations Heteroscedasticity: transfrms, weighting Correlation: robust variance	FEH Ch. 9; SW Ch 3.1; scan Chs 5 and 7
Thurs Feb 28	12	Problem Set 3 due 5 PM FEB 29	Prediction Inference for fitted values; sums of coefficients Colinearity Multiple R-squared Confidence bands / prediction intervals	SW Ch. 2.8.3; 10.1
Tues Mar 4	13		Prediction continued Overfitting; cross-validation Mallows’ CP (bias-variance tradeoff) PRESS	FEH: Ch 5
Thurs Mar 6	14		Model building strategies Parsimony Role of theory; variable groupings Data based methods: AIC, BIC Automated methods Extrapolation; Propensity scoring	FEH Ch. 4; SW Ch. 10.2-10.4
Tues Mar 11	15	Problem Set 4 due in class	Case study / review	FEH: Ch 7
Thurs Mar 13	16		FINAL EXAMINATION
FEH: Harrell, F.E. (2001), Regression Modeling Strategies, With Applications to Linear Models, Logistic Regression, and Survival Analysis, New York: Springer. SW: Weisberg S. (2005), Applied Linear Regression, 3rd. Ed., New York: John Wiley & Sons Home \| Schedule \| Classes \| Problem Sets \| Labs \| e-Mail Last edited: 14 January, 2008 ©2008, Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health All Rights Reserved