lect06 - Categorical Data Analysis Lei Sun 1 CHL 5210...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Categorical Data Analysis - Lei Sun 1 CHL 5210 - Statistical Analysis of Qualitative Data Topic: Logistic Regression ctd. and Poisson Regression Outline • Logistic regression ctd. • Poisson regression. – Log-linear model. – An example of Poisson regression. – An example of Poisson regression for rates. • Homework 3. Categorical Data Analysis - Lei Sun 2 Multiple regression analysis - model building. Model building (selection of variables, diagnostics and validation) for linear multiple regression models generally applies to logistic regression models. • To obtain valid estimates of associations: – Specify variables. – Univariate exploratory analyses. – Construct full model. How many covariates should we include in our model given that we will be limiting attention to large sample procedures? – Assess interactions. The Challenge of examining for interaction (too many). Hierarchically well formulated models (all lower order components of any term are included ). – Assess confounding. – Examine the fit of your final model(s). Categorical Data Analysis - Lei Sun 3 • The LBW example. – Risk factors of primary interest (3 variables). * Mother’s smoking status (SMOKE). * Mother’s age (AGE). * Number of physician visits during the first trimester (FTV). – Potential confounders and effect modifiers (5 variables). * Mother’s race (RACE). * Mother’s weight (LWT). * Mother’s history of premature labor (PTL). * Mother’s history of hypertension (HT). * Presence of uterine irritability (UI). Categorical Data Analysis - Lei Sun 4 – How should we model mother’s weight (LWT)? (Univariate exploratory analyses.) data x1;infile "bwt.dat"; input id low age lwt race smoke ptl ht ui ftv bwt; lwtq= (lwt le 109) + 2*(lwt ge 110 and lwt le 120) + 3*(lwt ge 121 and lwt le 138) + 4*(lwt ge 139); lwt2=0; lwt3=0; lwt4=0; if (lwtq=2) then lwt2=1; if (lwtq=3) then lwt3=1; if (lwtq=4) then lwt4=1; proc freq;table low*lwtq; proc logistic descending; model low=lwt2 lwt3 lwt4; run; Table of low by lwtq low lwtq | 1| 2| 3| 4| Total---------+--------+--------+--------+--------+ 0 | 21 | 37 | 34 | 38 | 130---------+--------+--------+--------+--------+ 1 | 21 | 13 | 13 | 12 | 59---------+--------+--------+--------+--------+ Total 42 50 47 50 189 Model Fit Statistics Intercept Intercept and Criterion Only Covariates-2 Log L 234.672 226.071 Categorical Data Analysis - Lei Sun 5 Testing Global Null Hypothesis: BETA=0 Test Chi-Square DF Pr > ChiSq Likelihood Ratio 8.6013 3 0.0351 Score 9.0246 3 0.0290 Wald 8.6347 3 0.0346 Analysis of Maximum Likelihood Estimates Standard Wald Parameter DF Estimate Error Chi-Square Pr > ChiSq Intercept 1-0.00010 0.3086 0.0000 0.9997 lwt2 1-1.0458 0.4463 5.4911 0.0191 lwt3 1-0.9613 0.4490 4.5845 0.0323 lwt4 1-1.1524 0.4526 6.4820 0.0109 Odds Ratio Estimates Point 95% Wald Effect Estimate Confidence Limits lwt2 0.351 0.147 0.843 lwt3 0.382 0.159 0.922 lwt4 0.316 0.130 0.767 What is the reference group?...
View Full Document

This note was uploaded on 02/23/2012 for the course CHL 5210H taught by Professor Leisun during the Fall '11 term at University of Toronto.

Page1 / 43

lect06 - Categorical Data Analysis Lei Sun 1 CHL 5210...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online