This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 1 BIOST/EPI 513 Spring Quarter 2011 Dr. McKnight HOMEWORK 6 See Appendix I for STATA commands and Appendix II for output. These problems ask you to use data from the Western Collaborative Group Study (WCGS) found in wcgs.dta and documented in wcgs.codebook.txt on the Assignments page of the class website. 1. Using the WCGS data and logistic regression, examine the evidence for whether systolic blood pressure (SBP) 125 mmHg is associated with the risk of a CHD event (any type) during followup, after adjustment for age in years. (a) Give an ageadjusted odds ratio for the association of SBP 125 with the risk of a CHD event (any type) during follow up. 2.2 (b) Give a 95% CI for this odds ratio. (1.7, 2.9) using either regular or robust SEs (c) Choose an appropriate hypothesis test for whether this odds ratio is one and: i. State your choice and give a short (one sentence) justification for it The likelihood ratio test, Wald, and robust Wald tests are equally acceptable. Results are given in black, blue and red , respectively. ii. Give the value of the test statistic 32.4 5.48 5.49 iii. Name the reference probability distribution to which you compare it to obtain a Pvalue Chisquared with 1 df, standard Normal , standard Normal iv. Give the Pvalue for your test < 0.001 < 0.001 < 0.001 v. State your conclusions from the hypothesis test. Reject the null hypothesis of no association between SBP 125 mmHg and the risk of any type of CHD event after adjusting for age in years. 2. Using the WCGS data and logistic regression, examine the evidence that the odds ratio for the association of SBP 125 with the risk of a CHD event (any type) during followup depends on age. (a) Estimate three separate odds ratios for the association of SBP 125 with the risk of a CHD event (any type) during follow up, one for subjects 39 years old, one for subjects 49 years old and one for subjects 59 years old. 1.93 2.26 2.64 (b) Present 95% confidence intervals for these odds ratios. Table 1. Estimated ORs and associated CIs for subjects at various ages. Age (years) Estimated OR 95% CI 95% robust CI 39 1.93 1.16, 3.21 1.17, 3.19 49 2.26 1.68, 3.03 1.68, 3.03 59 2.64 1.40, 4.95 1.42, 4.90 2 (c) Choose an appropriate hypothesis test for whether the odds ratio depends on age and: i. State your choice and give a short (one sentence) justification for it The likelihood ratio test, Wald, and robust Wald tests are equally acceptable. Results are given in black, blue and red , respectively. ii. Give the value of the test statistic 0.38 0.62 0.63 iii. Name the reference probability distribution to which you compare it to obtain a Pvalue Chisquared with 1 df, standard Normal , standard Normal iv. Give the Pvalue for your test. 0.536 0.537 0.529 v. State your conclusions from the hypothesis test. Fail to reject the null hypothesis that the odds ratio for the association of SBP 125 with the risk of any type of CHD event during followup does not depend on age. during followup does not depend on age....
View
Full
Document
This note was uploaded on 02/09/2012 for the course STAT 513 taught by Professor Barbaramc.knight during the Spring '11 term at University of Washington.
 Spring '11
 BarbaraMc.Knight

Click to edit the document details