Assignment 7

pdf

School

Pennsylvania State University *

*We aren’t endorsed by this school

Course

504

Subject

Statistics

Date

Apr 3, 2024

Type

pdf

Pages

1

Uploaded by ChefLyrebird3739

Report
STAT 504 Assignment [L7] 1. Berkeley admissions data, available within R using the commands below, summarizes depart- ment, sex, and whether the person was accepted or not. data(UCBAdmissions); x = UCBAdmissions If we ignore Department A (so that dept has five levels), perform backward elimination on the full model with dept, sex, and their interaction, using AIC as the criterion for removal. Explain what happens at each step of this procedure, and report the final model. For conve- nience, the data without department A can be read in from the file “admitBCDEF.csv”. x = read.csv("admitBCDEF.csv", header=T) 2. The columns in the data set “Tbl6.15.txt” correspond to duration of surgery (in min), type of device used to secure the airway (0=laryngeal mask, 1=tracheal tube), and whether the patient experienced a sore throat on waking (1=yes, 0=no). (a) With sore throat as the event of interest, consider the logistic model with both main effects (duration and type) and their interaction. What would interaction imply about the odds ratio for sore throat when comparing the two type groups? (b) Explain why the interaction is not an important term to include in this model. To support your answer, report either the LRT or the Wald test statistic and the AIC for models with and without the interaction term. (c) Report the fitted model (additive) model without the interaction term. In terms of the odds of sore throat, interpret the parameter estimates. (d) Summarize the predictive power of this model using a classification table. Report and interpret the sensitivity and specificity. You may use π 0 = . 5. 3. Consider a 2 × 3 × 3 table representing variables A, B, C, where we use logistic regression to estimate the probability of success for the first variable (A). (a) How many parameters does the full (saturated) model have? Write down the logistic regression equation. (b) Write the logistic regression equation for the model predicting A from B alone. How many degrees of freedom will the goodness of fit test have? (c) Write the logistic regression equation for the model predicting A from B and C with no interaction. How many degrees of freedom will the goodness of fit test have?
Discover more documents: Sign up today!
Unlock a world of knowledge! Explore tailored content for a richer learning experience. Here's what you'll get:
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help