Week 7 Self-Assessment, Solutions

.pdf

School

University of Ottawa *

*We aren’t endorsed by this school

Course

1106

Subject

Statistics

Date

Feb 20, 2024

Type

pdf

Pages

5

Uploaded by ddddddddvv

Report
Logistic regression is flexible enough to allow you to predict a binomial outcome (survived: yes/no) with both categorical and continuous variables. Test whether chivalry existed on the Titanic as it sank. Set up your analyses using survived as the dependent variable, age as a covariate, and sex as a factor. Build the model in three blocks. Block 1 includes age, Block 2 sex, Block 3 sex*age (their interaction). Under Model Fit, make sure that AIC is checked as well as the overall model test. Under Model coefficients, make sure Likelihood ratio tests, odds ratio, and confidence intervals are all checked. Under estimated marginal means, you can make different plots. Set up three "Terms." Term1 Age, Term 2 Sex, Term 3 Age then Sex. Note. The following questions focus a lot on prediction. This means that the results are used to predict the odds and probability that a specific person would survive. The use of regression for prediction is one of its key applications and forms the basis of machine learning and predictive modeling. You have set up three different nested models. Based on the overall model Chi-squared tests, which models are significant? All of them are significant. Nested models are compared to see if the more sophisticated model is significantly better than the
simpler model. The model with the interaction in it significantly better that the model with just the two main effects. True There are multiple ways of assessing which model is best. The AIC is a common measure used to compare models. The smaller this measure is, the better the model. Which model has the smallest AIC? Model 3 Based on Model 2, females are more likely to survive the wreck than men.
True, when the reference level is set to Male, you can see that the LOGIT estimate is greater than zero (2.466 Based on Model 2, what is the odds ratio for a female surviving the wreck as compared to a man? Answer as X.XX The odds ratio is 11.77. Make sure Male is teh reference level Question 6 Assume your prof was on the Titanic. Let's try to predict whether he would survive! He is a 47 year old male, who could afford a second class ticket. Use Model 1, which only contains the age. Blank 1 What is the estimate (also known as the log-odds) of his survival? Answer as X.XXXX
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
  • Access to all documents
  • Unlimited textbook solutions
  • 24/7 expert homework help