R-Studio Course Work

R-Studio Course Work Order Description R-Studio These problems use the Copelan bone marrow transplant dataset, available in" rel="nofollow">in the KMsurv package. 1. Install and load KMsurv, then load the ‘bmt’ dataset. Create factors if appropriate for further analysis. You can fin" rel="nofollow">ind variable defin" rel="nofollow">initions with the command > help(bmt) 2. Generate appropriate plots and summary statistics to explore the dataset and check for problem data. 3. Perform logistic regression to model one year post-transplant survival. Are there observations without full follow up? Explain" rel="nofollow">in your strategy for dealin" rel="nofollow">ing with any missin" rel="nofollow">ing outcome data. In your report, in" rel="nofollow">include the code used and the output from your fin" rel="nofollow">inal model. For one of the significant variables, explain" rel="nofollow">in what the odds ratio is and how to in" rel="nofollow">interpret it. 4. Plot disease free survival time usin" rel="nofollow">ing Kaplan-Meier. Plot the curves and comment on what you see. What is the three year disease free survival of the three patient types? Check whether these curves are different usin" rel="nofollow">ing the log-rank test. 5. Build a Cox model for post-transplant survival. Manually build the best model, explain" rel="nofollow">inin" rel="nofollow">ing why you in" rel="nofollow">included and excluded each variable. Compare your results to the logistic regression model. Which do you thin" rel="nofollow">ink is the better approach, and why? 6. Load the library ‘MASS’ and the dataset ‘birthwt’. For this section, assume that there is no missin" rel="nofollow">ing or bad data, no data transformations are required, and that the assumptions of each test are met. Answer the followin" rel="nofollow">ing questions. For each question, write a statement of the null and alternative hypothesis, in" rel="nofollow">include the code you used for the statistical test (this should not be more than 1-3 lin" rel="nofollow">ines), the R output from the statistical test, and whether the null hypothesis was rejected. You will graded based on in" rel="nofollow">inclusion of all required elements, appropriate choice of statistical test, and correct implementation and in" rel="nofollow">interpretation of the test. 7. Is there a relationship between race and smokin" rel="nofollow">ing status? 8. Is there a relationship between mother’s age and mother’s weight? 9. Is there a relationship between number of previous premature labors and low birthweight? 10. Create a multivariate model predictin" rel="nofollow">ing birth weight A. What are the significant predicators? B. Explain" rel="nofollow">in why this is not a very clin" rel="nofollow">inically useful model 11. Create a multivariate model predictin" rel="nofollow">ing if a child will have low birth weight A. What are the significant predictors? B. What is the in" rel="nofollow">increase in" rel="nofollow">in the odds of havin" rel="nofollow">ing a low birthweight baby if the mother smokes?