Reference no: EM133907515

Biostatistics

Assignment 1

Question 1

Examine the relationship between cholesterol (CHOL) and age (AGE) using Proc GLM to fit a linear model for CHOL that allows a quadratic relationship with AGE (i.e. include AGE and AGE*AGE).

Write down the algebraic representation of the fitted (i.e. with estimated coefficients) quadratic relationship.

Is the relationship between cholesterol and age curved or straight? Provide evidence from your output for your answer.

Use Proc GLM to fit a linear model for CHOL that allows separate quadratic relationships with AGE for men and women.

Write down the algebraic representation of the fitted quadratic relationships separately for men and women.

Perform (and interpret) a test of whether the fitted quadratic curve for men is significantly different from the quadratic curve for women.
Obtain estimates (from the fitted model) of mean cholesterol for men and for women with age 40, 50, 60, and 70 years and describe the difference in estimated mean cholesterol for men and women at each age.

Use Proc GLM to fit a linear model that tests whether there is a difference in mean CHOL for people with and without diabetes.

What is the estimated difference (and 95% confidence interval) in mean CHOL for people with and without diabetes.

What is the p-value for testing if there is a difference in mean CHOL for people with and without diabetes.

Use Proc GLM to fit a linear model that tests whether there is a difference in mean CHOL for people with and without diabetes after adjustment for sex and age (use your results from (a) and (b) to decide how to fully adjust for sex and age).

What is the estimated difference (and 95% confidence interval) in mean CHOL for people with and without diabetes after adjustment for sex and age. Boost your grades with expert assignment help today!

What is the p-value for testing if there is a difference in mean CHOL for people with and without diabetes after adjustment for sex and age.
Describe how adjustment for sex and age has changed your results comparing mean CHOL in people with and without diabetes.

Question 2

Use Proc GLMSELECT to conduct a stepwise (backward) search of predictors for FEV from among the following list of potential predictors: sex, asthma, smoking, diabetes, age, height, bmi, and exercise. In your analysis consider sex, asthma, smoking and diabetes as categorical variables and the remainder as quantitative variables. In your search of predictors consider main effects and interactions with sex only (but do not consider squares of quantitative variables) and use p=0.01 criteria for dropping variables. Provide output that shows the order in which terms were dropped and the output showing the fitted final model.

Use Proc GLM to fit the final model from GLMSELECT and obtain and interpret the estimated effect (on FEV) of each variable in the final model.

Write down the algebraic representation of the fitted model from (b).

Obtain and interpret a measure of how well this model predicts FEV.

If the final model was used to predict FEV, what would be the estimated difference in predicted FEV for the following male and female:
sex=male, asthma=no, smoking=never, age=50, height=1.8, BMI=25, exercise=7.
sex=female, asthma=no, smoking=never, age=50, height=1.7, BMI=26, exercise=7.