STATS 330 Advanced Statistical Modelling Assignment | University of Auckland

STATS 330 Course Learning Outcomes(CLO Outcomes)

• Define the generalised linear model and describe its underlying assumptions
• Explain and use a variety of statistical tools and procedures that determine the appropriateness of a fitted statistical model
• Summarise an appropriate modelling procedure, outlining how this is driven by the aims of the analysis
• Identify an appropriate candidate model to fit to a particular data set that is capable of answering the questions of interest
• Evaluate the appropriateness of a fitted statistical model, and take sensible steps to improve a model that is found to be inappropriate
• Write their own R code to carry out each step of the modelling procedure
• Communicate the findings of an analysis accurately and concisely

Question 1

At a research center for cats, scientists conducted a study to understand the health ofdomestic cats and the factors influencing their well-being. The study aimed to understandhow diet, activity, and other traits interact to predict health outcomes in cats.

The research involved cats from across the region. Each cat was monitored for the following attributes:

1. cat id: A unique identifier for each cat in the study.

2. diet: The type of diet consumed by the cat, categorized into:

• dry: Cats consuming primarily dry food.
• wet: Cats consuming primarily wet food.
• mixed: Cats consuming a combination of dry and wet food.

3. weight: The weight of the cat in kilograms.
4. activity level: The number of hours per day the cat is active in average, recordedusing a tracker.
5. age: The age of the cat in years, ranging from 1 to 15.
6. health score: A composite score reflecting the overall health of the cat, measuredon a scale from 0 to 5, with higher values indicating better health.
7. has illnessA: A binary response variable indicating whether the cat had illness A inthe last year:has illnessA = (1 if the cat had illness A,0 otherwise.
8. illness last year: A count variable representing the number of illnesses a cat hashad in the last year.

The researchers hypothesized that diet, weight, activity level, age, and overall health arerisk factors, or protective factors for illnesses.