Wednesdays 2-5pm BL 325

The course will focus the using and interpreting advanced statistical methods with applications in a number of different areas. This course is designed for Master and PhD students in Statistics, and is REQUIRED for the Applied paper of the PhD Comprehensive Exams in Statistics. The course is a mixture of theory and applications, and will include a number of projects which will involve computing with R.

Statistical methodology to be covered includes:

- generalised linear models and maximum likelihood estimation
- linear mixed models for longitudinal and hierarchical data
- Bayesian inference using INLA
- generalized linear mixed models
- semi-parametric regression
- survival analysis

Additional topics will include:

- Scientific writing
- Reproducible research with R

- Multivariable calculus
- Matrix algebra
- Upper-division courses on probability and statistical inference (i.e. STA 347, STA 422)
- Methods of Applied Statistics 1
- Statistical Computing using R

Four homework assignments will each account for 15% of the final grade, with a final exam providing the remaining 40%.

- 38 Jan: Generalized linear models
- 27 Feb: Longitudinal data
- 13 March: Generalized linear mixed models
- 2 April: Survival analysis or semi-parametric modelling

23 February 2019 (?)

The course will not adhere closely to any single text, though the following will be referred to frequently.

- Maindonald & Braun,
*Data analysis and graphics using R: An example-based approach*, 2010 - Davison,
*Statistical models*, 2003 - Wakefield,
*Bayesian and frequentist regression methods*, 2013

Students will need to have access to a computer running a recent version of R, and several additional packages for R will be installed. Those not familiar with R are encouraged to become so within the first few weeks of the course.

- Course web page: pbrown.ca/teaching/astwo
- email:
`patrick.brown@utoronto.ca`

office:

- SS 6026C Wednesdays and Thursdays
- Centre for Global Health Research, St. Mike’s Hospital at other times

office hours: Wednesday 9-10, Thursday mornings 10-11

- Frequentist inference with Generalized Linear Models
- Models and inference
- Applications and interpretation

- Applied statistics in practice
- scientific writing
- reproducible research

- Linear mixed models and longitudinal data
- Mixed models, maximum likelihood estimation and REML
- Applications and interpretation
- Correlation in time
- Random coefficient models

- Generalized Linear Mixed Models and Bayesian inference
- Bayesian inference and INLA
- Random effects models for non-Gaussian data
- Applications and interpretation
- Advanced INLA

Smoothing and semi-parametric models

Survival analysis

Spatial statistics

Davison, A. (2003). *Statistical models*. Cambridge University Press.http://books1.scholarsportal.info/viewdoc.html?id=/ebooks/ebooks1/cambridgeonline/2012-11-08/1/9780511815850

Maindonald, J. & Braun, W.J. (2010). *Data analysis and graphics using R: An example-based approach* (3rd edition). Cambridge: Cambridge University Press.https://www.cambridge.org/core/books/data-analysis-and-graphics-using-r/E04AEC5BCEF09D2E51A63EB5A8CB0680

Wakefield, J. (2013). *Bayesian and frequentist regression methods*. Springer New York.