Academic Year 2018/19 - Term 2

MATH20802 - Statistical Methods 
Frequently Asked Questions


Q1) How much has the course syllabus changed in comparison to previous years?

The module has changed substantially. Essentially, all material relating to probability and not to statistics has been removed (e.g. on moment generating functions). Instead, new material (linear regression, Bayesian statistics) is now included.


Q2) Are the old lecture notes still relevant?

Parts of the old lecture notes are still relevant. In particular all the material about likelihood inference.


Q3) What material is examinable?

Generally, all content presented in the lectures and in the tutorials are examinable, except noted otherwise.  A list of topics relevant for the exam is found in the section "lecture contents" of the course web page - see https://strimmerlab.github.io/courses/2018-19/MATH20802/index.html#contents


Q4) Will there be revised typed typed lecture notes, and when will they be released?

Yes, there will be new typed lecture notes (on likelihood, regression and Bayesian statistics). These will be based on the handwritten slides and will be released in chunks during the course of the term.  You will find the new typed notes on Blackboard.  If you find any typos or errors in the new lecture notes please let me know.


Q5) What is the best way to prepare for the exam?

The best strategy is to revisit all the lectures by studying the lecture notes and revisiting the material from the example classes and the R sessions.  In addition, it is highly recommend to also consult a textbook (several of which are suggested on the course web page). To facilitate this use the keywords provided at https://strimmerlab.github.io/courses/2018-19/MATH20802/index.html#contents . Note that you will not be asked to do any R programming in the exam but you will need to be able to interpret R output.


Q6) Will the in-class test require using R or will it be similar to the format of the exam?

The in-class test in week 7 is an online test on Blackboard.  The examinable material includes everything taught up to week 5 (i.e. likelihood estimation and inference).  There will be no question about R, and no questions about regression and linear models (week 6 lectures).


Q7) What will be the format of the final exam?

It will be a written test of length 2 hours, counting 80%. See also the Section "In-class test and exam" at https://strimmerlab.github.io/courses/2018-19/MATH20802/index.html . A template of the final exam will be provided on Blackboard.


Q8) What is the format of the online mid-term exam? 

The in-class test is an online assessment on Blackboard and will take place in week 7 (worth 20%). See also the Section "In-class test and exam" at https://strimmerlab.github.io/courses/2018-19/MATH20802/index.html


Q9) Do we get a mock exam?

No, there will not be a mock exam.   However, the exam problems will be similar to the problems in the example sheets.  In addition to questions with calculations there will also be questions concerned with conceptual understanding. See also Q5 for suggestions how to prepare for the exam.


Q10) When is the material for the weekly tutorials and computer labs released?

The example sheets / instructions are released on Tuesday noon, and the solutions on Friday noon each week.


Q11) Is there a formula sheet provided during the midterm and the final exam?

No, both the in-class test and the final exam will closed book and you will not get a formula sheet.  However, note that all necessary details about distributions will be given as part of the question so you do not need to memorise the probability mass/density functions.