Method Paper

Your grade in this course will be determined on a paper and presentation describing an additional statistical topic. You need to clear your planned project with the professor early in the semester, and you are encouraged to seek help and show early drafts.

Some suggested topics to cover include:

  1. Bayesian estimation (14.4, additional sources)
  2. Bayesian testing (14.4, additional sources)
  3. Non-parametric testing (14.1-3; pick one test and describe in detail)
  4. The chi-square test and its applications (Chapter 13)
  5. ANOVA: Comparing more than two means
  6. ANOVA: Using linear models
  7. Multiple hypothesis corrections
  8. Permutation testing

At the end of the semester, you will have produced an essay and a classroom presentation that will contain

  1. A description of the method
  2. Any theorems and proofs required to motivate or enable the method
  3. Conditions for use, applicability and rules of thumb
  4. An example of the method in use

Evaluation criteria

Each of the four paper tasks tasks is scored:

Acceptable is awarded for a completed report, with no more than 4 minor flaws. Any submission that misses the course deadline will be awarded this grade.

Good is awarded for a completed report without any flaws.

Excellent is awarded for a completed report without any flaws that also contains a description and evaluation of underlying philosophical differences to the topics covered.
[eg, if Bayesian, describe and evaluate difference to Frequentism; if non-parametric, describe and evaluate difference to parametric]

Minor flaws

As minor flaws count:

  • Significant language or readability issues, making it hard to understand the text
  • Arithmetic errors in an appropriately chosen statistical method

Major flaws that award a failing grade to the paper and thereby the course include:

  • Missing a deadline
  • Not covering some of the four requested topics at all
  • Readability issues to the point of our not being able to understand the text

Rubric

Task 1 - Method description

Good

Written in comfortably readable English, with proper academic citation styles for any relevant references.

Describes the method, when it is applicable, what assumptions it makes and what interpretations or conclusions can be drawn from its output.

Excellent

Describe what the main competing techniques are, and how or when this method excels.

Task 2 - Theory

Good

Written in comfortably readable English, with proper academic citation styles for any relevant references.

Describes the underlying theory – the why it works of the method.

Excellent

When applicable state any central theorems and outline their proofs.

Task 3 - Practice

Good

Written in comfortably readable English, with proper academic citation styles for any relevant references.

Work at least one complete example, on synthetic or real data.

Excellent

Work one complete example on real data, check assumptions and interpret the results.

Task 4 - Presentation

Good

Communicated clearly with a well-prepared talk using relevant and legible illustrations (slides or blackboard, as you prefer).

Respond to questions from the audience.

Excellent

The presentation puts the method in context, demonstrates the method “live” and handles questions with noticeable mastery of the material.