Messages - Page 2

1. On Wed Feb 24, Nils lectured from the first half of Ch 5, with 1-parameter extensions of given models, leading to tolerance radius characterisations etc. Celine went through 4-hour Exam Exercises 1, 2 from the 2015 set. 

We've uploaded "Nils, Lecture 7" and handwritten notes for the exercises ("Week 7") to the course site.

2. Exercises for Wed March 3 (this message will be edited and completed later): 

(a) You visit an island where everyone is fond of Model 0, namely the N(xi, 1). You creatively suggest the N(\xi,\sigma^2) model. What is the tolerance radius around M0? 

(b) Nils Collection Exercise 9, the kurtosis extension of the normal. 

(c) Gerda-Nils Book Exercise 5.8, an extended Poisson model.

Starting at 12:15: https://uio.zoom.us/j/63282276479 

** 1. On Wed Feb 17, Celine lectured on Ch 4, with AIC vs BIC behaviour. The AIC wins one competition, but BIC wins another one. Also, check the first few pages of Leo Breiman's "The Two Cultures" (and perhaps you should read C.P. Snow 1959 too).

** 2. Nils went through the two exercises called (a) and (b) in last week's message. An R script will be placed on the website for the Bayes & BIC exercise.

** 3. Note the existence and usefulness of the Hjelpetr?d A, now also Hjelpetr?d B.

** 4. For next week, do Exercises 1 and 2 from the four-hour exam 2015 -- you'll find this exam set at the 2017 website.

Starting at 12:15: https://uio.zoom.us/j/63754776945

In addition to (a) weekly teaching, so far via zoom, but perhaps by physical presence in Auditorium 4, in some weeks, and (b) weekly notes, from Celine and Nils, for lectures and exercises, we wish to open an additional communication channel (c): "hjelpetr?der", using ordinary mail, where comments and questions can be posed, discussed, answered.

This has been done, conveniently and successfully, in other recent Nils courses. It does not necessarily work well for Big Courses, but we're relatively few active participants on this occasion.

So, if interested, please send your mail address to Celine and Nils. Thanks.

1. On Wed Feb 10, Celine lectured on Ch 3, with the BIC and on hazard rate regression models. Nils went through some generalities and various details regarding Exam Project 2019, Exercise 1, the Swedish accidents dataset.

2. We've now placed on the course site (a) Celine's notes du jour, (b) Nils' "lecture 5" pdf, and (c) Nils' "com11a", the R script for the exam exercise. Run it, go through it, make sure you understand what goes on in its different parts, so that you may copy and modify for similar tasks later on.

3. Lars Henry Berge Olsen, mail address lholsen-at-math.uio.no, is our course's kurstillitsperson. We're a smallish group, so if you have concerns or comments (or advice!) to the course lecturers, you may by all means contact Celine or Nils directly ...

1. On Wed Feb 3 rounded off the material of Ch 2; see the pdf with "Nils lecture 4". Key words include the math behind AIC and AIC^*; cross validation; et frempek to the O(1/\rootn) modelling framework.

1b. Zoom-link for the lecture on Wed Feb 10:  https://uio.zoom.us/j/65023987156. The topic will be i) the derivation of BIC (3.2), and (ii) generalities on survival analysis (3.4). 

2. Celine went through several exercises, both extra C exercises and "N 20", on birds on islands, with Poisson regression models. See also her R scripts.

2b. Note that a new version of "Solution to additional exercises" has been uploaded: some mistakes have been corrected, and explicit expressions for all the expectations in C2 have been added (thanks to a comment from Lars). 

3. For Wed Feb 10, go through as much as you manage for Exam 2019 Project, Exercise 1. You need to...

1. On Wed Jan 27 Nils continued on Chapter 2 themes (ML behaviour outside models, delta method, sandwich variance matrix, AIC, AIC^*). Check out the slides. C¨¦line went through exercise N2 and commented on parts of N1, some R code and handwritten notes have been made available.

2. We are still not allowed to have physical lectures next week, so we will continue with zoom lectures for now (in the assigned lecturing hours: Wednesdays from 12:15 to 15): https://uio.zoom.us/j/68175596867. Some key words for next lecture: finishing Chapter 2 (more AIC, AIC*, cross-validation). 

3. Exercises for Wed Feb 3: go back do exercise N4 (Egyptian lifetimes) and compute p* and AIC* for the four candidate models; do exercises C1 and C2 from the pdf document "Additional exercises (by C¨¦line)"; do question a)...

1. On Wed Jan 20 Nils gave a general overview of maximum likelihood theory (under the assumption that the given model is correct), along with a sketch of some proofs. Check out the slides. C¨¦line went through exercise N4 with various details, some R code and handwritten notes have been made available.

2. We are still not allowed to have physical lectures next week, so we will continue with zoom lectures for now (in the assigned lecturing hours: Wednesdays from 12:15 to 15): https://uio.zoom.us/j/64015240359. Note that these are live lectures, without any recording. Some key words for next lecture: ML behaviour outside models, delta method, sandwich variance matrix, AIC, AIC^*, examples. 

3. Exercises for Wed Jan 27: do exercise N2 from the Nils collection, wh...

1. On Wed Jan 13 we talked about some practicalities, gave a brief overview of the course content and started on some basics from Ch 2, concerning maximum likelihood and the AIC formula. Check out the slides. 

2. We are not allowed to have physical lectures next week, so we will do it digitally, on zoom: https://uio.zoom.us/j/65518671874. From 12:15 to 14:00 there will be lecturing, then exercises from 14:15 to 15:00.

3. Exercises for Wed Jan 20: do exercise 4 from the Nils collection. You can find the Egyptian lifetimes dataset under the "Data, exercises and R programmes folder" (on the left). Feel free to add some more candidate models for example the extended Gamma-Weibull model from exercise 7 (don't do the rest of Ex 7 yet), or a model of your own invention. 

We have uploaded Version A of Exercises & Lecture Notes, as of 12-i-2021 -- there'll perhaps be Versions B, C, D later on during this semester. You should print out a copy for yourself. We'll go through quite a few of these exercises during the course (and also several from the Claeskens-Hjort Model Selection book).

Nils and C¨¦line will have a joint zoom lecture on the 13th of January: https://uio.zoom.us/j/61338838456?pwd=WmRLb3dSeFBVSEY2UDN2OFAzQW9Xdz09. From 12:15 and lasting approximately one hour. We will discuss practicalities, the content of the course and start a bit on chapter 2 from the Claeskens and Hjort book. 
 

Nils and I would like to know a little bit about *you* (our students), so we made a small questionnaire for you: https://nettskjema.no/a/179536. There we ask a bit about your background, motivation and preferences concerning teaching. This is completely voluntary and anonymous.