PDA

View Full Version : Metrics textbooks



Gerter
03-17-2009, 10:13 AM
I just found out that I'm allowed to take textbooks and other material to my metrics exam. We're using Greene as the text but I'd also like to bring something else. I think it would help to have a book that lists the most common distributions and their properties (mean variance, MoM, MLE estimators, etc.) and also good comparative examples such as "Find this estimator for this distribution..". I'm also wondering what kind of notes would be helpful in the exam, I believe typical questions involve estimators, various distributions, hypothesis testing, etc. :hmm:

I think having the optimal material could definitely mean an extra 10-20%. Any ideas?

unitroot
03-17-2009, 10:29 AM
You really think that bringing a book you have never seen before will somehow help you pass some exam? Most books I have seen are hard to understand on first reading, much less useful when trying to lookup something in a hurry. Just bring your primary study aid (your class notes, Greene's book or whatever else you're supposed use to study for the test).

Gerter
03-17-2009, 10:44 AM
You really think that bringing a book you have never seen before will somehow help you pass some exam? Most books I have seen are hard to understand on first reading, much less useful when trying to lookup something in a hurry. Just bring your primary study aid (your class notes, Greene's book or whatever else you're supposed use to study for the test).

Why do you think I'll never have seen the book before the exam? I'm not going to decide on the book the day before the exam. I'm looking into this stuff now and the exam is in late May. That should give me plenty of time to get used to the book. I also find metrics books fairly straightforward with regards to content. You usually have a similar backbone and specific methods aren't hard to look up. I.e. if there was a Bayesian question on the exam, it shouldn't be hard to look up the relevant place in the book.

I'm also not taking a book to learn as I go during the exam. I'll know the methodology involved beforehand but would like a book with practical examples (with solutions) to compare to the exam questions (for example steps taken in a certain estimation technique). I find Greene lacking in that respect.

MorgieLilly
03-17-2009, 01:28 PM
I just found out that I'm allowed to take textbooks and other material to my metrics exam. We're using Greene as the text but I'd also like to bring something else. I think it would help to have a book that lists the most common distributions and their properties (mean variance, MoM, MLE estimators, etc.) and also good comparative examples such as "Find this estimator for this distribution..". I'm also wondering what kind of notes would be helpful in the exam, I believe typical questions involve estimators, various distributions, hypothesis testing, etc. :hmm:

I think having the optimal material could definitely mean an extra 10-20%. Any ideas?
I think Greene does that early in Chapter 2, lists the most comment distributions, their properties etc...

treblekicker
03-17-2009, 01:32 PM
you could just create your own reference sheet. if you've been paying much attention at all, it really shouldn't be too difficult.

Gerter
03-17-2009, 03:31 PM
Nope, nothing in chapter 2. There is a small distribution chapter in the appendix but not very thorough (no estimator properties).

And sadly paying attention in this course isn't enough. It's one of those teachers that likes to make exams with questions that weren't covered in the lectures or notes.

Dr. Who
03-17-2009, 05:34 PM
My math stats book has a table with general forms of the distribution, mean, variance, MGF for both discrete and continuous. I find it to be handy.

buckykatt
03-17-2009, 06:05 PM
If your exam will have any questions about actual applications, Peter Kennedy's book would be great to have on hand. It's available in a cheap paperback edition and it's actually one of those books you can pick up, look in the index, and find something useful and comprehensible about your topic of choice. It is no substitute for Greene, et al., but a wonderful supplement.