TestMagic

DismalScientist · 2007 July 7th, 09:44 PM

What is the bare minimum preparation (as far as math/stat) that is required to do well in first year econometrics courses? Is there any book (for preparation) that you guys recommend? Does math camp usually include refresher on the kind of math you might need for metrics?

~DismalScientist

asquare · 2007 July 7th, 09:55 PM

Absolute bare minimum is linear algebra. Be able to multiply matrices quickly. Otherwise, you will spend a lot of time trying to figure out what very simple expressions mean, and won't be able to concentrate on the more important concepts. Know what (x'x)^(-1)*(x'y) is

It's helpful to know basic probability -- joint probability, conditional probability, Bayes Rule, etc. But you will probably get a review of that at the start of your metrics sequence.

Also, if you've never used STATA, you might want to take a look at an online tutorial. Ditto for MATLAB, though fewer people know MATLAB coming in to the PhD program and you are more likely to get instruction in how to use it. You certainly don't need to master MATLAB yourself over the summer.

(Remember -- they let you in. Therefore, they think you can do the work. You don't have to try to teach yourself everything before you even start. Enjoy your summer; it's the last truly free time you are going to have for a while!)

notacolour · 2007 July 7th, 10:42 PM

I believe I've mentioned this elsewhere on the site, but I went into econometrics having never taken a stats course. It was a little tough at first, but I taught myself everything I needed to know from the mathematical appendices of Greene and Wooldridge.

My advice would be to get whatever econometrics book you'll be using, and look over the mathematical appendix. If you don't know big chunks, now would be a good time to learn.

werther · 2007 July 7th, 11:01 PM

Quote:

Originally Posted by asquare

Know what (x'x)^(-1)*(x'y) is

it sucks trying to translate that from the summation form. it's like speaking another language and trying to translate into another. i really wish my undergrad metrics prof taught us metrics using matrices, but oh well.

Golden Rule · 2007 July 8th, 01:14 AM

Quote:

Originally Posted by werther

it sucks trying to translate that from the summation form. it's like speaking another language and trying to translate into another. i really wish my undergrad metrics prof taught us metrics using matrices, but oh well.

I entered my econ phd program having taking several undergrad stats and courses, a few grad stats courses, and one grad and undergrad econometrics course... and it still took me until the middle of my second semester of first year to really comfortable jumping between matrix and summation notation, and doing the right algebra with each. It's something that profs and TAs I have had either don't spend time on or don't teach well. So point is your situation is hardly unique.

One day while studying for my final, I just spent a whole afternoon working with the appropriate section of the math appendix of Greene, and making sure I could derive every result -- I got by without doing that before -- but I thought that really helped a lot, and I probably should have done that sooner.

shootermcgavin7 · 2007 July 8th, 06:18 AM

Quote:

Originally Posted by DismalScientist

What is the bare minimum preparation (as far as math/stat) that is required to do well in first year econometrics courses?

I made it through with time & alcohol.

My suggestion would be a Math Stats II course.

2007 July 7th, 09:44 PM	#1 (permalink)
DismalScientist Let's Go Hokies! Join Date: Jan 2006 Posts: 192	1st year Econometrics! What is the bare minimum preparation (as far as math/stat) that is required to do well in first year econometrics courses? Is there any book (for preparation) that you guys recommend? Does math camp usually include refresher on the kind of math you might need for metrics? ~DismalScientist

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

2007 July 7th, 09:55 PM	#2 (permalink)
asquare TestMagic Guru Moderator Join Date: Mar 2006 Location: Ann Arbor Posts: 1,725	Absolute bare minimum is linear algebra. Be able to multiply matrices quickly. Otherwise, you will spend a lot of time trying to figure out what very simple expressions mean, and won't be able to concentrate on the more important concepts. Know what (x'x)^(-1)*(x'y) is It's helpful to know basic probability -- joint probability, conditional probability, Bayes Rule, etc. But you will probably get a review of that at the start of your metrics sequence. Also, if you've never used STATA, you might want to take a look at an online tutorial. Ditto for MATLAB, though fewer people know MATLAB coming in to the PhD program and you are more likely to get instruction in how to use it. You certainly don't need to master MATLAB yourself over the summer. (Remember -- they let you in. Therefore, they think you can do the work. You don't have to try to teach yourself everything before you even start. Enjoy your summer; it's the last truly free time you are going to have for a while!)

2007 July 7th, 10:42 PM	#3 (permalink)
notacolour TestMagic Guru-in-Training Join Date: Jul 2005 Location: North Carolina Posts: 805	I believe I've mentioned this elsewhere on the site, but I went into econometrics having never taken a stats course. It was a little tough at first, but I taught myself everything I needed to know from the mathematical appendices of Greene and Wooldridge. My advice would be to get whatever econometrics book you'll be using, and look over the mathematical appendix. If you don't know big chunks, now would be a good time to learn.

TestMagic

Free GMAT, GRE, SAT Test, TOEFL practice tests