Teaching myself maths (for Econometrics)?

[IceBus]

OK, I know we have boatloads of CS/talented maths folks on OcUK so I figured this would be as good a place as any.

I've just started a Graduate Diploma in Economics with the University of London International system. Amongst my core modules is the dreaded Econometrics. Unfortunately, it wasn't clear when I signed up how high the level of pre-existing maths knowledge is meant to be - as I only needed to show I had GCSE Maths (which I got an A in about ten years ago). I'm not (too) dumb, but maths is certainly not my forte and my undergrad was in political science.

To be able to start Econometrics I need to know multivariate differential calculus and some statistics (sampling distribution of an estimator, hypothesis testing and principles of statistical inference). Currently, all of this is greek to me.

I've signed up to www.mathtutordvd.com as it seems like a pretty good site, but I was wondering if anyone could offer any further advice or reading on getting my (very) rusty maths up to that standard?

Has anything else done something similar?

Books or any other resources are very much appreciated!

Thanks guys.

 

[muon]

This module is run by LSE and I assume they are making you do the equivalent of EC220.

http://www.lse.ac.uk/resources/calendar/courseGuides/EC/2010_EC220.htm

You should have at least A-level Maths (imo) and done S1-S3 or a university level (first year) statistics course.

If its this module

http://www.lse.ac.uk/resources/calendar/courseGuides/EC/2010_EC221.htm

then you will seriously struggle. It requires first year university level Mathematics with complete confidence in using Linear Algebra and Calculus.

-------------------------------------------------------

Luckily for you, decent Econometrics textbooks cover the bare essential Mathematics pre-requisites to a decent level (still need A-level Mathematics). I would recommend this textbook by Maddala

http://www.amazon.com/Introduction-Econometrics-G-S-Maddala/dp/0471497282

The course textbook for ec220 is actually this

http://www.amazon.co.uk/Introductio...1?ie=UTF8&s=books&qid=1294945319&sr=8-1-spell

 

[iamTakeo]

Look in your library for 'Maths for Economics', most of it was teaching me what I missed at A-Level maths. Also, learn Greek. I mean the alphabet, makes econometrics less scary. If your using econometrics packages there will be very little actual maths, just an understanding of what each particular test is doing and why and learning to make econometric models (those long scary equations with no numbers).

 

[muon]

If you really want to learn the Mathematics that first year economics students at LSE will learn before taking an introductory econometrics course, then you will need a decent linear algebra textbook, calculus textbook and statistics textbook.

I recommend FREE lectures you can get run by MIT via itunes.

http://ocw.mit.edu/courses/#mathematics

18.01 covers single variable calculus (A-level maths/fmaths imo)

18.02 covers multivariable calculus

18.03 covers differential equations (again cross between A-level and university level)

18.06 covers introductory linear algebra.

 

[muon]

Look in your library for 'Maths for Economics', most of it was teaching me what I missed at A-Level maths. Also, learn Greek. I mean the alphabet, makes econometrics less scary. If your using econometrics packages there will be very little actual maths, just an understanding of what each particular test is doing and why and learning to make econometric models (those long scary equations with no numbers).

I'm fairly sure this module will be equivalent to EC220 (linked above) and the exam will be almost entirely theoretical.

 

[iamTakeo]

I liked econometrics for things like what influences on house prices, but it has the potential to destroy your soul. I've had some horrific lectures in econometrics.

 

[IceBus]

Yup it looks like it is the EC220 course (and it is LSE run/administered) with the same core text etc.

I think I'm eventually going to need the stats and algebra for grad school anyway (US grad schools love quant heavy stuff) so it's probably as easy to learn it properly now and apply it to econometrics, rather than just learning the econometrics relevant parts and having to come back to it later.

Thanks a lot for the suggestions so far guys