Quantitative analysis tools for the beginner

Posted: March 30, 2010 in Uncategorized
Tags: , , , , , , , ,

My main goal here is to provide you with a basic understanding of the basic tools of the quant in the industry.  This list is far from comprehensive or exhaustive by any means.

Excel – basic spreadsheet system that is used for prototyping basic models and has the ability to add more functionality thru add-ins.  If you are extremely new to finance…..LEARN EXCEL!  It will be sitting on the desk of every trading shop that you visit (by shop I mean proprietary trading firm, arcade, etc.).  Hedge funds and many other firms (Broker-Dealers, Mutual Funds) have their own in-house software that replaces these items…see further below.

Useful EXCEL Add-ins:

  1. PopTools (http://www.cse.csiro.au/poptools/) – Stellar add-in that allows the user to easily perform many different mathematical operations.  This is probably the best of the shelve tool for performing Excel based principle component analysis in my opinion.
  2. MATRIX (http://digidownload.libero.it/foxes/SoftwareDownload.htm) – Awesome add-in with many of the functions of PopTools.  I prefer PopTools.

Matlab – the next step up from Excel; there are many things you can do here such as writing your own functions, libraries, etc that are very difficult in Excel but the cons are the costs for the software is very high and the toolboxes that you often need are very expensive as well.

R – the open source competitor to Matlab that I actually like much better.  There are a plethora of packages for R and cover most of the same realm in finance as their Matlab counterparts but the big difference is price.  Everything that I have seen for R has been free!  I really prefer R and whole-heartedly recommend you learn it if you are new to the game.  If you don’t learn R then learn Matlab they are similar enough that knowledge carries over.

Octave – the open source alternative to Matlab; although it is available for Windows, this product is mainly a Linux cult software.  Not much good to say about it nor much bad to say about it other than there is not much in the way of toolboxes available like its competitors.

Python – I know its a scripting language but when I was in college long ago I tended to use it for pretty much everything before I learned about Matlab or R.  I actually tend to recommend learning Python to most beginners as it builds your resume, teaches you debug, and forces you to learn how to actually program useful things such as robust statistics!

JMulti – I don’t recommend it as it is hard to use.  If you have the time to learn it then it is actually very fully functional and FREE!  The graphical user interface does have some rather quirky issues at times.

Gauss – Old statistics package, at one time they were pretty much the market for co-integration, etc.  The torch was passed and this thing has been collecting dust.

SAS – Expensive and sluggish in my opinion and doesn’t really compete against the speed of developing your own packages in R or Matlab.  One thing that you will learn about me as we go on is that not many off the shelve software packages meet my stringent needs so most of the programs that I run are either custom or heavily modified.  This will cost you three arms, 10 legs, and an ovary….start growing extra limbs.

EViews – a pleasant but not robust econometrics package.

All said learn Python and get R.  If you know these 2 then learning anything else is a cake walk.  Once you have a handle on Python, check out PyTables for next discussion…..You’ll thank me later when you are dealing with TICK data!

Again, if you have question then ASK!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s