There are two components to DIY quantitative investing. The first is know-how in building models and systems. The second is data to fuel the analysis, model or strategy.
It is this second component that often impedes otherwise capable analysts because finding valid, usable, well formatted data is either time consuming or expensive. We can't all afford Bloombergs @$20k a pop.
The democratization of quant demands as a prerequisite, the democratization of data. This is the mission that www.quandl.com is on. Quandl is part data platform, part search engine. It is a search engine in that it uses bots to find numerical data from all over the internet. It is a platform because it parses and organizes the data it discovers thus making it easy for Quandl users to find and use numerical data.
By bringing all numerical data into one place and building a smart index and interface Quandl opens up quantitative information to everyone. Quandl currently offers 4 million financial and economic time series for free. This includes FX rates, futures, commodities, interest rates, stock prices, indexes as well as business and economic indicators. A large fraction of the data an analyst needs to build models like the ones Gray and Carlisle discuss can be easily downloaded from Quandl.
Quandl has thousands of users already, but is still in its nascence. TKA readers should check it out if they have a few minutes.