5
$\begingroup$

Does anyone know of a decent quant/stat factor website, distribution(public or private) or publication that tracks performance of "many" of traditional quant/stat factors? By that I mean would show daily, weekly, or monthly performance of enough traditional factors just to get a general idea of what is working lately/been trending for a long time/etc.. (e.g. 3 day reversion strategies worked quite well the last two weeks and open/close bracketing strategies got crushed last month.)

I've seen the occasional distribution from a bank that will show a handful of factors they run on a quarterly/annual basis, and Bloomberg has some functionality (e.g. BTST) that leaves a lot to be desired, but I've got to imagine there's something more systematic/open source out there, especially with the (slow) rise of more cloud based backtesting services, since what I'm looking for really just requires some person/group having coded enough factors up in a platform and a simple report on top. (The data quality is not too critical, but naturally the more flawed the less useful!)

Any pointers?

Much appreciated all! I'll certainly post anything I find as well.

$\endgroup$
  • $\begingroup$ quantopian.com offers something of the sort. The algorithms are open source and submitted by the public so I'm not sure about the level of quality but you might be able to find the occasional gem and follow it as it performs over time. $\endgroup$ – Cristian Dima Jun 30 '13 at 11:17
  • 1
    $\begingroup$ quantpedia.com also offers something of the sort. Some of the website is behind a paywal but you can try a free account to get a basic idea of what the site offers. $\endgroup$ – Cristian Dima Jun 30 '13 at 11:56
1
$\begingroup$

Have you looked at Kenneth French's website or the AQR data library ? They do a good job of tracking the classic ones.

If you are looking for paid content, such as the one pointed out below there are plenty of vendors that provide that data although most of them are for institutional clients - look at factset, s&p capital iq, thomson reuters starmine, qsg, apt. Also Barra and Axioma. It is a highly commoditized product at this point.

$\endgroup$
0
$\begingroup$

I use Portfolio123. It is a North American-focused equity research and trading platform targeted towards retail investors. There are tools for researching and tracking factors, designing and testing systems, screening, performance tracking, portfolio simulation, and even trading live algorithms. Some users follow others' strategies, others adapt off-the-shelf systems, while more advanced users can define their own unique systems. The focus is on North American (US and Canadian) equities and ETFs, but there are additional tools for macro-factor research and back-testing.

A major selling point is that the company fundamental data, coming from S&P Capital IQ, is institutional quality and designed for point-in-time back-testing. Data quality makes a huge difference in mitigating look-back and survivor bias. Premium fundamental data is also, arguably, much richer in alpha than commodity fundamental data.

It would be more work to setup a factor tracking system than a pre-configured off-the-shelf dashboard. But within a few weeks, you could easily design your own tracking system for a number of factors and using a number of tools. And you also might find something unique and therefore be able to fend off market efficiency longer than many other better known quant factors.

There are a range of user types and prices. Basic membership is intended to let user follow others and tinker around with ideas. More premium membership allows access to more advanced tools, a larger repository of systems, and longer back-test periods.

Also, +1 on @Cristina-Dima's suggestions regarding Quantopian and Quantpedia. Quantopian is more of a competitor with Portfolio123, while Quantpedia is surely a supplement -- I use it frequently for idea generation.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.