Recently I've read some books about quantative approach to fundamental investing: - What works on Wall Street - James O'Shaughnessy - Quantitative Value - Wesley Gray, Tobias Carlisle - Quantitative Strategies - Richard Tortoriello Basically, their research methodology, can be summarized as, we have a set of indicators: - value (E/P, EBIT/TEV, S/P, ...) - momentum (RSI, ...) - quality (Piotroski score,...) - growth (PEG, ...) We rank stocks and assign to deciles. We decide how often we rebalance portfolio (rather low frequency) and which strategy to apply. We calculate return,cagr, sharpe etc. for every decile/strategy.

I'm looking for free/open-source framework/library to reproduce similar research. I can't use yahoo data (non-yahoo stock exchange), so I need to load my own data. I consider to use python pandas for this, but maybe a better solution exists. Unfortunately, I've only found libraries for pair trading and technical analysis for single stock.

  • $\begingroup$ What analysis do you need to do that you can't do with numpy and pandas? Why don't you try going through the documentation and playing around with it a bit more? $\endgroup$ – John Mar 24 '14 at 22:08
  • $\begingroup$ I alwyas try to find better tool before I start to work. So if somebody has a similar problem and know better approach I'll be grateful for help. I hate reinventing the wheel. Btw. I've found this:SIT Maybe somebody implemented this or similar model in python/pandas? $\endgroup$ – Quant Christo Mar 25 '14 at 11:14
  • $\begingroup$ Python is a full programming language so it has a lot of potential, but you'll probably have to roll up your sleeves and program some of the stuff you want. There is dedicated backtesting software out there, but you'd have to pay for it. $\endgroup$ – John Mar 25 '14 at 13:04
  • $\begingroup$ I start off backtesting stuff on exactly what you are doing now, but long back in 2001. One advice I would give is to use a language that has a data frame support sooner than later. Back in those days when I started, R was the only option. It doesn't matter if you use pandas in python, or data frame / data table in R, tables in Matlab, deedle in C# / F#, these days tools can't get more easier for you. I wish pandas in python or tables in Matlab was available in 2001. I wouldn't have wasted precious years. $\endgroup$ – uday Mar 26 '14 at 18:34
  • $\begingroup$ Thanks @uday for your comment. I wonder if libraries like: QSTK, PyAlgoTrade or Zipline could be helpful with such approach? The second question is about pandas, I consider to implement in such way: i) we have data panel of dataframes indexed by time, ii) dataframe is indexed by equity names iii) columns in dataframe represents various parameters and ratios: price, P/E, PEG, P/S etc. Is it right approach? Maybe more complex dataframe is enough? $\endgroup$ – Quant Christo Mar 28 '14 at 8:07

Quantopian provides both the fundamental data (from Morningstar), as well as the backtest platform to reproduce results from the books you mentioned. Here's the introduction to our fundamentals offering: https://www.quantopian.com/posts/fundamental-data-from-morningstar-now-available-for-backtesting

(disclosure: I'm the ceo of quantopian)

| improve this answer | |

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.