Back to Community
Trading Strategy Ideas: FMA 2015 Papers

Hey all,

I go to academic conferences regularly now, and there are always some interesting papers. I'm sure there are some good trading strategy ideas to be found in them. I am currently at FMA 2015, and will comment on this thread with each interesting paper I see.

The full conference program can be found here: http://www.fma.org/Orlando/OrlandoProgram.htm

One of the drawbacks of academic research is that testing out of sample is very difficult, and as a result I suspect that many of these papers are overfit. One of the beauties of Quantopian, however, is that we can reproduce these results and check. If anybody successfully reproduces, or provides evidence against, the findings of a paper, I would be happy to try to get in touch with the author and show them the new results. Maybe we can even get some profs commenting in the forum.

Thanks,
Delaney

Disclaimer

The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.

26 responses

Not from this conference, but cited by a paper here. This paper discusses why betting against beta may be a good idea.

http://www.econ.yale.edu/~af227/pdf/Betting%20Against%20Beta%20-%20Frazzini%20and%20Pedersen.pdf

This paper attempts to explain some movement in dividends and stock returns by M&A events. This is especially testable now that we have the EventVestor M&A dataset available in the store.

http://www.fma.org/Orlando/Papers/RiccardoSabbatucci_FMA15.pdf

This paper claims it's important to include global risk factors in your consideration, even when only locally invested. It uses GARCH models to estimate risk exposures. A lecture on GARCH models co-developed with a professor at MIT Sloan is available as part of our Lecture Series.

http://www.fma.org/Orlando/Papers/Pricingtogether_Akbari_Carrieri.pdf

This paper uses multivariate factor models to improve returns forecast for industries. This type of paper lends itself very well to trading, as they give an existing factor model. I'm wondering if a long-short equity strategy could be adapted in some way.

http://www.fma.org/Orlando/Papers/Bessler_Wolff_(2014)_-_Return_forecasts_in_portfolio_optimization_06_01_2015.pdf

This paper attempts to do volatility forecasting on a set of 10 stocks. Would be interesting to see how it performs on other assets, as with such a small sample size there's a high risk of it being overfit.

http://www.fma.org/Orlando/Papers/CovEstBenefits.pdf

Really cool idea contained in this paper. There are results in here claiming that stocks heavily shorted by the market tend to underperform compared to stocks that are not short sold. This is in essence a momentum strategy based on a short-selling volume signal. It's predictive over 15 trading days, so the corresponding strategy would be to periodically check how many short positions were held by every stock on the market, and then go short the most short and long the least short. Hold for 15 days and you should see a good return on that portfolio.

The paper uses Markit Equity Lending data. Wonder if there are other sources.

http://www.fma.org/Orlando/Papers/AP_Saffi_DeleveragingRisk_FMA.pdf

This paper says that positive news events about the large and well-known firms in an industry sector cause positive returns in other lesser known companies in that sector. The natural trading strategy here is to wait for positive news events for well known firms like Apple, and then buy tech stocks and short another industry.

http://www.fma.org/Orlando/Papers/DoFirmLevelCompanyNewsEventsGenerateEquityComovementPatterns.pdf

hey Delaney, thanks for this thread. Besides conferences, where else can we find the latest papers?

I find conferences and journals a good source. The nice thing about those is that the information is sorted already, and you can, for example, skip to the papers that discuss asset pricing. The conferences I attended last year were WFA, EFA, and FMA. I will also be attending AFA in January.

Aside from the conference websites it can be tricky to access the papers, as they often fall behind some kind of publisher paywall.

Thank you Delaney, some interesting reads here. Are there any that you favor right now?

Dear Delaney,
what journals di you recommend for trading strategies?

I recommend looking at asset pricing articles from the following conferences:

FMA
EFA
AFA
WFA

Keep in mind that a tremendous amount of overfitting exists in academic finance. Because all results are historical it's really hard to know what's meaningful and what is a random artifact of historical data.

Here is a good paper for building a market timing strategy:

http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2609814

Dear Delaney and Frank,
Thanks for your information.
Frank did you backtest these papers or are you aware of a backtest?

Cheers

Can you provide the links to the other past conferences you cite: EFA, AFA, WFA

Hi Bodo,

No I have not tried to implement either paper into an algo. Working on a strategy using the CAPM. I will share it when I am complete.

Here is a crude implementation of the Capital Asset Pricing Model, which is probably at the core of the financial literature:

(note, there are a lot of calculations done in this algo that I myself would challenge, just wanted to post something that attempts to implement the CAPM)

Clone Algorithm
25
Loading...
Backtest from to with initial capital
Total Returns
--
Alpha
--
Beta
--
Sharpe
--
Sortino
--
Max Drawdown
--
Benchmark Returns
--
Volatility
--
Returns 1 Month 3 Month 6 Month 12 Month
Alpha 1 Month 3 Month 6 Month 12 Month
Beta 1 Month 3 Month 6 Month 12 Month
Sharpe 1 Month 3 Month 6 Month 12 Month
Sortino 1 Month 3 Month 6 Month 12 Month
Volatility 1 Month 3 Month 6 Month 12 Month
Max Drawdown 1 Month 3 Month 6 Month 12 Month
# Backtest ID: 5713e1c1bf91800f858cf317
There was a runtime error.

Very cool, would be interesting to see it implemented in pipeline as a long short factor. Betting for and against beta is a factor many academics have studied. Unfortunately it may be tricky before we have a correlation/regression operation elegantly implemented in pipeline.

Thanks Delaney. Here is my first attempt at migrating for Q2 :

Clone Algorithm
21
Loading...
Backtest from to with initial capital
Total Returns
--
Alpha
--
Beta
--
Sharpe
--
Sortino
--
Max Drawdown
--
Benchmark Returns
--
Volatility
--
Returns 1 Month 3 Month 6 Month 12 Month
Alpha 1 Month 3 Month 6 Month 12 Month
Beta 1 Month 3 Month 6 Month 12 Month
Sharpe 1 Month 3 Month 6 Month 12 Month
Sortino 1 Month 3 Month 6 Month 12 Month
Volatility 1 Month 3 Month 6 Month 12 Month
Max Drawdown 1 Month 3 Month 6 Month 12 Month
# Backtest ID: 57197c133ad6380f75a04017
There was a runtime error.

Dear Frank,
thanks for sharing this interesting algo, may I ask what is the meaning of the variable

 # Makeshift adjustment for risk free rate...will fix later  
    Distribution_Adj = 0.85  

Thanks
Francesco

Francesco,

That was sheer laziness. I wanted to use the bond ETF to calculate a dynamic risk free rate for inclusion in the CAPM. The inclusion of the 0.85 was a shortcut around adding in the distributions from the ETF. Essentially I just hard coded in a a risk free rate of 0.85, which is added to the rolling annual appreciation/depreciation calculation of the bond fund.

Regards

Frank

I'm a mid-career data scientist and statistician. After a few pages of http://www.fma.org/Orlando/Papers/Bessler_Wolff_(2014)_-_Return_forecasts_in_portfolio_optimization_06_01_2015.pdf, this is the closest to what I plan to do. I'm new to Q, I've back-tested a few. As best I understand from the forums, I need to learn the research environment. I welcome advice.

You might find the lecture series useful for learning the research environment. Once you get towards the end you'll find examples closer to what you see in the paper. And yes the vast majority of the work a good quant does will be in the research environment. Backtesting is too easy to overfit in my opinion and should be viewed as the final step in an otherwise research-dominated workflow.

https://www.quantopian.com/lectures

“(Jacobi) knew that it is in the nature of things that many hard problems are best solved when they are addressed backward,” Munger counsels.

So what happens when you skip over the current market move and think about the next one instead?

This strategy idea looks for an extreme in VIX, waits for a pause, and then buys equities.

http://henrycarstens.com/101-trading-ideas-trade-the-next-move-strategy-101tradingideas/