Is Your Trading Strategy Still Working?

The Challenge of Validating Strategy Performance

One of the challenges faced by investment strategists is to assess whether a strategy is continuing to perform as it should.  This applies whether it is a new strategy that has been backtested and is now being traded in production, or a strategy that has been live for a while.

Fig 6All strategies have a limited lifespan.  Markets change, and a trading strategy that can’t accommodate that change will get out of sync with the market and start to lose money. Unless you have a way to identify when a strategy is no longer in sync with the market, months of profitable trading can be undone very quickly.

The issue is particularly important for quantitative strategies.  Firstly, quantitative strategies are susceptible to the risk of over-fitting.  Secondly, unlike a strategy based on fundamental factors, it may be difficult for the analyst to verify that the drivers of strategy profitability remain intact.

Savvy investors are well aware of the risk of quantitative strategies breaking down and are likely to require reassurance that a period of underperformance is a purely temporary phenomenon.

It might be tempting to believe that you will simply stop trading when the strategy stops working.  But given the stochastic nature of investment returns, how do you distinguish a losing streak from a system breakdown?

Stochastic Process Control

One approach to the problem derives from the field of Monte Carlo simulation and stochastic process control.  Here we random draw samples from the distribution of strategy returns and use these to construct a prediction envelope to forecast the range of future returns.  If the equity curve of the strategy over the forecast period  falls outside of the envelope, it would raise serious concerns that the strategy may have broken down.  In those circumstances you would almost certainly want to trade the strategy in smaller size for a while to see if it recovers, or even exit the strategy altogether it it does not.

I will illustrate the procedure for the long/short ETF strategy that I described in an earlier post, making use of Michael Bryant’s excellent Market System Analyzer software.

To briefly refresh, the strategy is built using cointegration theory to construct long/short portfolios is a selection of ETFs that provide exposure to US and international equity, currency, real estate and fixed income markets.  The out of sample back-test performance of the strategy is very encouraging:

Fig 2

 

Fig 1

There was evidently a significant slowdown during 2014, with a reduction in the risk-adjusted returns and win rate for the strategy:

Fig 1

This period might itself have raised questions about the continuing effectiveness of the strategy.  However, we have the benefit of hindsight in seeing that, during the first two months of 2015, performance appeared to be recovering.

Consequently we put the strategy into production testing at the beginning of March 2015 and we now wish to evaluate whether the strategy is continuing on track.   The results indicate that strategy performance has been somewhat weaker than we might have hoped, although this is compensated for by a significant reduction in strategy volatility, so that the net risk-adjusted returns remain somewhat in line with recent back-test history.

Fig 3

Using the MSA software we sample the most recent back-test returns for the period to the end of Feb 2015, and create a 95% prediction envelope for the returns since the beginning of March, as follows:

Fig 2

As we surmised, during the production period the strategy has slightly underperformed the projected median of the forecast range, but overall the equity curve still falls within the prediction envelope.  As this stage we would tentatively conclude that the strategy is continuing to perform within expected tolerance.

Had we seen a pattern like the one shown in the chart below, our conclusion would have been very different.

Fig 4

As shown in the illustration, the equity curve lies below the lower boundary of the prediction envelope, suggesting that the strategy has failed. In statistical terms, the trades in the validation segment appear not to belong to the same statistical distribution of trades that preceded the validation segment.

This strategy failure can also be explained as follows: The equity curve prior to the validation segment displays relatively little volatility. The drawdowns are modest, and the equity curve follows a fairly straight trajectory. As a result, the prediction envelope is fairly narrow, and the drawdown at the start of the validation segment is so large that the equity curve is unable to rise back above the lower boundary of the envelope. If the history prior to the validation period had been more volatile, it’s possible that the envelope would have been large enough to encompass the equity curve in the validation period.

 CONCLUSION

Systematic trading has the advantage of reducing emotion from trading because the trading system tells you when to buy or sell, eliminating the difficult decision of when to “pull the trigger.” However, when a trading system starts to fail a conflict arises between the need to follow the system without question and the need to stop following the system when it’s no longer working.

Stochastic process control provides a technical, objective method to determine when a trading strategy is no longer working and should be modified or taken offline. The prediction envelope method extrapolates the past trade history using Monte Carlo analysis and compares the actual equity curve to the range of probable equity curves based on the extrapolation.

Next we will look at nonparametric distributions tests  as an alternative method for assessing strategy performance.

Posted in Monte Carlo, Performance Testing, Portfolio Management, Stochastic Process Control, Strategy Development, Systematic Strategies | Leave a comment

Volatility ETF Strategy March 2015: +2.04%

HIGHLIGHTS

  • 2015 YTD: + 7.29%
  • CAGR over 40%
  • Sharpe ratio in excess  of 3
  • Max drawdown -13.40%
  • Liquid, exchange-traded ETF assets
  • Fully automated, algorithmic execution
  • Monthly portfolio turnover
  • Managed accounts with daily MTM
  • Minimum investment $250,000
  • Fee structure 2%/20%

 VALUE OF $1000

STRATEGY DESCRIPTION
The Systematic Strategies Volatility ETF  strategy uses mathematical models to quantify the relative value of ETF products based on the CBOE S&P500 Volatility Index (VIX) and create a positive-alpha long/short volatility portfolio. The strategy is designed to perform robustly during extreme market conditions, by utilizing the positive convexity of the underlying ETF assets. It does not rely on volatility term structure (“carry”), or statistical correlations, but generates a return derived from the ETF pricing methodology.

The net volatility exposure of the portfolio may be long, short or neutral, according to market conditions, but at all times includes an underlying volatility hedge. Portfolio holdings are adjusted daily using execution algorithms that minimize market impact to achieve the best available market prices.

Ann Returns

RISK CONTROL

Our portfolio is not dependent on statistical correlations and is always hedged. We never invest in illiquid securities. We operate hard exposure limits and caps on volume participation.

Sharpe

 

 

 

 

 

OPERATIONS

We operate fully redundant dual servers operating an algorithmic execution platform designed to minimize market impact and slippage.  The strategy is not latency sensitive.

MONTHLY RETURNS

Monthly Returns

 

 

(Click to Enlarge)

PERFORMANCE STATISTICS

PERFORMANCE STATS

 

 

 

 

 

 

 

 

 

 

 

 

 

(Click to Enlarge)

 

 

Posted in VIX Index, Volatility ETF Strategy, Volatility Modeling | Leave a comment

The Lazarus Effect

A perennial favorite with investors, presumably because they are easy to understand and implement, are trades based on a regularly occurring pattern, preferably one that is seasonal in nature.  A well-known example is the Christmas effect, wherein equities generally make their highest risk-adjusted returns during the month of December (and equity indices make the greater proportion of their annual gains in the period from November to January).

As we approach the Easter holiday I thought I might join in the fun with a trade of my own.  There being not much new under the sun, I can assume that there is some ancient trader’s almanac that documents the effect I am about to describe.  If so, I apologize in advance if this is duplicative.

The Pattern of Returns in the S&P 500 Index Around Easter

I want to look at the pattern of pre- and post- Easter returns in the S&P 500 index using weekly data from 1950  (readers can of course substitute the index, ETF or other tradable security in a similar analysis).

The first question is whether there are significant differences (economic and statistical) in index returns in the weeks before and after Easter, compared to a regular week.

Fig 1

It is perhaps not immediately apparent from the smooth histogram plot above, but a whisker plot gives a clearer indication of the disparity in the distributions of returns in the post-Easter week vs. regular weeks.

Fig 2

It is evident that chief distinction is not in the means of the distributions, but in their variances.

A t-test (with unequal variances) confirms that the difference in average returns in the index in the post-Easter week vs. normal weeks is not statistically significant.

Fig 3 It appears that there is nothing special about Index returns in the post-Easter period.

The Lazarus Effect

Hold on – not so fast.  Suppose we look at conditional returns: that is to say, we consider returns in the post-Easter week for holiday periods in which the index sold off in the  week prior to Easter.

There are 26 such periods in the 65 years since 1950 and when we compare the conditional distribution of index returns for these periods against the unconditional distribution of weekly returns we appear to find significant differences in the distributions.  Not only is the variance of the conditional returns much tighter, the mean is clearly higher than the unconditional weekly returns.

Fig 6


Fig 5

 

The comparison is perhaps best summarized in the following table.  Here we can see that the average conditional return is more than twice that of the unconditional return in the post-Easter week and almost 4x as large as the average weekly return in the index.  The standard deviation in conditional returns for the post-Easter week is less than half that of the unconditional weekly return, producing and information ratio that is almost 10x larger.  Furthermore, of the 26 periods in which the index return in the week prior to Easter was negative, 22 (85%) produced a positive return in the week after Easter (compared to a win rate of only 57% for unconditional weekly returns.

Fig 4

A t-test of conditional vs. unconditional weekly returns confirms that the 58bp difference in conditional vs unconditional (all weeks) average returns is statistically significant at the 0.2% level.

Fig 7

Our initial conclusion, therefore, is that there appears to be a statistically significant pattern in the conditional returns in the S&P 500 index around the post-Easter week. Specifically, the returns in the post-Easter week tend to be much higher than average for  periods in which the pre-Easter weekly returns were negative.

More simply, the S&P 500 index tends to rebound strongly in the week after Easter – a kind of “Lazarus” effect.

 Lazarus – Or Not?

Hold on – not so fast.   What’s so special about Easter?  Yes, I realize it’s topical.  But isn’t this so-called Lazarus effect just a manifestation of the usual mean-reversion in equity index returns?  There is a tendency for weekly returns in the S&P 500 index to “correct” in the week after a downturn.  Maybe the Lazarus effect isn’t specific to Easter.

To examine this hypothesis we need to compare two sets of conditional weekly returns in the S&P 500 index:

A:  Weeks in which the prior week’s return was negative

B:  the subset of A which contains only post-Easter weeks

 If the difference in average returns for sets A and B is not statistically significant, we would conclude that the so-called Lazarus effect is just a manifestation of the commonplace mean reversion in weekly returns.  Only if the average return for the B data set is significant higher than that for set A would we be able to conclude that, in addition to normal mean reversion at weekly frequency, there is an incremental effect specific to the Easter period – the Lazarus effect.

Let’s begin by establishing that there is a statistically significant mean reversion effect in weekly returns in the S&P 500 Index.  Generally, we expect a fall in the index to be followed by a rise (and perhaps vice versa). So we need to  compare the returns in the index for weeks in which the preceding week’s return was positive, vs weeks in which the preceding week’s return was negative.  The t-test below shows the outcome.

Fig 9

The average return in weeks following a downturn is approximately double that during weeks following a rally and the effect is statistically significant at the 3% level.

Given that result, is there any incremental “Lazarus” effect around Easter?  We test that hypothesis by comparing the average returns during the 26 post-Easter weeks which were preceded by a downturn in the index against the average return for all 1,444 weeks which followed a decline in the index.

The t-test shown in the table below confirms that conditional returns in post-Easter weeks are approximately 3x larger on average than returns for all weeks that followed a decline in the index.

Fig 8

Lazarus, it appears, is alive and well.

Happy holidays, all.

Posted in Mean Reversion, Pattern Trading, S&P500 Index, Seasonal Effects | Comments Off

Combining Momentum and Mean Reversion Strategies

The Fama-French World

For many years now the “gold standard” in factor models has been the 1996 Fama-French 3-factor model: Fig 1 Fig 5Here r is the portfolio’s expected rate of return, Rf is the risk-free return rate, and Km is the return of the market portfolio. The “three factor” β is analogous to the classical β but not equal to it, since there are now two additional factors to do some of the work. SMB stands for “Small [market capitalization] Minus Big” and HML for “High [book-to-market ratio] Minus Low”; they measure the historic excess returns of small caps over big caps and of value stocks over growth stocks. These factors are calculated with combinations of portfolios composed by ranked stocks (BtM ranking, Cap ranking) and available historical market data. The Fama–French three-factor model explains over 90% of the diversified portfolios in-sample returns, compared with the average 70% given by the standard CAPM model.

The 3-factor model can also capture the reversal of long-term returns documented by DeBondt and Thaler (1985), who noted that extreme price movements over long formation periods were followed by movements in the opposite direction. (Alpha Architect has several interesting posts on the subject, including this one).

Fama and French say the 3-factor model can account for this. Long-term losers tend to have positive HML slopes and higher future average returns. Conversely, long-term winners tend to be strong stocks that have negative slopes on HML and low future returns. Fama and French argue that DeBondt and Thaler are just loading on the HML factor.

Enter Momentum

While many anomalies disappear under  tests, shorter term momentum effects (formation periods ~1 year) appear robust. Carhart (1997) constructs his 4-factor model by using FF 3-factor model plus an additional momentum factor. He shows that his 4-factor model with MOM substantially improves the average pricing errors of the CAPM and the 3-factor model. After his work, the standard factors of asset pricing model are now commonly recognized as Value, Size and Momentum.

 Combining Momentum and Mean Reversion

In a recent post, Alpha Architect looks as some possibilities for combining momentum and mean reversion strategies.  They examine all firms above the NYSE 40th percentile for market-cap (currently around $1.8 billion) to avoid weird empirical effects associated with micro/small cap stocks. The portfolios are formed at a monthly frequency with the following 2 variables:

  1. Momentum = Total return over the past twelve months (ignoring the last month)
  2. Value = EBIT/(Total Enterprise Value)

They form the simple Value and Momentum portfolios as follows:

  1. EBIT VW = Highest decile of firms ranked on Value (EBIT/TEV). Portfolio is value-weighted.
  2. MOM VW = Highest decile of firms ranked on Momentum. Portfolio is value-weighted.
  3. Universe VW = Value-weight returns to the universe of firms.
  4. SP500 = S&P 500 Total return

The results show that the top decile of Value and Momentum outperformed the index over the past 50 years.  The Momentum strategy has stronger returns than value, on average, but much higher volatility and drawdowns. On a risk-adjusted basis they perform similarly. Fig 2   The researchers then form the following four portfolios:

  1. EBIT VW = Highest decile of firms ranked on Value (EBIT/TEV). Portfolio is value-weighted.
  2. MOM VW = Highest decile of firms ranked on Momentum. Portfolio is value-weighted.
  3. COMBO VW = Rank firms independently on both Value and Momentum.  Add the two rankings together. Select the highest decile of firms ranked on the combined rankings. Portfolio is value-weighted.
  4. 50% EBIT/ 50% MOM VW = Each month, invest 50% in the EBIT VW portfolio, and 50% in the MOM VW portfolio. Portfolio is value-weighted.

With the following results:

Fig 3 The main takeaways are:

  • The combined ranked portfolio outperforms the index over the same time period.
  • However, the combination portfolio performs worse than a 50% allocation to Value and a 50% allocation to Momentum.

A More Sophisticated Model

Yangru Wu of Rutgers has been doing interesting work in this area over the last 15 years, or more. His 2005 paper (with Ronald Balvers), Momentum and mean reversion across national equity markets, considers joint momentum and mean-reversion effects and allows for complex interactions between them. Their model is of the form Fig 4 where the excess return for country i (relative to the global equity portfolio) is represented by a combination of mean-reversion and autoregressive (momentum) terms. Balvers and Wu  find that combination momentum-contrarian strategies, used to select from among 18 developed equity markets at a monthly frequency, outperform both pure momentum and pure mean-reversion strategies. The results continue to hold after corrections for factor sensitivities and transaction costs. The researchers confirm that momentum and mean reversion occur in the same assets. So in establishing the strength and duration of the momentum and mean reversion effects it becomes important to control for each factor’s effect on the other. The momentum and mean reversion effects exhibit a strong negative correlation of 35%. Accordingly, controlling for momentum accelerates the mean reversion process, and controlling for mean reversion may extend the momentum effect.

 Momentum, Mean Reversion and Volatility

The presence of  strong momentum and mean reversion in volatility processes provides a rationale for the kind of volatility strategy that we trade at Systematic Strategies.  One  sophisticated model is the Range Based EGARCH model of  Alizadeh, Brandt, and Diebold (2002) .  The model posits a two-factor volatility process in which a short term, transient volatility process mean-reverts to a stochastic long term mean process, which may exhibit momentum, or long memory effects  (details here).

In our volatility strategy we model mean reversion and momentum effects derived from the level of short and long term volatility-of-volatility, as well as the forward volatility curve. These are applied to volatility ETFs, including levered ETF products, where convexity effects are also important.  Mean reversion is a well understood phenomenon in volatility, as, too, is the yield roll in volatility futures (which also impacts ETF products like VXX and XIV).

Momentum effects are perhaps less well researched in this context, but our research shows them to be extremely important.  By way of illustration, in the chart below I have isolated the (gross) returns generated by one of the momentum factors in our model.

Fig 6

 

Posted in Factor Models, Mean Reversion, Momentum, VIX Index, Volatility Modeling | Comments Off

Developing Long/Short ETF Strategies

Recently I have been working on the problem of how to construct large portfolios of cointegrated securities.  My focus has been on ETFs rather that stocks, although in principle the methodology applies equally well to either, of course.

My preference for ETFs is due primarily to the fact that  it is easier to achieve a wide diversification in the portfolio with a more limited number of securities: trading just a handful of ETFs one can easily gain exposure, not

Fig 3

 only to the US equity market, but also international equity markets, currencies, real estate, metals and commodities. Survivorship bias, shorting restrictions  and security-specific risk are also less of an issue with ETFs than with stocks (although these problems are not too difficult to handle).

On the downside, with few exceptions ETFs tend to have much shorter histories than equities or commodities.  One also has to pay close attention to the issue of liquidity. That said, I managed to assemble a universe of 85 ETF products with histories from 2006 that have sufficient liquidity collectively to easily absorb an investment of several hundreds of  millions of dollars, at minimum.

The Cardinality Problem

The basic methodology for constructing a long/short portfolio using cointegration is covered in an earlier post.   But problems arise when trying to extend the universe of underlying securities.  There are two challenges that need to be overcome.

Magic Cube.112

The first issue is that, other than the simple regression approach, more advanced techniques such as the Johansen test are unable to handle data sets comprising more than about a dozen securities. The second issue is that the number of possible combinations of cointegrated securities quickly becomes unmanageable as the size of the universe grows.  In this case, even taking a subset of just six securities from the ETF universe gives rise to a total of over 437 million possible combinations (85! / (79! * 6!).  An exhaustive test of all the possible combinations of a larger portfolio of, say, 20 ETFs, would entail examining around 1.4E+19 possibilities.

Given the scale of the computational problem, how to proceed? One approach to addressing the cardinality issue is sparse canonical correlation analysis, as described in Identifying Small Mean Reverting Portfolios,  d’Aspremont (2008). The essence of the idea is something like this. Suppose you find that, in a smaller, computable universe consisting of just two securities, a portfolio comprising, say, SPY and QQQ was  found to be cointegrated.  Then, when extending consideration to portfolios of three securities, instead of examining every possible combination, you might instead restrict your search to only those portfolios which contain SPY and QQQ. Having fixed the first two selections, you are left with only 83 possible combinations of three securities to consider.  This process is repeated as you move from portfolios comprising 3 securities to 4, 5, 6, … etc.

Other approaches to the cardinality problem are  possible.  In their 2014 paper Sparse, mean reverting portfolio selection using simulated annealing,  the Hungarian researchers Norbert Fogarasi and Janos Levendovszky consider a new optimization approach based on simulated annealing.  I have developed my own, hybrid approach to portfolio construction that makes use of similar analytical methodologies. Does it work?

A Cointegrated Long/Short ETF Basket

Below are summarized the out-of-sample results for a portfolio comprising 21 cointegrated ETFs over the period from 2010 to 2015.  The basket has broad exposure (long and short) to US and international equities, real estate, currencies and interest rates, as well as exposure in banking, oil and gas and other  specific sectors.

The portfolio was constructed using daily data from 2006 – 2009, and cointegration vectors were re-computed annually using data up to the end of the prior year.  I followed my usual practice of using daily data comprising “closing” prices around 12pm, i.e. in the middle of the trading session, in preference to prices at the 4pm market close.  Although liquidity at that time is often lower than at the close, volatility also tends to be muted and one has a period of perhaps as much at two hours to try to achieve the arrival price. I find this to be a more reliable assumption that the usual alternative.

Fig 2   Fig 1 The risk-adjusted performance of the strategy is consistently outstanding throughout the out-of-sample period from 2010.  After a slowdown in 2014, strategy performance in the first quarter of 2015 has again accelerated to the level achieved in earlier years (i.e. with a Sharpe ratio above 4).

Another useful test procedure is to compare the strategy performance with that of a portfolio constructed using standard mean-variance optimization (using the same ETF universe, of course).  The test indicates that a portfolio constructed using the traditional Markowitz approach produces a similar annual return, but with 2.5x the annual volatility (i.e. a Sharpe ratio of only 1.6).  What is impressive about this result is that the comparison one is making is between the out-of-sample performance of the strategy vs. the in-sample performance of a portfolio constructed using all of the available data.

Having demonstrated the validity of the methodology,  at least to my own satisfaction, the next step is to deploy the strategy and test it in a live environment.  This is now under way, using execution algos that are designed to minimize the implementation shortfall (i.e to minimize any difference between the theoretical and live performance of the strategy).  So far the implementation appears to be working very well.

Once a track record has been built and audited, the really hard work begins:  raising investment capital!

Posted in Cointegration, ETFs, Johansen, Long/Short, Portfolio Management, Statistical Arbitrage | Comments Off

Algorithmic Trading

MOVING FROM RESEARCH TO TRADING

I have written recently about the comparative advantages of different programming languages in the context of research and trading (see here).  My sense of it is that there is no single “ideal” programming language – the best strategy is to pick an appropriate tool for the job and there are usually several reasonable choices one could make.

If you are engaged in econometrics research, you might choose a package like RATS, Eviews, Gauss, or Prof. James Davidson’s excellent and inexpensive TSM, which I have used for many years and can recommend highly. For a latency-sensitive high frequency trading application, you will probably want to use something like C++, or possibly a 3rd party algo system like Apama or Tethys. But for algorithmic trading systems of intermediate frequency the choice appears almost unlimited.

The problem with retail traAlgoTradingding tools like TradeStation, Multicharts, or Amibroker, is that they are designed primarily for single-asset strategies.  That may be ok for futures trading,where more often than not the focus is on a single underlying, but in equities the opposite is true. Using one of these products to develop and implement a pairs trading strategy is a stretch.   As for portfolio analytics – forget it.

This is where more general, high level languages like R, Matlab or Mathematica come in:  their greater power and flexibility is handling large, multivariate data sets makes it much more straightforward to develop portfolio strategies. And they can often bridge the gap between R&D and implementation quite easily:  code that was used in the research stage can often be quickly re-tooled to work in a production version of the system.  As for production systems, there is now a significant cottage industry of traders who use Matlab in algo trading.  R has a similar following (see here).

In addition to parallelizing the code (for use with the Parallel Computing Toolbox) to speed up the research phase, you might also want to implement a hybrid system by re-coding the slower routines in C++, to create a mex file (for details see here). Matlab’s Profiler is a useful tool for identifying code bottlenecks.  In a recent piece of research in which I was evaluating over 30,000,000 cointegrated portfolios, I discovered to my surprise that the main code bottleneck was the multiple calls to Matlab’s std function, a problem easily fixed with a few lines of C++ code.  The resulting hybrid program executed at more than twice the speed – important when your run time might be several hours, or even days.

HOOKING UP THE EXECUTION PLATFORM

Matlab AlgoThe main challenge for developers using generic tools like Mathematica, Matlab or R is the implementation stage of the project. Providing connectivity to brokerage/execution platforms never seemed high on the list of priorities for Wolfram or Mathworks and things are similarly hit or miss with R.

Belatedly, Mathematica now offers a link to Bloomberg via its Finance Platform.  Matlab, meanwhile, offers a Trading Toolbox, which supposedly offers connectivity , not only to Bloomberg, but also Interactive Brokers and Trading Technologies, amongst other platforms.  Unfortunately, the toolbox interface to IB appears to rely on outdated 1990s ActiveX technology, which is flakey at best.  In tests, I was unable to make progress past the ‘not connected’ error message.

At that point I turned to Yair Altman’s  IB-Matlab product.  Happily, this uses IB’s Java api, which is a great deal more robust than the ActiveX platform.  It’s been some time since I last used IB-Matlab and was pleased to see that Yair has been very busy over the intervening period, building the capabilities of the system and providing very comprehensive documentation for it.  With Yair’s help, it took me no time at all to get up and running and within a day or two the system was executing orders flawlessly in IB’s TWS.  The relatively few snags I ran into were almost all due to IB’s extremely terse error messaging, which often gives almost no clue as to what the issue might be.  Fortunately, Yair is very generous with his time in providing support to his users and his responses to me questions were fast and detailed.

EXECUTION ALGOS

AD AlgoWith intermediate  systems trading at frequencies of, say, 5-minutes to daily, one has a choice to make as regards execution.  Given that the strategy is not very latency sensitive, it is certainly conceivable to develop one’s own execution algos in Matlab.  However, platforms like TWS are equipped with native algos, not only from IB, but also other providers like Credit Suisse and Jeffries.

Actually, I have found several of IB’s own algos such as Scaletrader and Accumulate/Distribute to be very effective. Certainly IB seems very proud of them – IB CEO Thomas Peterffy has patented at least one of them. Accumulate/Distribute, for instance, is quite sophisticated, allowing the user to randomize and slice the size and interval between individual orders, use passive or aggressive order types, and pause execution on a news alert, or when the price falls below a moving average, or outside a specified range.

There is much to be said for using algos native to the execution platform rather than reinventing the wheel, providing the cost is reasonable. So, while it is perfectly feasible to build execution algos in Matlab, it typically isn’t necessary – in most cases standard algos will suffice.

There are exceptions, of course.  IB doesn’t offer the  kind of basket-trading capabilities REDIthat are available in advanced algo platforms like Tethys or RediPlus.  In those systems, for example, you can set the level of long/short imbalance in the portfolio that you are willing to tolerate and the algo will speed up or slow down execution of trades in individual components of the basket to maintain the dollar imbalance within that tolerance.  You can also manage the sector risk dynamically during execution.

Those kind of advanced capabilities don’t come cheap and you wont find them at IB, or any other retail platform. If you need that kind of functionality, for example, because you are trading a long/short equity portfolio within a universe of 200-300 names, your best option is probably to switch to a different execution platform.  Otherwise you will need to code a custom algo in your language of choice.

For many quantitative strategies, (at least the low frequency ones) IB’s standard algos are often good enough.  The Accumulate/Distribute algo, for instance, will show a visual representation of the progress of the execution of individuals legs of a pairs trade, and it is easy enough to identify a potential imbalance and adjust the algo parameters in real time. If you are only trading pairs, or small portfolios of cointegrated securities, it probably isn’t worthwhile to develop the sophisticated logic that would be required to handle the adjustment of the execution of individual legs of a trade in a fully automated way.  A large portfolio would be a different matter, however.

MATLAB EXAMPLE

MatlabI thought it might be instructive to take a look at how you might implement the execution of a strategy in Matlab, using IB algos. In the Matlab code fragment below, the (2 x nTickers) array tradeActions contains, in the first row, the action we wish to take (1 = BUY, -1 = SELL, -2 = SELL SHORT) and in the second row the (absolute value of) the number of shares we wish to trade for tickers i =1:nTickers. We break each order up into hundred lots and odd lots, routing the former via IB’s Accumulate/Distribute algo and the latter as passive REL orders (note that A/D  will typically randomize the timing of each sub-order, while REL orders are posted directly into the market). The Matlab function AccumulateDistribute implements the most important features of IB’s A/D algo, including random size and time slicing of the order.  Orders are submitted as passive REL orders with zero offset (so they will sit on the current bid or ask) – obviously you would typically want to consider allowing some non-zero offset for less liquid securities.  It is not hard to envisage how one might further enhance the algo to monitor the progress of the execution and speed up or slow down certain orders accordingly.

A couple of IB api “gotchas” to be aware of:

(i) IB requires unique and monotonically increasing orderIds for each order. One way to do this, suggested by Yair, is to use orderId = round((now-735000)*3e5);  This fails when you are submitting a number of orders sequentially at high speed (say in a for loop), where the time increments are sub-second, so you need to pass the orderID back and force a minimal increment, as I have in the code below.

(ii) It is very important to specify the primary exchange of each security:  securities with identical tickers can be found trading on different exchanges.  Failing to specify the primary exchange in such a case will result in IB rejecting the order with a typically cryptic api message.

Continue reading

Posted in Algorithmic Trading, Interactive Brokers, Matlab, Time Series Modeling, TradeStation | Comments Off

Volatility ETF Strategy Feb 2015: +3.13%

HIGHLIGHTS

  • CAGR over 40%
  • Sharpe ratio in excess  of 3
  • Max drawdown -13.40%
  • Liquid, exchange-traded ETF assets
  • Fully automated, algorithmic execution
  • Monthly portfolio turnover
  • Managed accounts with daily MTM
  • Minimum investment $250,000
  • Fee structure 2%/20%

 

VALUE OF $1000
STRATEGY DESCRIPTION

The Systematic Strategies Volatility ETF  strategy uses mathematical models to quantify the relative value of ETF products based on the CBOE S&P500 Volatility Index (VIX) and create a positive-alpha long/short volatility portfolio. The strategy is designed to perform robustly during extreme market conditions, by utilizing the positive convexity of the underlying ETF assets. It does not rely on volatility term structure (“carry”), or statistical correlations, but generates a return derived from the ETF pricing methodology.  The net volatility exposure of the portfolio may be long, short or neutral, according to market conditions, but at all times includes an underlying volatility hedge. Portfolio holdings are adjusted daily using execution algorithms that minimize market impact to achieve the best available market prices.

Ann Returns

RISK CONTROL

Our portfolio is not dependent on statistical correlations and is always hedged. We never invest in illiquid securities. We operate hard exposure limits and caps on volume participation.

Sharpe

OPERATIONS

We operate fully redundant dual servers operating an algorithmic execution platform designed to minimize market impact and slippage.  The strategy is not latency sensitive.

 

MONTHLY RETURNS

Monthly Returns

 

(Click to Enlarge)

PERFORMANCE STATISTICS

PERFORMANCE STATS

(Click to Enlarge)

 

 

Posted in VIX Index, Volatility ETF Strategy | Comments Off

Successful Statistical Arbitrage

communication

I tend not to get involved in Q&A with readers of my blog, or with investors.  I am at a point in my life where I spend my time mostly doing what I want to do, rather than what other people would like me to do.  And since I enjoy doing research and trading, I try to maximize the amount of time I spend on those activities.

As a business strategy, I wouldn’t necessarily recommend this approach.  It’s just something I evolved while learning to play chess: since I had no-one to teach me, I had to learn everything for myself and this involved studying for many, many hours alone.

By contrast, several of the best money managers are also excellent communicators – take Roy Niederhoffer, or Ernie Chan, for example. Having regular, informed communication with your investors is, as smarter managers have realized, a means of building trust and investor loyalty – important factors that come into play during periods when your strategy is underperforming. Not only that, but since communication is two-way, an analyst/manager can learn much from his exchanges with his clients.  Knowing how others perceive you – and your competitors – for example, is very useful information.  So, too, is information about your competitors’ research ideas, investment strategies and fund performance, which can often be gleaned from discussions with investors.  There are plenty of reasons to prefer a policy of regular, open communication.

As a case in point, I was surprised to learn from  comments on another research blog that readers drew the conclusion from my previous posts that pursuing the cointegration or Kalman Filter approach to statistical arbitrage was a waste of time.  Apparently, my remark to the effect that researchers often failed to pay attention to the net PnL per share in evaluating stat. arb. trading strategies was taken by some to mean that any apparent profitability would always be subsumed within the bid-offer spread.  That was not my intention.  What I intended to convey was that in some instances, this would be the case  - some, but not all.

To illustrate the point, below are the out-of-sample results from a research study applying the Kalman Filter approach for four equity pairs using 5-minute data.  For competitive reasons I am unable to identify the specific stocks in each pair, which result from an exhaustive analysis of over 30,000 pairs, but I can say that they are liquid large-cap equities traded in large volume on the US exchanges.  The performance numbers are net of transaction costs and are based on the assumption of a 5-minute delay in execution: meaning, a trading signal received at time t is assumed to be executed at time t+5 minutes.  This allows sufficient time to leg into each trade passively, in most cases avoiding the bid-offer spread.  The net PnL per share is above 1.5c per share for each pair.

Fig 0 While the performance of none of the pairs is spectacular, a combined portfolio has quite attractive characteristics, which include 81% winning months since Jan 2012, a CAGR of over 27% and Information Ratio of 2.29, measured on monthly returns (2.74 based on daily returns).

Fig 2

Fig 3

Finally, I am currently implementing trading of a number of stock portfolios based on static cointegration relationships that have out-of-sample information ratios of between 3 and 4, using daily data.

 

 

Posted in Cointegration, Kalman Filter, Pairs Trading, Statistical Arbitrage | Comments Off

A Comparison of Programming Languages

Towards the end of last year I wrote a post (see here) about the advent of modern programming languages, including the JIT compiled Julia and visual programming language ADL from Trading Technologies.  My conclusion (based on a not very scientific sample) was that we appear to be at the tipping point, where the speed of newer, high level languages  languages is approaching that of the fastest compiled languages like C/C++.

Now comes a formal academic study of the topic in A Comparison of Programming Languages in Economics, Aruoba and Fernandez-Villaverde, 2014.  Using the neoclassical growth model, the authors conduct a benchmark test in C++11, Fortran 2008, Java, Julia, Python, Matlab, Mathematica, and R, implementing the same algorithm, value function
iteration with grid search, in each of the languages. They report the execution times of the codes in a Mac and in a Windows computer and briefly comment on the strengths and weaknesses of each language.

periodic-table-of-programming-languages-1-728The conclusions from the study mirror my own thoughts on the subject very closely. The authors find that:

  1. C++ and Fortran are still considerably faster than any other alternative, although one needs to be careful with the choice of compiler.
  2. C++ compilers have advanced enough that, contrary to the situation in the 1990s and some folk wisdom, C++ code runs slightly faster (5-7 percent) than Fortran code.
  3. Julia delivers outstanding performance. Execution speed is only between 2.64 and 2.70 times slower than the execution speed of the best C++ compiler.
  4. Baseline Python was slow. Using the Pypy implementation, it runs around 44 times slower than in C++. Using the default CPython interpreter, the code runs between 155 and 269 times slower than in C++.
  5. Matlab is between 9 to 11 times slower than the best C++ executable.
  6. R runs between 475 to 491 times slower than C++. If the code is compiled, the code is between 243 to 282 times slower.
  7. Hybrid programming and special approaches can deliver considerable speed ups. For example, when combined with Mex files, Matlab is only 1.24 to 1.64 times slower than C++ and when combined with Rcpp, R is between 3.66 and 5.41 times slower. Similar numbers hold for Numba (a just-in-time compiler for Python that uses decorators) and Cython (a static compiler for writing C extensions for Python) in the Python ecosystem.
  8. Mathematica is only about three times slower than C++, but only after a considerable rewriting of the code to take advantage of the peculiarities of the language. The baseline version of the algorithm in Mathematica is considerably slower.

C++ still represents the benchmark for speed, but not by much.  It is barely faster than the old stalwart, Fortran, and only 1.5 – 3 times faster than up-and-coming rivals amongst the higher level languages (especially when you allow for hybrid programming to speed up the slowest algorithms).

So, as regards developing financial models and trading systems, my questions are (as before):

  • Why would anyone prefer Python, given that there is a much faster, free alternative in Julia, which is just as easy a language to program in?
  • What justification is there for preferring R to Matlab, other than cost?
  • Why does anyone bother with Java?  If speed is the critical issue, there are faster alternatives.  If you like the relative simplicity of the syntax, Julia is cleaner, simpler and just as fast in execution.

When you reach a point where a high level language like Matlab is only around 1.5x – 2x slower than C++, you really have to question whether the latter is an appropriate choice.  Yes, of course, in mission-critical applications where you need access to the hardware layer for speed purposes, C++ is the way to go.  But for so many applications, that just isn’t the case.

What matters, far, far more, are the months of costly and laborious programming effort that is often required to reproduce basic functionality that is already embedded in higher level languages like Matlab or Mathematica.  Not only that, but the end result of a C++ /Java development effort is likely to be notoriously inflexible by comparison.  That’s a huge drawback.  Rarely, if ever, does a piece of research translate flawlessly into production – it requires one to iterate towards a final solution, often making significant changes to the design of the system in the light of practical experience.

If I had to guess, based on my experience, I would say that 80% or more of development tasks in quantitative research and trading would produce a superior result if preference was given to using a higher level language for the initial development.  When the system is sufficiently stable to put into production, you simply create a hybrid application by recoding any mission-critical components for which speed is an issue in C++.

Finally, where does that leave my beloved Mathematica?  To be fair, while you don’t have the joys of strong typing to contend with, Mathematica’s syntax is just as demanding and uncompromising as C++ – a missed comma or incorrectly placed bracket is just as critical.  But, the point is, while in C++ the syntactical rigor is just annoying, in Mathematica it’s worth putting up with because the productivity is so much greater.  A competent programmer can produce, in a single line of Mathematica code, a program that would require hundreds, if not thousands of lines of C++ code to accomplish.  Sure, he might get the syntax wrong at first:  but it’s only a single line of code and the interactive gui interface makes debugging very simple.


mathematica fn

That said, while Mathematica can be very tedious to use for procedural programming, it excels in three areas:

1.  Symbolic programming. Anything involving mathematical symbols and equations – Mathematica is #1

2.  User interface.  In Mathematica, it is trivial to build a  sophisticated, dynamic gui in no time at all, again, often in 1-2 lines of code

3.  Functional programming. Anything that can be thought of as a function, Mathematica handles extremely well.  We are not talking about finding a square root here:  I mean extremely complex functions that, again, might take hundreds of lines of code in another language.

It is also worth pointing out that Mathematica comes supplied with functionality that Matlab provides only through numerous, costly add-on packages.

CONCLUSION
Before I allow a development team to start mindlessly coding up a system in Java or C++, I want to hear their reasons why they aren’t going to do it 10x faster in another, higher level language.  “We always use C++/Java for production” is not a reason.  Specifically, which parts of the system require the additional 1.5x speed-up, and why can’t they be coded as dlls (Matlab mex functions)?

Finally, on a cost-benefit basis, ask yourself how much  you might benefit if the months and tens (or hundreds) of thousands of dollars wasted on developing in C++ were instead spent on researching and developing new trading ideas.

 

Posted in Algo Design Language, Algorithmic Trading, Julia, Mathematica, Matlab, Programming | Comments Off

ETF Pairs Trading with the Kalman Filter

I was asked by a reader if I could illustrate the application of the Kalman Filter technique described in my previous post with an example. Let’s take the ETF pair AGG IEF, using daily data from Jan 2006 to Feb 2015 to estimate the model.  As you can see from the chart in Fig. 1, the pair have been highly correlated over the last several years.

Fig 1Fig 1.  AGG and IEF Daily Prices 2006-2015

We now estimate the beta-relationship between the ETF pair with the Kalman Filter, using the Matlab code given below, and plot the estimated vs actual prices of the first ETF, AGG in Fig 2.  There are one or two outliers that you might want to take a look at, but mostly the fit looks very good. Fig 2

 Fig 2 – Actual vs Fitted Prices of AGG

Now lets take a look at Kalman Filter estimates of beta.  As you can see in Fig 3, it wanders around a lot!  Very difficult to handle using some kind of static beta estimate. Fig 3

Fig 3 – Kalman Filter Beta Estimates

  Finally, we compute the raw and standardized alphas, being the differences between the observed and fitted prices , i.e. Alpha(t) = AGG(t) – b(t)* IEF(t) and kfAlpha(t) = (Alpha(t) – mean(Alpha(t)) / std(Alpha(t)   I have plotted the kfAlpha estimates over the last year in Fig 4.   Fig 4

Fig 4 – Standardized Alpha Estimates

  The last step is to decide how to trade this relationship.  You might, for example, trade the portfolio in proportion to the standardized deviation (i.e. the  size of kfAlpha(t)).  Alternatively, you might set a threshold level, say +/- 1 Sd, and trade the portfolio when  kfAlpha(t) exceeds this the threshold.   In the Matlab code below I use the particle swarm method  to maximize the likelihood.  I have found this to be more reliable than other methods.

Continue reading

Posted in Cointegration, Matlab, Statistical Arbitrage | Comments Off