A reader of this blog (hey – I’m as surprised as you are!) sent me an email recently detailing a strategy they’d developed. While the details of that strategy aren’t relevant here, they sounded good and they got me to thinking about the process of selling a trading strategy. This is an activity that I’ve spent some time on and have decided just isn’t for me.
There are a lot of difficulties with selling a trading strategy. One of them is a consequence of the foundational problem of back-testing about which I first started posting on this blog. For any given period of time (that has already elapsed!), it’s not difficult to generate a good number of pretty impressive strategies. All you have to do is try a good enough number of random strategies and some of them will prove to be too good to be true.
Presumably, any credible person who might be listening to your pitch will be at least intuitively aware of this fact and will thus be highly suspicious of any back-tested results you might present. For this reason, it’s impossible to sell a strategy on the basis of back-tested results. Only auditable, real-world returns will be considered valid by any serious person. Of course, you might find someone who’s less particular, but then you’re flirting with fraud rather than a legitimate sale.
So let’s say you have impressive, verifiable results. You still have to answer the question:
If this strategy is so good, why are you selling it? Why not just trade it yourself?
Read more…
hedge funds, startup, strategy development
rebranding opportunity?
A friend of mine pointed out an article he came across on his bloomberg terminal today which reminded him of a strategy I’d described to him sometime back and which we’ve been trading over the past year or so with good results.
To the great chagrin of some of my partners, I even wrote a few posts about the phenomenon underlying our strategy and its evolution as we capitalized on it. Eventually, they persuaded me to shut up already, but the outline was there for all – including Goldman! – to see.
My first post on the topic, “unsung virtues of a dynamic hedge” published June 4th of last year, was pretty coy and didn’t mention the source of alpha itself but talked about enhancing it with a dynamic hedge.
My next post on the topic, “to dream” was published July 14th of last year and laid out the exploitable discrepancy of the market’s behavior. Interestingly, the data I provided in that posting went back the same amount of time as in Goldman’s piece.
I explicitly wrote one last time about the strategy in “evolution of a strategy” wherein I detailed the process by which we’d been evolving the strategy.
Now, one of the more entertaining things about having a blog is that you get to see who is viewing your content. I’m happy to note that all of the major IBs are represented including a variety of distinct IPs within Goldman.
Now, I’m not accusing them of stealing my ideas or anything untoward like that… but I’ll admit that I am wondering how long it’s going to take them to make similar observations across markets beyond US Equities…
Read on for the Bloomberg article…
Read more…
back-testing, dereferenced, strategy development
During some recent travels, I read William Poundstone’s ramblingly entertaining Fortune’s Formula. It had been sitting on my shelf after I’d originally gotten it, perused it and offhandedly discarded it as yet another of these science-is-fun-and-full-of-wacky-characters books for the butch humanities student. My initial impression was a bit harsh as the book proved entertaining and covered a lot of ground including significant coverage of Ed Thorp and his stat arb alchemy (see here for his own papers on the topic).
One of the more compelling segments of the book relates Claude Shannon’s demon which is a nice thought-experiment / trading-strategy which illustrates the tractability of the problem of trading on a random walk market with fixed properties. I wrote the above applet to explore the impacts of applying friction and otherwise modifying the behaviors of the market and the demon.
The original demon posited a world with no friction in which the market contains one instrument which doubled or halved in value each day. Shannon’s demon looks to take advantage of this volatility by maintaining a portfolio which was rebalanced each day to ensure a 50/50 split between cash and the market. The applet implements a very simple monte-carlo test-bed for Shannon’s Demon. You can configure the demon and the marketplace along a variety of parameters, and then run many instances of the demon, each on its own self-contained random-walk market.
Although Shannon’s demon is a highly “stylized” case in the sense that it operates on a very synthetic, unrealistic and favorable formulation of a random-walk marketplace, it has spawned a great deal of interest and serious research.
Most of all, it’s a revealing illustration of the kind of reasoning one must embrace in order to address stat arb strategy development. Enjoy.
—
Updated: March 4th – made price axis logarithmic to better reveal mc paths.
books, dereferenced, monte-carlo methods, strategy development
One of my favorite tools for strategy development is the distribution of returns a strategy will generate. As I’ve discussed before (and here and here), it’s an easily quantifiable characterization of a strategy’s “underlying nature” and can be used to engineer strategies that fit appropriate markets.
Given the enduring value of return distributions, I found this morning’s post in ft.com/alphaville especially interesting. They cite a Dresdner study examining the distribution of returns for Goldman Sachs’ prop trading in 2003 and 2008. Eye opening stuff.
normal
not so much
dereferenced, performance analysis, strategy development
As I’ve written before, I’m not a particularly big fan of technical analysis or any of the many and varied charting techniques people espouse. That said, we are working with a proprietary futures trading company and some of the successful (non-algo) trading that they do involves point-and-figure charts. Although a trading algorithm doesn’t care about graphical representations, I wasn’t familiar with the technique and decided that the best way to understand it was to try to implement it, which is how I spent my Saturday evening …
The above applet re-uses the one I’d written previously in discussing simple stochastic processes. This time, it illustrates a point & figure chart below the regular line chart. Point & figure charts expose two characteristics: a “box size” (in ticks) and a “reversal” (in boxes). The applet allows you to vary both and then generate a day’s worth of random/synthetic data to view it. One of the nice features of JFreeChart is that you can easily “zoom” into a chart by dragging within the chart. I’ve disabled this in the line chart but you can try it in the p&f chart. (Note: you should right-click and “Auto-Range-Both Axes” before you generate new data or you’ll stay in the zoomed segment of the chart.)
Now that I think I understand the basics of point & figure charting, it will be interesting to see what an algo might do with it…
open-source software, strategy development, technology
Inevitably one of the first ideas people have when they start thinking about how to write a trading algorithm turns out to be among the hardest: trading the news. The problems are many and in some cases not so obvious…but the natural appeal of the idea seems universally compelling.
Just after the dot.com craze, a brilliant friend of mine (who had just sold his web consulting startup) decided to write a book. The premise was glorious. A bunch of clever college-age kids formed a startup to predict the stock market. The method they used was to constantly comb the web with ultra-sophisticated algorithms which would run across giant server farms overnight and ultimately generate tomorrow’s headlines. Based on the headlines that their system generated, they would place trades that would take advantage of these predicted events.
Sadly, my friend never went on to complete his book, so I don’t know how it all turned out. (Instead, he went on to start another successful company, this time in the field of robotics.) While he was writing it, I loved getting new drafts as they were filled with clever ideas. But the core idea of predicting headlines and then using those headlines to trade always struck me as especially cute.
For those of us without access to news-predicting algos, writing strategies based on the news is rather less straight forward, though there are a growing variety of products and services aiming to fill the gaps. Today must have been trading-the-news-day as I found a few articles on the topic in my mailbox and even received a cold call from a vendor, Need to Know News, with just such an offering. Below I’ll look at some of these offerings and consider some of the issues involved in writing trading strategies based on the news. Read more…
back-testing, market data, startup, strategy development, technology
I came across this gem of a quote in a comment on the big picture and it reminded me, somewhat circuitously, of another one of the things I view as axiomatic about algorithmic trading.
In The Alchemy of Finance, George Soros observed that one of his advantages as a trader was that while he held beliefs strongly, he was also capable of abandoning or even reversing them quickly as conditions evolved. An algorithmic trader needs something like this but more so – an automated trader is best served free of opinions entirely.
I think this is why the sunken ship quote made me think of this. While charts are an effective means of quickly communicating potentially a great deal of information to a human viewer, the cult of chart technicians and the endless supply of books, lecture series and training materials they actually make their money on might convince you that you can form your opinions based on chart patterns…
Read more…
books, strategy development
I’ve written here about exchange simulation in service of back-testing trading algorithms and briefly mentioned the difficulties of simulating the behavior of an order book. I just came across “A stochastic model for order book dynamics” by Cont, Stoikov and Talreja of Columbia and Cornell financial engineering groups. (I’ve also saved a local copy of the paper here.) While their focus isn’t on simulation for the purpose of back-testing but on probabilistic reasoning in real-time for high-frequency strategies, they illustrate a variety of models/methods for such reasoning. The equation depicted above is part of their description for reasoning about the likelihood of being able to make the spread in a stat arb strategy which places orders simultaneously at the bid and ask. It’s very technical, but interesting even if only to illustrate the kinds of tools being wielded in the service of algorithmic trading!
EMS Internals, dereferenced, strategy development
A friend recently asked me what I considered to be the “axioms” of alpha-seeking trading strategies. I think there are a few, but probably the one that seems to me most important is that the atomic element of a trading strategy should always be a portfolio as opposed to a single instrument.
In a scenario of perfect knowledge, this wouldn’t be true. If you somehow *know* with certainty that crude will go up or that Citigroup will go down, then concentrating all of your resources into a position based on that belief might be reasonable. But knowledge seldom comes in such a neat package (and will frequently be illegal to act upon when it does!).
Instead, knowledge will typically come in more conditional and less certain forms: “commodities tend to rise during periods of FUD [Fear-Uncertainty-Doubt]” or “companies who announce stadium naming rights deals tend to under-perform.” In some cases, perhaps the knowledge on which you’ll base your strategy can be quantified probabilistically.
Depending on the nature and quality of the knowledge or hypothesis that forms the basis for a given strategy, one can adapt one’s portfolio construction/optimization based on customized relationships amongst the potential portfolio constituents. But one doesn’t need to be so fancy to see the concrete benefits of our first axiom. Below I detail a simple strategy I’ve put together to explore the forces involved.
Read more…
performance analysis, portfolio management, strategy development
I’ve been trading an increasing amount at the open and close of the equity markets using market-on-open (MOO) and market-on-close (MOC) order types and have found that the quality of executions varies enormously between the two types and have spent a bit of time analyzing the differences which I share below.
The quick scoop is that MOC orders almost invariably fill at the exchange’s published closing price, while MOOs vary very substantially from the published open price. Below I quantify my findings in a bit greater depth.
Read more…
back-testing, execution quality, performance analysis, post-trade analysis, strategy development
