I’ve filed for a divorce

I think, and hope, that we can still be friends, but my days of being married to high-probability discretionary trading are over. This separation is amicable and mutually agreed upon. We had a good run, but things ran into trouble last year when the market gave us something we’ve never seen before. Our relationship has never been the same since. The screaming, the accusations … oh, it got ugly. I thought maybe we could work it out this year, but this latest 3-week rally has put a permanent end to our relationship.

High-probability discretionary trading is the mistress or spouse of many traders. She is seductive, beautiful and addictive (lady traders, feel free to visualize her as a him, if you’d like). She promises, and oftentimes delivers, low risk for high reward. There is nothing so alluring as visualizing a setup that has a defined risk (the place where the trade no longer works) and its potential reward, which can be five times larger or more than the risk.

High-probability discretionary trading’s father is Poker. He taught her growing up that it’s all about betting based on the size of the pot and your individual odds. He taught her that you don’t need to play every hand, and sometimes it even makes sense to fold a good hand to get an idea of how others are playing.

Where my relationship with high-probability discretionary trading went bad was in my expecting something from her that she could never deliver. Probabilities that were actually based on reality. You see, her probabilities are a little problematic, and she will be the first to admit it. I know there are many successful traders out there who are still very happily married to high-probability trading, and maybe things would have been different with me if I had just listened more. Who knows. It is what it is.

Even though I’ve had relationships with fundamental trading, intuitive trading and emotional trading, I found high-probability trading worth getting married to. Now that I’m moving on, I’d like to introduce you to my future spouse: system trading.

I know she’s kinda boring compared to the others, but I love her for that. She’s a little frumpy, awkward in public and can’t play sports, but she’s a bit more real to me the than the flashy others.

System trading takes a notion or idea and puts it through a scientific process of validation. It includes backtesting, optimization and walk forwards. It involves compiling a statistical profile for a trade system and the market to be traded. When she fires off a signal, you take it. No discretion. All the work you’ve done on understanding the system depends on you taking every signal. It’s a very committed relationship that way.

If all you like to talk about is the latest reality show and what your neighbor is selling his house for, then you probably won’t like system trading. If you like Wittgenstein, Tolstoy and Gabriel Garcia Marquez, then you may have found your mate. We talk about things like intermarket correlation, genetic algorithms, neural networks, particle swarm optimization and fuzzy logic over breakfast. It’s fun, but then that’s me.

Advertisements

13 Responses to “I’ve filed for a divorce”

  1. Anonymous Says:

    MilkTrader, I am a system trader and have been for 9 years. My system is "boutique" and have "upgraded" my neural network as the time and technology has changed. The system has beaten the S&P consistently. I basically build trading models (stocks, ETF's, etc) off of over 200 different indices. That is I look for correlations between how a model reacts to various indices over short periods of time.

    Thanks for your blog. I am curious what is your system based off of?

    Sincerely,

    Anotherrecord

  2. Milk Trader Says:

    I will be developing a range of systems starting with trend-following and breakout systems, and then moving to intermarket correlations.

    How do you know when a correlation gets broken, or when there is a decoupling? Do you use a certain range (like .30) in Pearson correlation? Also, are you familiar with predictive correlation. I haven't used it yet, but it's in the fridge defrosting for dinner some time soon.

  3. gappy3000 Says:

    Are you sure you know what you are doing? Most "systematic" traders don't know what a Bonferroni correction is, or why leave-one-out cross validation should not be performed on large datasets, or why fitting autoregressive time series as a linear regression on lagged variables is plain wrong. From your comment above, I understand you are not trained in statistics. I don't mean to be adversarial. Just a word of caution because I have seen people blow up, and they had PhDs in CS from top schools.

  4. Milk Trader Says:

    Are you referring to my mention of predictive correlation?

    Bonferroni correction: will research.
    Leave-one-out cross: will research.
    Linear regression issue: agree.

    In any case, a statistically invalid or naive system will not make it past a walk forward on out-of-sample data.

    From you comment above, I understand you're not trained in social interaction.

  5. gappy3000 Says:

    "In any case, a statistically invalid or naive system will not make it past a walk forward on out-of-sample data." That is also not strictly true. It depends very much on how you design your out-of-sample validation procedure, and it's easy to choose incorrect metrics, and/or poor tests on these metrics. Just to get an idea, check out Lo's paper on the estimation of Sharpe Ratio. In any case, we agree that a "systematic" approach is the one of two viable ones for retail investors, the other being solid fundamental analysis.

    A very solid quantitative training is a necessary prerequisite. Having said that, even people with very quantitative backgrounds and good pedigrees make big methodological mistakes. Only you can assess your skills, and I am not implying you're not up to the task.

    And, no, I don't have a PhD in social interaction. But believe it or not, I am trying to be helpful. In fact, didn't I send you a link on Twitter a while back on leveraged ETF decay (from Barclays' researchers)?

  6. Milk Trader Says:

    Yes you did and your comments are always welcome.

    I will blog on the methodology I use to backtest, optimize and walk-forward a system. My first system is called Bumblebee and it's logic and code have already been blogged. I have done backtesting, but I haven't blogged about the results yet because I'm giving it a very critical and skeptical eye. As I lay out the specifics of the methodology, please don't hesitate to comment.

    My latest statistical reading is Applied Regression Analysis (Dielman) and I realize it's very basic. If you can recommend a market-relevant statistical tome(s), it would be appreciated.

    I'm also starting to play around with R and have purchased a statistical book on how to use the program.

  7. gappy3000 Says:

    I am a big R advocate, even though I need something more scalable, so am looking at functional languages.

    One good book I recently purchased and am reading is "Statistical Models and Methods for Financial Markets" by Tze Leung Lai and Haipeng Xing. It's short, but has links to monographs. There are no implementation details, but those are too platform-specific.

  8. Anotherrecord Says:

    MilkTrader,

    I use Momentum Swing trades and correlate stock price change to a particular index I have chosen (can even trade off of VIX, VXN, etc. – I look at about 175 indices). I look for correlations of index to stock change 0f 0.95 over a three month (previous 3 month) before I trade the next month. I have chosen to target ETF's since the gaps in individual stocks are just to risky. I only do end of day trading and will make my trades if the "system" tells me to. I have designed an autoexecution system that I upload my signals into every evening for the next day of trading. It works beautifuly in the respect that I have linked my system into Interactive Brokers through and API and set my trades to execute 20 seconds before the market close (unless my stops or limits are hit first). I trade 10 models (stocks) per month and decide the ration of short to long ETF's based on the Bull/Bear ratio from Timers Digest.

    Thanks,

    Anotherrecord

  9. Milk Trader Says:

    You must have a large amount of data on the system. Does it have a smooth equity curve?

    The auto-execution is an interesting technology. Did you program the execution API yourself and what language did you use?

    As far as using VIX, VXN, etc, I've always thought it interesting to compare option volatility with actual volatility of the underlying. Does current VIX predict SPX based on some spread threshold? In other words, if VIX is 10% higher than actual price volatility, is that a predictor of future price action? I would calculate price volatility as the standard deviation of each day's range over a period of 30 trade days, btw, though I'm not sure that would match up with VIX, which is 30 calendar days. Maybe 21 trade days?

  10. Anonymous Says:

    Funny, I am doing the opposite. I found myself shutting down my system before the recent rollover, and again before our very unstandardly deviant rocket lift off, and ended up the better for it. Of course, I don't have the training of Max Dama, so there you go. I really think you have to look before leaping, besides having rules and measures; I believe marketsci's systems are traded similarly.. Just my .0002 cents

  11. Anonymous Says:

    Milk Trader…excellent scribe!. The best advice and statistic I could offer; discount 99.9% of any anonymous unsolicited advice regarding trader systems (I guess that would included this post as well…dog gone). Motives are never (ever) clear or altruistic. Next best suggestion, read Perry Kaufman's "New Trading Systems and Methods 4th edition"… an outstanding resource (required reading for the CMT Level II exam)for system traders. Last, number one golden bestest rule for system development: less is more. The esoterica of exotic systems and ancillary tests make for interesting conversation but usually fail when the first bullet is fired. Think LTCM and (and recently) Renaissance Technologies. The most successful systems are those that are eloquently simple. Best of luck, sounds like you'll do very well with you past discretionary experience as a guide.

  12. Charles Maley Says:

    Why are we such suckers for prediction?
    I think we love preditions because if we make predictions, and/or concoct explanations for those events we predicted wrong, then we won't feel like victims of randomness. We feel more in control. But are we more in control or just intoxicated by some illusion of control?

    http://www.viewpointsofacommoditytrader.com

  13. Milk Trader Says:

    There is a large part of the trading community that does predict market direction (up, down, sideways) and places trades based on this analysis and/or opinion. They are fundamentalists and intuitives, and need to be comfortable with being wrong (it doesn't destroy their confidence or accounts).

    Another part of the trading community claims to have no opinion of market direction, and do not place trades based on opinions. They will not work well in a system that forces them to predict which way, if any, the market is headed. They prefer a trading world of probabilities. They don't mind a losing trade because they realize no system will produce 100% wins.

    System trading is more like the second group, though fundamental economic data can be incorporated into a neural network that feeds a system.

    Illusion of control over the markets doesn't last long and that trader will soon find that not only do they lack control of the markets, they lack control over themselves.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: