Thanks, that is reasonable, I suspected you would recommend that. Maybe grouping instruments into synthetic vs non-synthetic and imposing constraints might also help besides pre-filtering. I'm worried about robustness of instrument weights. I have noticed you trade some signals where the signal strength is dependant on a group of markets, like global trend. In that case, do you force the forecast weights within a group to be the same, so that positions are proportional, or you maintain some relative positioning property that is assumed in your forecast? In this case, global trend would be the tide that lifts all boats, so your positions in each market shouldbe proportional, such as "all markets trading global trend shall have a 5% forecast weight to global trend forecast".
If you're worried about robustness use equal weights. On the second point, I don't do any forcing of any kind. All the grouping means is that all equities will have the same forecast for that particular rule on a given day. But I don't force them all to have a 5% forecast weight to that rule. Why should I? The actual amount they contribute to the rules total forecast will depend on (a) the instrument weights and (b) the forecast weights. Notice I also have relative value forecasts. This also means that the net forecast across all instruments from the relative value rule might not be zero (because of the same issues). Some people get really worried about this. I don't know why. GAT
Hi GAT, You've had stellar performance since launching your strategy. I note that your launch fortuitously coincided with one of the strongest trending markets ever! What would you have been thinking if you launching in e.g., 2011/12, which were generally not great markets for trend-orient strategies? I note you have more than just trend, but this is the largest part of your portfolio. One thing I am concerned about is 3-4 years of sideways markets after launching and going straight into a large drawdown. How do you personally deal with these periods?
It's true I've been very, very lucky. Trend following sucks because you spend most of your time in a drawdown. Because I've been in this business since 2006 I've got used to that. I think there are two key things. The first is to ensure that the swings in your p&l aren't so big that they disturb you. I think the biggest losing day I have had is about a £25K loss. Now don't get me wrong, that wasn't pleasant, but I could live with it. I know for a fact that I couldn't cope with a £50K loss. For starters that limits my risk target. I couldn't increase it anymore. It also limits the amount I have in my trading account. My trading account should have all my household assets in it because it has a higher sharpe than long only investments. So if for example it had half my household assets in, and I transferred everything in, then it would have cash risk that was twice as much and I'd hate that. [I could also avoid looking at my p&l daily, but I like to know whether I'm within my risk limits. This means for example that I tolerate a larger loss on my long only portfolio just because I don't know its happening - it's just too much of a pain to work out long only p&l that I only bother once a year] Interestingly I don't tend to think about my drawdown in cash terms, but as a percentage. Again this limits my risk target. If I believe my backtest I should expect to see a 35% drawdown at worse. Again so far I've been lucky and not got above 20%. The second thing is to partition your trading account, and ensure you don't have to live off it month by month, and treat it as a hedge fund. Let me explain. My maximum risk is 400K. If I lose money below that I degear. But if I make money above that I don't increase my capital at risk (more here). This means effectively my trading account acts like a hedge fund with a 100% performance fee. Any money made above the high water mark gets paid to the investor. Again in the long run this is sub optimal, since I would make more money if I kept all my profits in there. But it has some nice properties. It means for example I can't lose more than 400K (barring a massive overnight gap before I could liquidate anything). All the money I've made above the HWM I've taken out of the account and it can't be lost. This also means I can think of my trading account as a very strange investment that periodically pays out dividends (money made above the HWM), but in some years might not pay out at all. So right now for example the trading account is slightly down since the start of the tax year. Unless things pick up it's going to be a year without a payout. In practice this also means I need to have enough other income to keep myself going from a cashflow perspective potentially for several years. This means if you trade for a living you eithier need to be in a position where you don't need the income from your trading account, or where you have enough money in cash to cover an N year drawdown, or a combination of the two (for example if your other income will cover half your outgoings then you only N/2 in cash). Hope that makes sense. GAT
Hi GAT, New to this thread but have caught up on all 53 pages, in addition to reading your fantastic book and following your very informative blog. Thank you for all your intellectual generosity. I work within the fund industry and it seems bizarre to me that someone hasn't tried to lift you out of 'retirement' yet... Two questions for you if I may: 1. Once you have a systematic futures strategy which you have implemented with healthy Sharpe over the first 12 months (talking >1.5 net of costs), has demonstrated good diversification benefit, has gone through up and down periods and demonstrated robustness over its initial life - what are the next steps that a trader should take wrt this strategy? I ask this because it seems to be easy at that stage to spend time in one of the 2 extremes: either (1) spending all your time trying to remove every single historic drawdown from what is clearly already a pretty good strategy and add 0.0001 of Sharpe by overfitting, OR (") leaving the strategy complete alone and resting personally in a zone of complacency. How should I be spending my time in this phase? 2. I notice there is some oblique discussion of systematic option strategies on this thread (including the very useful link to Palaro's research - thanks for that). It seems from this and also from browsing various academic papers online that the bulk of research surrounding systematic vol trader is centered mostly around systematic short selling strategies. I am wondering if you had any links/book recommendations/guidance on building more fully rounded systematic vol strategies - i.e. ones which not only build on systematic short selling, but also look at stuff like momentum in implied vols, trend in implied/realized spreads, trend in vol term structure, trend in skew etc as signals when to buy/sell implied vol or inform cross-sectional vol strategies? I have not come across anything related to this online or in systematic trading books and any guidance would be appreciated as I look to build something like this out. Thanks in advance GT.
Oh believe me they've tried. But I'm resisting manfully. Doing nothing is best (I think you knew that from the way you phrased the question - had we been in court you would have been in trouble for leading the witness). Having said that if you are bored it's might be worth looking at something from a completely different trading style to run alongside what you already have. Which brings me on to: I am afraid I don't know of any book or website that addresses this. Most (good) options traders seem to be discretionary in nature. GAT
Thanks GAT. Doing nothing is harder than backfitting in extremis but as you say, it's probably the right thing to do. Interesting what you say on the options strats. Speaking to various market participants over the past 6 months, it seems there is definitely an emerging sector of the systematic sector that is looking at/developing systematic vol strats (diversified long/short) based on various vol signals. I will share any resources on this topic if I find them and if you are able to do so too (should you come across anything of this nature) that would be much appreciated. BTW - if you don't mind me asking - why the resistance to getting back into the industry? I'm sure there are plenty of places that would multiple your capital significantly. Even if payout was in the range of 10-20% (and not 73% after tax as you have currently), at some capital multiplier, this trade off becomes worth it (particularly with the positive carry of a salary). Or you just don't fancy reporting to someone?
Oh sure it's definitely an emerging trend and plenty of people are doing it (UBS O'Conner for decades, AHL since 2007). But people aren't writing about it. Office politics play a part of it, but also I don't want to commute anymore, and I bluntly value my time more than the extra money. GAT
I whole heartedly agree with this notion and hope to move in a similar direction. The 'New Rich' are rich in time as per Tim Ferriss. Working for 40 hours a week for 40 years to retire is the bigggest misnomer of our time. In this day and age the 40 should also be changed to at least 60 in each figure....