C (45:25)
Yes, so, so, so Rob put out a blog post because we've, we've had quite a few heated discussions about fixed income. What a surprise. He used to be a fixed income trader. I think it was in the city before he went dark and went to the systematic universe. So he was like a. And, and in fact I followed in his footsteps. In fact our wives share birthdays. So it's like, it's really kind of, of spooky a little bit. And, and, and I say well, I think actually I can do better in fixed income than if I concentrate on, on that then I can do better than just, than just doing the full cta. And Rob of course is, doesn't do, doesn't do that. Rob trades sort of 200 futures and so forth and doing very well as well. So you know, full kudos to, to Rob and he's always trying to challenge me. So he put out this paper which is really, really nice. And an amazing thing about Rob is the speed at which he does research. And that was even in ahl it took him he can prototype something very, very quickly and get to the heart of the problem. So he does two things in this blog. The first bit is looks at Curry and he says does Curry add value? You know how much Curry does add value in into the asset class? Because of course I, I like to trade fixed income. I like to trade effects especially now when interest rates have, you know, there's a gradient of interest rates across different economies. And he says like how valuable is carry historically? And he puts plots lots, multiple asset classes and multiple markets. And what you find is that if you were to look at the sharp of spot versus the sharp of what we trend followers doing, the sharp of the spot doesn't really explain what we get. It's a sort of flat feat. And then it does the same thing for the carry returns. And again it doesn't explain particularly well what we do. And then you look at the sharp of the underlying asset total return series. So both carry and spot moves and, and trend obviously you get this smile effect that if the sharp of the underlying market is plus one, you will get maybe 0.8 sharp into your, into your P and L of your trend following system. And the same if the, if the sharp is minus one, you will get again 0.8 or thereabout. And in the second part of the paper he also observes that different asset classes do actually perform differently. So if you have a sharp of One in your underlying asset, bonds and FX and metals seems to be actually producing, doing very well and actually able to get 0.8, whereas equities are unable to essentially to harvest the underlying trend in the asset. So and both observation by the way, I completely agree with and I think he's just in missing like the first interpretation that I would like to put is in terms of the the spot versus the carry. It's precisely what I say. What I say is that if we have an asset class which is essentially has got very little carry or it doesn't have carry at all, then the spot doesn't really explain our own sharp and it's not a particularly good. You don't actually perform particularly well. And I'm not saying that carry is like the be all and end all. And if you have curry in your system you'll get a sharp 7 then that's not the case. What we are trying to do is we're trying to take a market, right? If you think about what a CTA does, this sharp of a single market that we have may be somewhere between 0.2 and 0.3. And then we diversify across our portfolio and that gives us a reduction of volatility of say 33 1/2 x and that gives an increase in the Sharpe of the overall portfolio. So my 0.2 Sharpe Sharp asset, if that's my average Sharpe and I multiply it by three and a half, I get a 0.7 sharp CTA. And that's really the mathematics of what is going on here. And what I'm saying is carry just gives you this little extra right in terms of instead of the average being 0.2, the average is 0.3. And that when you multiply it by three and a half now gets me from, you know, to 1.05 sharp. And I kind of like that. So the question is, you know, do you want to tilt your portfolio to assets which are high carry does carry. Explain this in the same way that I would might ask do I want to tilt my location to markets which right now are in supply shock or supply demand. The structural reason why they are more likely to trend. Maybe there are more physical players in the market. Various reasons why we might think about it. And the aim is not to get to say, oh, I'm taking my sharp from 0.2 to 0.7. The aim is to say, can I tilt? Do we have historical reasons or fundamental underlying reasons why we understand why there is why there is a trend. And I think what is nice about his blog is that the answer is yes, if you look like just a spot market, you get a relatively mediocre shop. But if you do, if you put the carry into the price time series, then you get a slightly higher one. So that's the first part of his paper and it's a really nice one. The second one is even more interesting because he just makes this observation and he says, well, I don't really understand why equity seems to have trended worse. Right. Long only equity has been on a tear, right. So if I look historically in the last 50, I can't remember 30 years that he's looked at really nice shop, but trend hasn't performed that well. And I think here is where a lot of people miss out in terms of the mathematics. So first of all, let's do the mathematics and then we can talk about the fundamental reasons why different asset classes have it. So I put out another paper with Adam pedal about the mathematics of, of trend following. And you know, most people say, oh, if there is a trend in the market, you should be able to make money money. But actually the trend quality, the autocorrelation of the underlying price is really important. Trend is about positive autocorrelation. And you will have markets where this auto correlation is very muted in equities, it's actually negative. Well at different times, at different times and at different horizons. But that negative autocorrelation translates into a lower performance. So you're not able to harvest as a trend follower necessarily using a generic trend following machinery you will be able to get less trend. Right? Because the auto correlation will basically means price goes up, you put in a position and then you suffer a day later when price comes back down. That's what negative auto correlation means. It's like on a short term basis, up, down, up, down, up, down. That's not particularly good for trend. Okay. And what is observed which is true is that in fixed income we see actually a much higher autocorrelation and that means that we're able to harvest it. So the paper we put with Adam and myself, we actually did the mathematics for that and we kind of can calculate what is the worth? What is 1% of positive autocorrelation worth to you as a trend follower? And then there's the second question, which is a much more deeper question, is like why are different markets different autocorrelations? And I think here is really a structural question which requires you to understand what's going on in terms of, for example, rebalancing in equities in Terms of value traders in equities, in terms of day traders or zero day to expiry the option market. So you have different players and they will have different dynamics that will feed into the price. And once you understand that, you can think about, oh, how do I modify my trend to be able to address the structural nature of the market? And by the way, the structure can change over time. So it's a moving target. But I think it's really important from our perspective is when you look at different asset classes, like, you know, you can close your eyes and say, yes, I'm going to have an inventory system on the side, but if you're saying I'm committed to trend, I really want to trade trend, then it's really important to you for you to understand, okay, what are the things that are structural in this market and how they play? And how can I combine them into trend in a way that I'm still got this positive convexity, this trend dynamics, this beautiful risk management system and at the same time takes into account the physical nature of what the actual thing is and who the other players in the market are. So I think it's a really great paper by Rob. I really enjoyed that and I think I'm looking forward to another coffee with him later on where we can discuss this blood.