In this episode I speak with Giuliana Bordigoni, Director of Specialist Strategies at Man AHL.
In her role, Giuliana oversees the firm’s strategies that require specialist knowledge. This includes, for example, alternative markets, options trading, credit, and machine learning.
We spend a good deal of time discussing alternative markets, a focus of Giuliana’s in both her current role and her prior as the Head of Alternative Markets. We discuss the potential benefits and challenges of introducing alternative markets to existing CTA programs, unexpected roadblocks in doing so, and the opportunities that Giuliana is most excited about today.
We also discuss machine learning, which is treated as its own unique class of strategy rather than as a technique, and why Giuliana is so excited about systematic credit today.
I hope you enjoy my conversation with Giuliana Bordigoni.
Transcript
Corey Hoffstein 00:00
Okay, Giuliana, are you ready? I am ready. All right 321 Let’s jam. Hello and welcome everyone. I’m Corey Hoffstein. And this is flirting with models, the podcast that pulls back the curtain to discover the human factor behind the quantitative strategy.
Narrator 00:22
Corey Hoffstein Is the co founder and chief investment officer of newfound research due to industry regulations, he will not discuss any of newfound researches funds on this podcast all opinions expressed by podcast participants are solely their own opinion and do not reflect the opinion of newfound research. This podcast is for informational purposes only and should not be relied upon as a basis for investment decisions. Clients of new found research may maintain positions and securities discussed in this podcast for more information is it think newfound.com.
Corey Hoffstein 00:53
If you enjoy this podcast, we’d greatly appreciate it. If you could leave us a rating or review on your favorite podcast platform and check out our sponsor this season. It’s well it’s me. People ask me all the time, Cory, what do you actually do? Well, back in 2008, I co founded newfound research. We’re a quantitative investment and research firm dedicated to helping investors proactively navigate the risks of investing through more holistic diversification. Whether through the funds we manage the Exchange Traded products we power, or the total portfolio solutions we construct like the structural Alpha model portfolio series, we offer a variety of solutions to financial advisors and institutions. Check us out at www dot Tink newfound.com. And now on with the show. In this episode, I speak with Giuliana bordoni, Director of specialist strategies at man hl. In her role, Juliana oversees the firm’s strategies that require specialist knowledge. This includes for example, alternative markets, options trading credit and machine learning. We spend a good deal of time discussing alternative markets a focus of Giuliana is in both her current role and her prior as the head of alternative markets. We discussed the potential benefits and challenges of introducing alternative markets to existing CTA programs, the unexpected roadblocks in doing so and the opportunities that Giuliana is most excited about today. We also discuss machine learning, which is treated as its own unique class of strategy rather than as a technique. And why Giuliana is so excited about systematic credit today. I hope you enjoy my conversation with Giuliana bordoni. Juliana, welcome to the show. Very excited to have you here. You have the distinct honor of being the last guest this season, though certainly last but not least, so very excited to have you here. Thank you for joining me,
Giuliana Bordigoni 02:59
thank you for inviting me. And I’m glad to be the last one of this season.
Corey Hoffstein 03:03
Let’s start for the listeners who maybe don’t know who you are with a quick introduction in your background.
Giuliana Bordigoni 03:08
So I’ve been a manager for about 15 years now, before joining HR, I was doing a PhD. So I came straight from University. My PhD was about utility maximization. So it was about mathematical finance. And it was fairly theoretical. So after the PhD, I wasn’t really sure what I wanted to do. And I said, Okay, let’s give this a try to the industry for a few months. And then if I like if I stay otherwise, I just go back to academia. So I took an internship at many cello for six months. And you see, after 15 years, I’m still in the same place, but not doing the same job in my defense. So obviously, having been 15 years in a place, I’ve covered several roles, starting from more like the signal generation and doing more like a real core research. And and now I would say that is more about managing people and projects than actually me doing the research itself, which I missed a bit but enjoy the rest as well.
Corey Hoffstein 04:12
Now in your career at man HL, you’ve worked on momentum strategies mean reversion strategies, volatility strategies, credit strategies, alternative markets and machine learning strategies. What was the most challenging project you’ve tackled so far in your career?
Giuliana Bordigoni 04:28
I feel quite lucky. And that’s to have been exposed to so many projects. That is the reason why me so I really enjoyed that part. But that has been one of the main challenges. So just to change topic, every time and every time, learn about the new asset class or learn about the new kinds of strategies and helping in support and maybe other people but also being productive because it’s not that I could change just for the sake of my learning, right. The point is to be productive and that that’s been definitely one of my main challenges. But I’m not trying to avoid to answer your question of question is that which one, so let me tell you which one I find the most challenging, and I think most likely is machine learning. The reason is, because is the one that is less than my background, I told you that my background is more on like takasi processes, mathematical finance. So it’s quite different from machine learning and these kinds of techniques. So for sure, that is part of the story. The other part of the story is that it is the one that is less transparent. If you think about all the other projects I worked on, I mean, I can always explain, you know, what the strategy is doing, I can have an expectation, or what the strategy would do in certain scenarios. And if you think of machine learning, you don’t have that ability, if you have it, to me is not the point on machine learning, because then you can use something simpler to make, you want to use machine learning to try to discover things that you can’t see. And therefore it makes it less transparent. That is the other part of the challenge. But at the end is also like to think that I’m a quant, I like details. And in machine learning because I told you know my background, I need to rely on other people. And that’s also like the other part of the struggle is so poor results, a personal struggle in accepting that you need to delegate to people who are more expert than I am,
Corey Hoffstein 06:27
we’re definitely going to touch on machine learning later in the episode. But it’s rare for me to have a guest on its head, both such a breadth of experience in different areas of quantitative finance. But he’s also gone quite deep in several of them. So I want to spend a little bit of time on the full arc of your career. And maybe we can start with your role as head of alternative markets where I know you spent a considerable amount of time trying to incorporate asset classes that weren’t historically touched by CTAs. At a high level, can you explain maybe both the why, and the what behind this? I’m really particularly curious as to how you thought about identifying alternative asset classes that might have the properties you are looking to exploit in the trading systems you already had set up.
Giuliana Bordigoni 07:13
So you are right, I like data. And that is probably coming from my background for this character, I don’t know. So I hope that I can give you some details about what I’ve learned in the last 15 years going to the alternative market side, the why is easiest. So the why is diversification, you can think of momentum. And when you think about momentum, we know that is relatively weak affect the data PRC market at different times. And the best you can do is to try to capture it on as broad number of assets as possible. So the why is the search for diversification, the difficulties the water, because you need to come up with a list of markets that are really diversifying your portfolio and you need to search in the liquid space. So here we are talking when I was doing alternative market release, we were talking about trading momentum in this new market. And therefore, for the momentum system, I would say that you need to be able to trade on a daily basis, which define the space of the market that you can trade. And then they want you to find new market is like looking at you can go through the exchanges and see what is traded, you can talk to brokers and find out what you are missing what they think there could be an interesting asset. And obviously, we have our internal experts that they will suggest market. So the list is fairly large. And we time we have gone through I would say quite a big part of it. You said you’re curious to know about the properties that we’re looking at that I’m looking for in this market and I told you one which is daily liquidity. Second, I told you already is diversification meaning that I don’t think that you want to trade alternative markets to capture a more expensive effect that you can capture in traditional space. So liquidity diversification, and then it needs to be suitable for momentum, meaning that I believe that momentum can arise in different markets in all the markets at different times for periods of time. The point is that, but there are some markets that they may have some other drivers like a peg currency for which a momentum system is not suitable. So that is the other properties easy momentum being suitable without wanting to cherry pick back sort of broad fundamental properties. And then obviously, there are the ways like is this market or the bid offer in this market tight enough that it will be profitable after the cost? Some would say these are the major question that I try to answer.
Corey Hoffstein 09:51
I’d love to stick with that question of is momentum suitable? Because for many of these assets that we’re talking about, they might trade OTC The end you might not have easily available data for them. So my question would be when you’re thinking about exploring whether a new asset can fit in the system? Are you trying to come up with a hypothesis as to why momentum or trend might emerge, for example, because of the players or the structure of the market? Are you ultimately letting the historical data speak for the asset class?
Giuliana Bordigoni 10:23
Well, I would say a bit of both in the sense that I don’t want to cherry pick. So I think, as I said earlier, you can find momentum in markets a different time, if I just look at a back test, it might be that that market is going to start playing tomorrow, and I’m going to miss out. So I’m really looking for diversification for different drivers. That is the main thing that I look at. But as I say, you need still to be comfortable, that there is a chance for that addition to be profitable, meaning if I made the example of a pegged currency, you can’t know if there is a momentum system can’t know if the central bank is going to intervene on the currency itself. So in that case, it doesn’t make sense. So when I mean, fundamentals, I’m trying to look at reason why it can’t trend, really, and then you to be like sort of obvious, big results. If it makes sense.
Corey Hoffstein 11:18
I would suspect that one of the biggest challenges to this whole endeavor was in gathering and cleaning data, as well as setting up the operations to trade some of these asset classes, particularly those that are trading OTC. And just to make things a little more concrete for listeners, some of these asset classes might be, for example, credit default swaps, or inflation swaps, or even crypto potentially. And so my question to you would be, how do you face these challenges? And how do you weigh the trade off of the costs in doing so particularly when you might not know the costs upfront, versus the Potential portfolio benefits that come from introducing these new asset classes,
Giuliana Bordigoni 11:59
I think that you arrive come in especially the more you do this job, it means that you have gone through the markets and they are a higher capacitor, they are more liquid. And the more you do this, the more you’ll end up with the cooler market that is less liquid, still daily trading. But less liquid than the other two have added already which me and they might be more difficult for many reasons, you might need some pricing code that you don’t have. And the capacity that adding might be limited. I think that era big firm have an advantage. Because obviously they have like more like larger, they tend to have larger team, which means that there might be more appetite to do the extra work even if the added value is less. But obviously the way in which you do it is prioritizing it start from the ones that are going to add more, and then you go through the list. But I still think that there is loads of appetite on my side to look also at the one that where the trade off, it means that you’re going to work a lot and like is more market. That’s the thing that we’re fifth,
Corey Hoffstein 13:04
were there any unexpected challenges that you faced? I know as researchers, we often hand wave things like what goes on in the back office and account operations. As you moved from the data and research to actual implementation. Are there any interesting roadblocks or challenges that you hadn’t expected?
Giuliana Bordigoni 13:22
Oh, absolutely. And you’re right. When you do research, and especially as a quant, you always think about hardware more than this price serious? Or do I get into the system? And how am I going to trade it on a daily basis? That is your concern. But there is way more to it. And you mentioned the coffee’s. But there is also the legal side. So the legal and compliance side. And one second ago, you mentioned crypto. And when you look into crypto, the legal and compliance are the really the big hurdle. So to cross, I’m not saying anything controversial if I say that crypt are more susceptible to financial crime than other asset classes. And therefore, if you want to add it to your portfolio, you really need to work with experts like specialists on the legal team and the compliance team that will need to make sure that you are not inadvertently breaching any regulations. But also, you have risks that you might not think of when you trade financial assets like theft risk, or even more simply, how to actually all the coins overnight. If you have a certain type of investor like us invest in they need to have a US custodian. So these kinds of risks are a bit outside the usual one space, but you even have other risks you have like sort of the risk that Ebola Tidemark think of like a market that you can’t trade. It’s not clear that you have trade for example bilateria And then what happens after is I volatility in the market and then bank margins will rise. Now what does it mean that bank margin rises? It means that the capacity of the bank to offer intermediation is reduced. So again, And you have to deal with the fact that you might have to control your position in this environment. And obviously there is the biggest challenge. When you look into the smaller markets, that is liquidity. Liquidity might be changing, you might have done a good job assessing it, but simply my Riah while you’re trading it. And so again, you need a bit of flexibility to keep assessing it and keep adjusting whatever your assumptions have been.
Corey Hoffstein 15:25
One of the points that you’ve made a couple times now is that the markets you’re looking to incorporate should have daily liquidity for momentum and trend strategies really have two follow up questions to this. The first is, is that a requirement or constraint that you think can be relaxed without necessarily losing the efficacy of the strategies? And second, how do you think about assessing that need in OTC markets where both live and historical liquidity may not be easily assessed?
Giuliana Bordigoni 15:57
I think that for momentum, you need the way in which as the momentum is like you want to capture a trend, but there will be a reversal and you need to be dealing with it, which means that you need to be scaling your position and possibly turn it around. Which means that if you look at markets that are less liquid than there, you might get stuck in a position when you want to turn around. And I just don’t think is the kind of property that you want in a momentum strategy. So I think the for me, a momentum Strategy is a strategy to capture this trend in a wide pool of market but this pool of markets are like sort of daily traded in terms of the OTC market. The point is that how do you assess the liquidity because it might be more opaque than for markets that are exchange traded. I would say that in general in the last 10 years after the financial crisis, loads of OTC Markets are now clear them, which means that they are less opaque than they were 10 years ago, but there will be still some that they are not. And the way in which you deal with it is like trying to be as engaged with the brokers, the people you are executing with as much as possible. So most of these markets will have to be traded via a trader, they are not automatically executed, which means that the trader has a relationship with the executing broker that is requesting a quote from. And so there is a sort of constant assessment of the liquidity. But as I said, as of today, there are quite a few volume data that you can find, for example, interest rate swaps, or for likely default swaps and some of the asset classes that you mentioned earlier. So they are becoming more transparent from that point of view. A consistent
Corey Hoffstein 17:43
theme this year in my discussions with different systematic futures traders, has been the idea of pure trend versus more broadly diversified future strategies coming out of 2008 Pure trend was obviously really attractive. And then I think the struggles of the 2000 10s caused a lot of managers to start incorporating diversifying strategies. How do you think about these trade offs? And what do you think the implications are for allocators looking to incorporate one approach versus the other within their portfolio?
Giuliana Bordigoni 18:13
When you mentioned broadly diversified strategy? Do you mean like sort of moment Toumani? Why they’re like sort of maternity market Umino to like, continent trend strategies, you mean both of
Corey Hoffstein 18:25
them? So including things like seasonality and carry and value and mean reversion? Really just going beyond traditional trend and momentum.
Giuliana Bordigoni 18:34
Okay, so I think they’re first of all, I still believe in diversification, both in markets and in signals, there is value in a portfolio to three different effects. And if you look at these er, you saw the pure trends, meaning like trend in the traditional space, and like relatively slow trend have done fairly well. And then you have the authority market space and the non trend space. In general, the maybe historically they have had even higher returns, and then maybe this year, they have lagged a little bit different space. And my point is that historically, this year, it might have like a bit different trend. But the point is that I think that we don’t have really a crystal ball to know at the beginning of the year, what is the effect? What is the signal or even what is the set of markets that are going to work best in the coming year? Because we don’t have a crystal ball. My point is that you diversify as much as possible. And so I see a combination of all these things working well in a portfolio.
Corey Hoffstein 19:39
Your title at man HL is director of specialist strategies. As far as titles I’ve encountered in the past this is a pretty unique one. Can you explain to me what the title in the role entails?
Giuliana Bordigoni 19:52
What do we mean by specialty strategy is any strategies which require some specialist knowledge data It comes from my boss, actually. And I quite like it. But it’s not my idea. So what does it mean? That requires some specialist knowledge, it could be like an asset class for which you require a specialist knowledge. Because maybe you have to price it, you have to be some total return or you need some special knowledge for the signals that you’re trading similarly can be like in the more traditional asset classes like future and cash equities. And then at that point, it needs to be like, You need to read some specialist knowledge in the strategy itself, like, for example, machine learning. So effectively, what it means that I look after the alternative market space, which does not include future photo and cash, equities, and machine learning, in summary, so it is quite a varied group of strategies,
Corey Hoffstein 20:49
beyond maybe colonizing space and selling Moon property, do you think we’ve ultimately hit the final frontier of alternative assets? Or do you think there’s still new markets to be traded?
Giuliana Bordigoni 21:01
So I told you that for momentum, we need daily liquidity, and therefore, because of that you need so this space is not infinite. I’m not claiming that there is an infinite pool of market where we can draw from, but I don’t think that personally, I don’t think that I’ve looked at all of them. Yeah. And so there are still some that haven’t looked at it yet. And in general, I think that there will be some, they’re not liquid today, and they will become liquid. But I can give you some example, you can think, for example, lithium and cobalt, we know that they are going to be an important for electric vehicle for like the energy transition for the transition to cleaner energy. But at the moment, they are not very liquid, do I think that they might become liquid in the future, I don’t see ya Oh, and the other thing is that there might be markets that you are not able to access today, because of regulation, I can give you an example, an example is India, you can’t access from offshore loads of Indian markets in your commodities. But recently, India is started to open up to opening up. And what they have said they have said, they have open up like a very limited number of contracts. And admittedly at the moment, they are like contracts that there is a global equivalent, and they are fairly correlated them to their global equivalent. So maybe at the moment, they are not very interesting. But it’s a sign towards the direction of opening, which means that maybe in the future, there will be more market that we will be able to access there. So this space is not infinite, is not over, but is not infinite. But I think that is also a space that is changing through time. So as some markets, but you have the other direction as well, there might be some markets that we trade today, and then tomorrow might not be liquid, and you will have to remove it to the portfolio. And my point is that if you want to be in the space, you need to be flexible. And you shouldn’t consider removing a market like a failure. Because maybe we started earlier, we spent maybe loads of effort to add these new markets, then we take it out and you say, I failed, you haven’t failed, they simply that you need to be flexible, and the liquidity is changing, and you just need to adapt, but in both ways. So it might go the other way as well.
Corey Hoffstein 23:20
Given the diversification potential of these newer, less liquid markets, do you think there’s a benefit in introducing them in a smaller size, even though you’re capacity constrained? Or for a large firm like man HLD, you really need to think of it more as a binary exposure?
Giuliana Bordigoni 23:36
No, I don’t think so. I believe in trading the size, you don’t leave a footprint in the market. So if it’s a small market, you’re just small. So I strongly believe in that. But I would still add a small market and I would rather grow it as long as it’s tradable and respects the property that we discussed earlier. I don’t have a problem in putting in a small market as long as we the smaller location, obviously. So yeah, I’m all for that.
Corey Hoffstein 24:03
Speaking of markets that are opening up like India, in July 2021, you were on a podcast where you spoke at length about the opportunities related to investing in China and I thought was a really fascinating podcast episode, I was hoping you could explain as the director of specialist strategies, why you find China so appealing as a market and whether your enthusiasm has waxed or waned since you recorded the episode last year.
Giuliana Bordigoni 24:30
The last question was about this more or less market if I think that even a small market added to a portfolio will make a difference. And now so you have to think that for money I’d say look the really the like small niche market. And then you look at China, which is like deep and diversifying. So it seems like very natural space to move towards. So Chinese markets are bigger. If you look at the commodity space, and you take the 10 was traded contracts, by number of contracts. I mean, in the war, you see that in the commodity space, or the intellect, about eight out of 10 are Chinese commodities, if you look at the size of the Chinese equity market is second to the US by market cap. If you look at that outstanding, like, again, China is amongst the biggest. So we all know that is a huge economy. And therefore, we have like contracts, you can trade markets that are big. But you know, these because you saw my post anyway. So they are big, but are also diversifying. And that’s true for all the asset classes are mentioned. So it’s true for the equity space, that if you look at what is the most correlated in this index, today, CSR 300, for example, you get the Hang Seng, but the correlation or give or take is going to be around 60%. If you go to like other major economies, you’re probably below 50%, then you go to the bond market, you have debt, the US and Europe have just come out of very low interest rates, if not negative. And what you saw, you saw that China actually stayed above like around 3%. So it didn’t really see these low interest rate environment, the US or Europe have lived. And now when most of the countries in the world that I came, not all of them, but most of them are it what you see, you see the Chinese in sort of an easing cycle, which makes even the bond market we are talking about diversification. But then obviously, the most interesting bit to me is commodities. And it is commodities because you have markets which are unique to China that they don’t have an equivalent, and you have markets that they might have an equivalent, but for reasons like local results, or economical reason. They are different from the global equivalent. So I gave you a long explanation. But the two points as they are, the markets are deep and diversified.
Corey Hoffstein 27:00
So one of the things I actually learned in that episode was that there are some really, truly unique markets trading within China. So for example, eggs and apples are markets that you don’t find anywhere else in the world. Can you expand on some of these unique markets that most of us aren’t even aware of, and the opportunities for diversification,
Giuliana Bordigoni 27:19
you are looking at the agricultural sector there, you have, like I mentioned eggs and apples but there are like other markets as well, me at least x and Apple, we can agree that we know what they are. Now, if I tell you another one, which is Jujubee it’s a berry, but what I mean is that there is a list of agriculture that are like that are not traded as well as there is like some of them that they are traded elsewhere. Like you have soybean right soybean you can trade it in other countries in the world. But the point is that when you go to the agricultural says I was trying to say earlier, you have like local weather, you have harvest season, even transportation, right that entry into the equation. So even the markets in general and agriculture, even the markets that thought they might have inequivalent, they are diversifying, compared to the euro equivalent. And then the asset class, which for me is extremely fascinated is industrials. And the reason why I find it fascinating is because you don’t have many the strength traded elsewhere. And binders. I mean, like material, there are methods that are used really in the industry. They are markets that are important to the Chinese economy, meaning that they’re using the manufacturing they’re using in construction, and therefore they are fundamental to the Chinese economy, you can trade them. If you look outside China, the thing you can think of in that space is steel. But steel is not a big contract outside China. When you’re going in China, you have an entire sector send examples. I know you want some examples. The most fascinating one is probably glass, but you have an entire sector for that. And then you have more metals, then you can trade outside China’s was. So all I’m trying to say is that there are plenty of unique markets. That is part of the fascination really. But at some point, I don’t know if and you asked me if my enthusiasm towards China is waned through time or comfort when I did the podcast? I don’t think I’ve answered it. I can tell you actually did not. There is an exact problem a year ago things have changed them compared to a year ago. We can now trade via Q fi licenser. Some of the markets, some of the futures, the Commodity Futures listed in China from offshore, and that’s really huge development. This is something that is relatively recent. So okay. It’s not the full list of markets that you can trade on shore, but it’s a start and I would say that it is maybe a third or something of that order. To me, that’s a great sign. Hopefully it’s a sign that in the future, we will be able to access all of them from Most shorter. So definitely my enthusiasm actually mentees asked me is that peak? Let’s see, I think
Corey Hoffstein 30:05
we would all certainly hope that access is an arrow that really only moves in one direction. And it only gets greater over time. That said, from a risk management perspective, how do you think about accessing these markets knowing that there’s the potential that regulators could reverse course, at any time,
Giuliana Bordigoni 30:23
I am that thing that you need to be flexible, as I said earlier, so to me, you capture what you can until you can, if further is reverse course, you will adjust and you will, you’ll stop trading. So that’s our field, obviously, also believe in being transparent, which means that if you want to invest in China, you need to be very clear that this is the kind of that this can happen, the portfolio will have to adjust to these kinds of changes.
Corey Hoffstein 30:50
One of the consistent patterns we see in markets over time is that as market access is commoditized, we see the markets themselves become more highly correlated to traditional more liquid markets. Do you think that the opportunity for additional diversification coming from these markets in China will ultimately be short lived? Or do you think that there are potentially structural barriers to entry are true fundamental differences in the markets themselves that will continue to make them attractive as diversifies over the long run?
Giuliana Bordigoni 31:22
I agree that if you have a market that is additive to access, you don’t have that risk, and so, not having that risk makes your life much easier. Now, what do I think about specifically about Chinese commodities is that they are quite big fundamental drivers to them, I mentioned earlier like local weather service the economy. So, I do believe that these are like sort of the biggest driver to the market. And they will stay and I also think that if you look in general at global markets, as an example, which sector is the most diversifying, usually comfort in like in a sort of macro portfolio, they tend to be commodities, they tend to be a diversifying compared to like financials also between each other, right. And here, what we are talking about, you’re talking about the portfolio like that is dominated the sector market that is dominated by commodities. So I see that risk can be less in this space. But more in general, I would say that if you have a market that becomes commoditized, what you see you tend to see like faster momentum, the disappear. So what I mean is that it depends even which kind of speed you trade. So if you think about let’s take the most liquid stocks that you probably don’t want to trade fast momentum. But we know that again, at the different times, we may have very strong trend of the medium to low speed inequities as well. So to go back to your question, yes, or the to access market, you don’t have to be scheduler. China specifically is commodity dominated. But in general, I think that depends on what do you want to trade in, in Moscow?
Corey Hoffstein 33:01
Alright, pivoting topics entirely. Here, I want to go back to machine learning for a moment, when you were laying out the topology of strategies, you overseeing your role you placed machine learning as a category unto itself, which strikes me is pretty unique as most of the quants that I’ve spoken to, in the past, consider machine learning really to be a tool or a technique, rather than its own unique category of strategies. So I was hoping you could explain why you treat it as a unique class of strategies.
Giuliana Bordigoni 33:30
So your artistic techniques and approaches like you can apply everywhere. And you could have just machine learning search inside each different team. In practice, the point is that when you do machine learning a lot is about IU is about the infrastructure that you have to build is about the models that you have to build. It’s about how you select the models, or you combine the models or you validate the models. So if you spread out many teams, what happens is that you are like other duplicating the third, fourth, or you are not really getting the most out of the economy of scale, I would say. So if you have like a team, which is focused on that and focus on this is sort of the technique that we’re going to use, this is sort of the methodology that we are going to use and then these sort of becomes a basis in my be more effective. The other part of the answer is that is also focused, meaning that if you leave machine learning as a technique inside another team, you will go with the other priorities. And so it depends how much you believe in a project, right? Because if you really believe that a project can deliver performance, you really want to have like focus and you want to really get it prioritized and that things are there. If you split it inside many teams, I also believe a lot in collaboration, meaning that it’s not that the machine learns is isolated or doesn’t work with the team. So for example, if they’re looking at machine learning and cash equity As the what happens is that they will work with the cash equity teams. So there will be like regular catch up. And it will be like regular updates on how the project is going. And the cash equity team will be exactly on top of the research and and machine learning is doing. So we are trying to get the most of the faculty and the focus team, in the mean time working with the other teams that are especially in the area that where this technique is applied.
Corey Hoffstein 35:27
How do you think about the problem of preventing machine learning from finding what’s already been discovered by the other teams left unsupervised, it’s really no surprise that machine learning is going to identify things like the major style, premium trend value and carry and other such well documented anomalies in the data. How do you make sure that you’re going after the idiosyncratic quote less transparent opportunities,
Giuliana Bordigoni 35:55
you can analyze the strategy output so you can analyze your output and look at how it correlates with everything else that we do, you can even neutralize some effects of a priority. What I’m trying to say is that you can do it in the design stage. But I want to believe that you always double check, which means that at the output stage, you should check what you’re really capturing and how it correlates with like you say like factor so moment to more what you have in another portfolio. Because the last thing you want to do, as I said earlier, you don’t want to duplicate without realizing on something you trade already, my view on machine learning is that he needs to be adding something and doing something different. Otherwise, I’d go for some technique which is simpler and more transparent.
Corey Hoffstein 36:43
From our pre call. I know that you’re particularly excited about opportunities and systematic credit today, which sort of runs counter trend for most of the other major quant firms I’ve spoken to who were really enthusiastic about systematic credit a few years ago, and their enthusiasm seems to have waned a bit. And they’ve largely gone quiet ever since. Why are you so excited today?
Giuliana Bordigoni 37:07
You’re telling me that I got late into the space? I’m a little late? No, I think that you’re right, there were quite a few launches of content found a few years ago. But if you look in general credit space is dominated by discretionary traders, I think I think it’s fair to say that, despite these few launches, is still dominated by discretionary traders. So the quantity automatic credit is still in minority in this space. We did start working on it a few years ago. But it takes a bit of time to get comfortable and ready with the new asset classes. It takes time, because you need to build a framework to trade it. And we are talking about hundreds of bond that you want to screen. Now this is like the main difference graduate system with discretionary. You want to trade a broader set of that, and therefore the framework takes a bit more time. In addition, you are in a space that is not the traditional space of quant meaning that it’s a bit less liquid. We are used that we studied many times in these podcasts that normally we trade on a daily basis. Er, you might not be able to trade on a daily basis, you might send a trade that is not going to be filled. So you need to have some mechanism to deal with it as well. So what I’m trying to say is that you have to be the framer, but you also need to be comfortable with the execution with a different profile of the strategy. And that takes time. So the reason why I’m excited today is because we have been working on it for a couple of years. And I feel that we have overcome these challenges, and started becoming comfortable with this asset class and with differences to my more natural face. But as I said, I don’t think we are late at the game because quants are still in minority,
Corey Hoffstein 39:05
knowing that you have a wide berth in the types of projects that you oversee. As you look forward today, what are the projects that you’re really the most excited about?
Giuliana Bordigoni 39:16
I tend to get excited very easily. That’s one of my problems. But I think that the last 15 years I’ve worked in a space in which there was of daily trading and then credit is the first project that I’ve done in a space that wasn’t the feeling. So what excites me today is to try to go to spaces that are less liquid, maybe not doing momentum, maybe doing some other strategies, but I would like to go to trade in a systematic way asset classes which are no quant space and not just CTA just not cold space. But I still am excited by and doing things and machine learning. I told you earlier bizarre that these things don’t excite me anymore they do. But as you My next challenge is really to try to go to spaces that are where the fourth assumption where we started, this color has been removed. So spaces that we can’t read daily. Well,
Corey Hoffstein 40:13
we come to the last question of the podcast. And it’s the same question I’ve asked all my guests this season. As you look back on your career, what do you think your luckiest break was?
Giuliana Bordigoni 40:23
My first luck is to have ended up in a place where I’ve enjoyed my job for 15 years. Because I don’t think it’s that easy to find a job that you really like, for some money, and not just the job, even the place. So I do think that I’ve been extremely lucky with the place. Now, I ended up here, I have to say that is also a bit of a lucky coincidence, because I did an interview about several months before finishing the PhD. And they were looking for somebody with a slightly different profile. They wanted somebody with option experience, and somebody available relatively soon. So I was clearly not fitted for that role. But they fit for that role. And but they said, OK, when you’re finished, you get back to us. So I would say that honestly, I’ve had many interviews. And when I finished the PhD, I was fairly tired. I just said you know what, I’ll just give it a go as I told you earlier, and I ended up too busy in finishes. So I would say that part of was a bit lucky because if I was less tired, maybe interview with a little bit more. So I think that my main life has been in debt.
Corey Hoffstein 41:37
Well, Giuliana, thank you so much for joining me. This was really great.
Giuliana Bordigoni 41:41
Thank you for having me. It was a pleasure.