Today I am speaking with Omer Cedar, CEO and co-founder of OmegaPoint.

One of the significant trends in quant equity over the last decade has been the attempt to better control for unintended bets and idiosyncratic risks. At OmegaPoint, Omer comes at the problem from the opposite direction: helping fundamental managers better focus on their idiosyncratic risk and recognize the factor risks they may be unintentionally taking.

We discuss how quantitative investors have impacted markets, how fundamental managers should think about factors, the low-hanging fruit for optimization, and surprising lessons Omer has learned in evaluating fundamental portfolios.

The idea of embracing idiosyncratic returns is, arguably, the antithesis of traditional quant investing. But in discussing the lessons learned about unintended bets from the opposite direction, I think there are important ideas that quants can take away.

I hope you enjoy my conversation with Omer Cedar.

Transcript

Corey Hoffstein  00:00

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  00:18

Corey Hoffstein Is the co founder and chief investment officer of new found research due to industry regulations he will not discuss any of new found researches funds on this podcast all opinions expressed by podcast participants are solely their own opinions 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 newfound research may maintain positions and securities discussed in this podcast for more information is it think newfound.com.

Corey Hoffstein  00:50

Today I am speaking with Omer, cedar, CEO and co founder of omega point. One of the significant trends in quant equity over the last decade has been the attempt to better control for unintended bets and idiosyncratic risks. At Omega point, Omer comes at the problem from the opposite direction, helping fundamental managers better focus on their idiosyncratic risk and recognize the factor risks they may unintentionally be taking. We discuss how quantitative investors have impacted markets, how fundamental managers should think about factors, the low hanging fruit for optimization, and surprising lessons Omer has learned and evaluating fundamental portfolios. The idea of embracing idiosyncratic returns is arguably the antithesis of traditional quant investing. But in discussing the lessons learned about unintended bets from the opposite direction, I think there are important ideas that quants can take away, I hope you enjoy my conversation with Omar cedar. Omer cedar, really excited to have you here for maybe a little bit of a different perspective what we normally get on the podcast, I tend to be super quant focus, you’re gonna bring a little bit of a fundamental mix into this and how quantum fundamental maybe can be baked together. So really excited to have this conversation today, where I was hoping to start was maybe get a quick introduction from you to our guests, a bit of your background and a high level overview of what you do at omega point.

Omer Cedar  02:20

Well, thank you so much, Corey, thank you for having me, this is a fantastic opportunity to share something that I’m a little bit passionate about, and has taken me on a journey for the last seven years and even started before that. So a little bit of background on myself, my name is over cedar, I founded a company called omega point. But prior to that I was actually at a systematic multi strategy fund called two sigma investments. I joined two sigma, when they were a small loft in Soho, and really was sort of the origin of a lot of the systematic quant world began as we know it at two sigma d. Shan, a few of the other well known sort of funds, but a two sigma, there was some interesting changes that were occurring in this was around 2005 to 2008. They really wanted to pursue the area of combining or thinking about how to combine traditional fundamental investing with quant and that was gave birth to sort of an area called the Alpha capture program at two sigma. And my background, I actually had come from grad school into two sigma, where I had done some work more on the fundamental and the business side combining that was sort of quantitative finance. And they really liked that I had sort of that background. So they said, you know, why don’t you give it a shot. And really, on the back of a napkin, we designed an idea around how to take ideas from fundamental analysts, portfolio managers, I mean, talk about sort of anyone kind of out there that has the notion of how to analyze a company in a very deep way, the notion of which companies are better positioned during a cycle, which companies have an ability to sort of weather potential political adjustments, and so on, and so forth. And these are things that we know that machines have a little bit of a harder time to discern also fundamental managers or fundamental analysts, I would say, because that’s probably more of a correct sort of term analysts spent a lot of time really reading other people and understanding is this really a management team that is going to succeed to run this company, so to speak, well had a lot of interest in that. And they said, you this is not something that we think are quite models can sort of pick up on their own, why don’t we design a process or a system to combine the two of them and really create sort of the best of both worlds. And that was kind of the birth, if you will, this alpha capture program, my role was to really build out the analyst community, and then use that to build really systematic models around that. We’ve launched that in 2008, right as the craziness was occurring in the markets, and then really grew that to have over 2000 contributors across 125 Different institutions across 25 countries. So it became a very sizable program and really helped highlight that indeed, there is this intersection that is very interesting between fundamental and quant what’s interesting about this whole sort of process is that as we were building this out, we really found that a lot of these fundamental contributors expressed a lot of interest and a lot of excitement about the fact that they were able to get access or interface with systematic quant powerhouse like two sigma. And they asked the question of, can we get a piece of this? Is there a quant lite version, right, that we can get access to that we can integrate into our fundamental strategies or our fundamental ideas? Do you know of any software system? Can you recommend something out there that we can work on? And we really tried, we looked around we said, You know what, there isn’t really anything out there. There’s this big white space where fundamental managers are supposed to do their thing, call managers are supposed to do their thing. But there’s no bridge in the middle. There’s nothing that really connect the two sides. And using a lot of the learnings that we had at two sigma, that’s really when I think this is around 2012 2013. I, personally, and a few others sort of took the plunge and said, you know, we really want to give this a shot to really build a platform that bridges fundamental inquiry. And that was really the genesis of omega point in our past

Corey Hoffstein  05:57

conversation, and one of the really interesting points you made to me was this idea that quants have sort of changed market dynamics in a way and in a manner that has actually impacted fundamental managers. Can you elaborate on that for me?

Omer Cedar  06:12

Absolutely. We’ve really seen this in mass over the last probably decade in 2007. When I’m sure the quants remember the quant quake of 2007. And certainly two sigma was in the heart of it. And it was a really important period of time for a lot of quants, survival. And fundamental managers didn’t even feel it. I mean, there was some turbulence in the markets, but it was still very much of a blip on the radar compared to what was going on afterwards in the next 12 to 18 months, right? And oh, wait no nine, when quants actually had some of their best years because of market volatility. So what’s interesting about that is that quants were still a very small component of the volume that was being sort of produced in the marketplace. And over the last 10 years, probably with the emergence of not just traditional, I would say quant investing, but also smart beta pension funds, sovereign wealth funds, think about all of these larger institutions who are trying to get access to low cost passive strategies really started investing in the idea that systematic rules based approach could really deliver value for them. And that combined with traditional quant really created a tremendous increase in sort of volumes in the marketplace that changed really the dynamic of how the markets function really today versus where they were 10 years ago. There’s obviously a lot of statistics out there that quants are what 1/10 Of the Aum globally, but nine tenths or longer bigger than nine tenths of the volume of the trading globally. So when you have 1/10 of the Aum representing 90% plus of the volume, you really start seeing a very different dynamic. And that’s created essentially a lot of what we would call headwinds or turbulent sort of markets for the fundamental managers who historically have built their strategies in their businesses on the basis of deep fundamental research with the understanding that if they deploy that fundamental sort of ideas into the marketplace, then the market will behave in a particular way, it will absorb those ideas, and as their ideas, their predictions, essentially have improved fundamentals or deteriorating fundamentals, however you want to look at that will take in place, you will actually see the stocks react appropriately. And that hasn’t happened. You know, I can tell you countless managers that have expressed their dismay at how that’s changed since they started investing a decade or two ago.

Corey Hoffstein  08:28

So for years, we’ve heard terms like quantum mental and fund detective start to emerge, firms certainly still seem for the most part to fall on one end of the spectrum or another. But why do you think that the idea of mixing quantitative and fundamental is so important going forward?

Omer Cedar  08:47

It’s a great question, because ultimately, there’s some very interesting behaviors. And we want to kind of maybe pull this a little bit apart, right? We’re scientists, we’re sort of as quants really want to understand what are the drivers of fundamental research and fundamental sort of analysis that is still very much important and active today? And what are the drivers of the quantitative approach that really makes it obviously strong and persistent and robust and complements the fundamental side? And ultimately, there’s been, again, I see it as sort of a question really where fundamental managers have built their careers on the basis of very deep sort of Benjamin Graham Warren Buffett style research, they identify companies or industries that are going through dislocations, or changes, they look at business models, they analyze management teams, and ultimately, all of these things are dynamic. And they change because people like us, right? I mean, management teams, CEOs and boards of directors make individual human life decisions. If we had machines running those companies, the outcomes might be a lot different, but you have essentially companies in the world that are being run by individuals that are making certain decisions, and then you have analysts on the other side of the equation that are investing Those companies based on a very similar sort of an idea framework of what makes a good business, what makes a not such a good business. And that intuition is very human, it’s very oriented towards people actually investing in other people very similar to the idea of the venture capital industry. And so while we can take a lot of those ideas and make them quantitative in nature, and maybe build strategies that will try to emulate that type of behavior, what we find is that deep fundamental research still exists, and still really adds a lot of value and a lot of alpha, because it’s human in nature. And it is really people investing in other people that approach and we’ve seen again, at two sigma and repeatedly again, and again, still creates a very long term source of alpha, most quant strategies we know are tend to be shorter horizon, we’re picking up on patterns in the market between a week, two weeks, maybe four weeks in length, and maybe sometimes longer. But really, there’s a sweet spot there, the fundamental ideas are really much more longer term, they tend to be more persistent, and they tend to really trend over long periods of time. And so you have this approach, where you can actually capture a much longer and robust source of alpha with a lot less sort of degradation. The challenge with that source of alpha is that it’s sort of impure, it’s sort of mining for that diamond. And you really have to kind of clear it out, polish it and really get the rough out and really find that diamond ended, because there’s a lot of impurities in the way that it’s being expressed today in the marketplace. And ultimately, that’s really where the machines come in. That’s really where quantitative investing driven by good systematic principles, robust risk management can take those ideas that are sort of diamonds in the rough, and really put them into a framework where they can succeed and be managed in a way that they’ll actually preserve the right kind of outcomes or create, if you will, the right kind of outcomes for the investors.

Corey Hoffstein  11:44

So let’s maybe talk about that framework for a bit. I always sort of felt like the cliche, fundamental manager playbook always has that upside down pyramid, right, where you’re screening the stocks down into the end portfolio. And I always said, Well, if you have a set of rules by which you do that, that’s just a systematic portfolio. And so it always felt like at the end of the day, the big difference between systematic and discretionary was maybe systematic, chose to manage risk by just having more diversification. And discretionary chose to manage risk by doing deeper fundamental dives into a more concentrated portfolio. But to me, those concepts seem a little bit at odds. It’s like where one stops, and the other one sort of begins. So how do you think about marrying these concepts together?

Omer Cedar  12:25

It’s a great question. I think that ultimately, the value of the deep research that you’re talking about is in the fact that there are some true what I would call longer term persistent, either business model, adjustments, or sort of evaluation concepts or things essentially, cannot be really replicated truly by sort of that filter that you’re describing, where it’s sort of a filter on various indicators, those sort of deeper fundamental pieces that we’re talking about, that are truly what I’m going to call bottoms up, they originate typically, from a thesis, a change in the environment, some real sort of what I’m going to call an exogenous kind of variable that typically is not easily sort of quantified using historical data. And that’s actually maybe where I’m trying to try to drive to as I’m sort of thinking more about this, as you have this notion where historical data, which is what quants use, we definitely think about it from the perspective of you have certain metrics, these metrics, obviously govern good companies versus less sort of good or poor companies. And you can observe those in the marketplace over repeated periods of time. Those certainly build these kind of quant models that we have, in many cases, on the fundamental side, it tends to come a lot more from something that is defined history, it doesn’t actually have the same historical data concept. It’s a new change, it’s a political environment, or maybe where we are right now with COVID. How is that going to change? A lot of these things, the models themselves are not going to be able to make those predictions because those events have not happened before. You don’t have the historical data. And you don’t have the statistical robustness. So I think we’re the real true intersection is when the fundamental ideas are essentially running contrary, in some ways to the traditional sort of quantitative framework that has been designed to sort of capture repeated behavior. And if you can capture those ideas, essentially, across a large scale of analysts, right, so we’re not just talking about I mean, most fundamental managers are going to create a portfolio of 20 or 30 names, if you can create a larger group that you’re working with, and they’re all coming out with these sort of unique ideas, then you actually can create that very diversified portfolio that we’re talking about where you’re screening out a larger number of stocks, you’re just doing it based on bottoms up fundamental, combined with a robust sort of a risk managed, optimized, sort of a strategy on the quant side.

Corey Hoffstein  14:39

So maybe we can ground this in an example this sort of theory of combining quantitative and fundamental together the Omega point platform that you make available to discretionary fundamental managers to give them sort of insight, perhaps into the quantitative impacts that they’re seeing in their portfolio. What does that look like? How does that play out for a fundamental manager who takes their portfolio and plugs it in? What are they seeing? What are they looking for? What are you ultimately trying to educate them on to help them either enhance returns or manage risk?

Omer Cedar  15:11

That’s a great question in terms of how we think about these sort of fundamental manager ideas, essentially, these are collections of ideas that they have, we recognize that sometimes you don’t have the power of diversification associated with those ideas. But what we can do is help them improve their overall sort of performance, sort of an Omega point, right, and their ability to sort of deploy capital in the market through a set of quantitative techniques, or overlays, if you will, that can help them sort of make those adjustments. So for example, let’s say I am a fundamental Manager with 20, or 30 securities, and I have a fairly concentrated portfolio, it’s bottoms up on a lot of deep analysis. And I’d like to deploy capital into those names. There’s sort of two questions that I’m starting to answer when I’m trying to sort of deploy capital. First of all, how do I size those names? How do I position and actually create the appropriate weights in those names? And secondly, how do I make sure that the risks that I’m trying to capture in those individuals sort of ideas are the actual risks that I’m going to be invested in and not something else? For example, if I’m going to have a view on Apple’s iPhone sales in the next month, or the next quarter? How do I know I’m not buying into China or buying into the semiconductor industry, and those bets could magnify my bet on actual Apple’s iPhones. So that’s really where factor analysis and as with quants, we understand sort of factor risk models and risks out there that can capture a lot of those bigger type of drivers can help us essentially isolate when we call the idiosyncratic the alpha in the managers ideas. And so I can take the sort of 30 ideas of the manager have, and I can actually cut them into how much idiosyncratic Alpha there is, and how much of the other factor risks that you may have. If you’re making a bet on the semiconductor industry, you’re really just making a China bet, if you’re going to just invest in this alone. And we can, you know, quantitatively really strip out these by either weighting towards the names that have higher idiosyncratic risk away from the ones that affect risk, or create what we call overlays, essentially customized hedge baskets, or ETF type of overlays that can reduce or target, pinpoint those exact risks and remove them. So the platform that omega point is created is essentially to take those fundamental ideas at their core, stripped them down into their alpha, construct them in a way that can give you the maximum alpha and hedge out as much of that factor risk those risks that the quants understand so that the managers left with the purest possible Alpha driven portfolio that is based on their ideas.

Corey Hoffstein  17:37

So how is this different than a lot of the other industry standard optimizers that are available to fundamental investors things like bar X el mar Northfield?

Omer Cedar  17:48

Yeah, I mean, I think that ultimately, there’s elements that all of these sort of optimization engines and factor models have that allow you to cook your own sort of stew, so to speak, one of the things that omega point has done is to really make a lot of that turnkey. So we actually are partnered today with bar our partner with axiom Well, we’re partnered with Wolf research, which is interesting sort of a data, right and alternative data and a factor model provider, and what we do with them, as we essentially take their factor model all their data, and we combine that obviously, with our software, which has all of the attribution analysis, a lot of the quantitative tools to be able to identify the signals to be able to strip out the alpha, as well as we have sort of fully built an optimizer, all of that gets done sort of in a turnkey fashion so that the manager, again, remember, this is a fundamental manager that doesn’t have necessarily the background in quant, he doesn’t understand optimization, he doesn’t really understand how to put all these pieces together, they’re essentially being done in a turnkey fashion for him. And ultimately, all they have to do is really provide their fundamental ideas. And the output is a portfolio that has the maximum number of alpha driven ideas in the portfolio and gives them really the insight into what decisions are being made by the computer. So again, the notion that bar axiom or Northfield have those tools is great when we found and they have found to in the industry is that quants will take those tools will understand how to work with them how to build a very effective strategy, but a fundamental manager will not. And the challenge is that really has created a gap in the marketplace for them. And that’s really the gap that omega point is trying to fill.

Corey Hoffstein  19:22

So talking about factors for a second, depending upon your interpretation of the literature, the factors, Xu is either just like a handful of definitions, or it’s hundreds of these different characteristics, which I have to imagine, as you get to that far end of the spectrum is perhaps incredibly overwhelming to a fundamental manager, but perhaps maybe has a little bit more intuition. They understand those metrics from a balance sheet perspective, or whatever it is. But as you think about translating factors into sort of actionable metrics for a fundamental manager to understand and think about how it impacts their portfolio, how do you sort of think about the factors who So

Omer Cedar  20:00

we know that there is essentially a lot of overlap between these factors. And depending on your choice of factor set that you start with, you may end up with one outcome with a set of sort of factors that you’d like to go with. And if you start with another set, you’ll end up with another outcome. And you have to deal with a lot of the sort of the correlation and a lot of the analysis there. Our approach, in particular with fundamental managers is to help them first and foremost, understand that there is a de facto standard in the market when risk sort of you like it or not, you have a Microsoft and Google so to speak of risk management. In terms of a factor model, you have Bara, you have, for example, axiom and those two are used today by over probably 6000, large asset managers, sovereign wealth funds, pension funds, as a framework to help them analyze risk in the market and make decisions on the basis of that. And by virtue of that so many other people are using it, it does actually the money flows do move in that direction. And you do see that you do see money flows follow the momentum factor for axiom. And so ultimately, then, if you want to come up with your own definition, that’s great. But that would be considered alpha, what you probably want to do is create a baseline of what you think is risk. And again, using these sort of de facto models, you have a starting point. And then it’s really a question of, what do you identify with in terms of your fundamental experience, and that’s really where the factors who can be filtered heavily to the areas that you care about. So for example, the factors who will have industry specific factors, it might have drilling rates that someone has sort of designed and an oil and gas name, you might have r&d to sales might be a better metric, when you’re thinking about technology companies than if you’re thinking about sort of banks. And if you’re a TMT manager, again, your fundamental manager usually tend to tilt more to some kind of an industry or some kind of a business model bias, you’re going to identify a lot more with those fundamental metrics, as you mentioned, quarry and those fundamental metrics, then you can overlay on top of the traditional model to have a baseline model like borrower X, Yama can be considered to be more alpha or more identifiable with your type of a strategy. And you can use them in both helping you manage risk, if you think that there’s a lot of people piled into the same ideas, or you can use it as actually an exposure that you’d want to have more of an alpha exposure that you want to have more of, and you want to tilt your portfolio to.

Corey Hoffstein  22:14

So let’s say I am an active manager, and I come to you with this pre baked book of 30 high conviction names and I started on the standard, I’m just going to equally weighed my high conviction names a little more my lower conviction names a little less, in your experience, even before you look at my book, having worked with so many fundamental managers and done this in the past, what would you expect the low hanging fruit to be for optimization either from a return enhancing perspective or a risk management perspective?

Omer Cedar  22:45

Fantastic question. We actually have an algorithm that we’ve designed, we call it focus on your alpha. And that is inherently the number one algorithm that we find that ultimately applies some kind of what I’m going to call a factor slash idiosyncratic optimization that works, I would say 70 to 80% of the time on a fundamental portfolio, fundamental portfolios are inherently constructed in a very similar way. They tend to be bottoms up driven, they tend to be very thoughtful in a particular area. And they usually tend to take bets in areas that the quants are not invested in. But on the other hand, they tend to have the sort of unintended of a will exposures in areas of quants are heavily invested in, for example, they tend to take on more risk, they tend to take on a lot more excess risk, because they think that that’s going to generate excess returns. So they’re going to look for companies where the risk reward from a fundamental perspective looks very exciting. But what actually happens there is that some of that excess risk can actually be explained by a factor that quants know is called the idiosyncratic volatility factor. And for those of you that may not be familiar with that, essentially, when you take volatility and you look historically, at volatility of a stock, you can say that some of the volatility can be described by its beta, the component that of its beta to the market. So yeah, it’s an energy stock, it’s gonna be more volatile. It’s an industrial stock, maybe a little bit less volatile. But then there’s a component with the volatility that is not explained by its beta. And that is called idiosyncratic volatility. And interestingly enough, that volatility factor alone represents a lot of the issues that fundamental managers sort of run into because they’re taking on this excess risk. But that idiosyncratic volatility that you don’t need to compensated for is where quants tend to spend a lot of the time shorting most quantitative especially sovereign wealth funds, pension funds, think of those sorts of the guys out there who are always looking for a stable low vol portfolio are actually shorting or underweighting, that factor in the market, and that creates a massive amount of flows. In fact, this is one of the most common factors that we’ve seen with a permanent Alpha structure, a negative alpha structure, where essentially it’s always returning negatively where high vol, stocks underperform low vol stocks, people have asked, what is that going to turn around? Well, when the sovereign wealth funds and pension funds decide if they want to invest in higher volatility names, but ultimately until that happens, you With this juggernaut, a volume that goes against the fundamental manager. So this focus on your alpha strategy, kind of pulling that back to that basically allows you to get rid of some of these unintended bets, right? That involve volatility and beta and other things that the managers just not aware of, and help them sort of focus on the alpha component of their portfolio, we run these again and again, across fundamental manager portfolios. And we find very similarly, those characteristics hold true, hence why we kind of call it sort of focus on your alpha.

Corey Hoffstein  25:30

So after evaluating such a large swath of different fundamental managers through your factor lens, have there been any surprising results that have emerged from this analysis?

Omer Cedar  25:43

I’m always surprised every day when I see because while there’s certain commonalities, there are so many different kind of what I’m going to call nuances in the fundamental strategies. And that’s actually what excites me about the continuum of working with fundamental managers, in order to actually build a niche in this market, you need to have an understanding of some kind of a corner of the market better than other people, you need to have that sort of edge. And that edge comes from a variety of different areas, you can have a good research edge, you can have a strong sort of, obviously an information edge and analytical edge. But ultimately, a lot of the times the edge comes from some kind of a nuanced understanding of a particular set of securities, whether it’s, again, it’s a shift in the political environment, there’s some kind of an undercurrent in lifestyle choices by consumer, there is a difference in the way that the underlying business models are changing. And the way that there may be deploying capital, all these types of things that they recognize are different ways of sort of investing in pockets in this very, very large ecosystem of global equities. And ultimately, that I would say is the part that continuously keeps me interested in surprises. Because really, when you think you’ve figured out what the fundamental strategy looks like, yet, there’s another manager out there who’s come up with a different view. And it’s another nuanced view that you can learn about.

Corey Hoffstein  26:58

So when it comes to these idiosyncratic bets, which you’ve expressed as sort of the right place for a manager to be focusing, finding their alpha, what’s the best way to think about measuring that idiosyncratic bet? When you have all these factors, you could try to say, do it via return regression, or you could try to extract out characteristic tilts within the portfolio that overlap. Or you could even construct these factor long shorts and look at holdings overlap in these mimicking portfolios. There’s all these sorts of different ways of maybe trying to extract idiosyncratic bets, how do you sort of think about the best way to identify those in an active managers portfolio?

Omer Cedar  27:34

I think that many of the approaches that you mentioned are very relevant. I think that the real question about isolating the idiosyncratic bet is, I think we sort of take a step back, we want to ask a couple of questions about what we’re trying to actually achieve from that extraction. Number one is, is this a piece of information that the manager is going to continuously try to either tilt their portfolio to or try to actually target when they’re filtering for ideas? So is this going to drive idea generation? Is this going to drive sort of more of a risk kind of a rebalancing, it could be both it could be driven sort of across multiple quants tend to have what they call an idiosyncratic return that we try to target, we define what we think that idiosyncratic return is, again, based on some components of the underlying return and some factors that we want to have access to, and some factors that we don’t have access to. And so ultimately, you can do that you can construct that ultimately that way. For fundamental managers, we tend to find that first and foremost, they actually do have really strong value in selecting sectors. So it turns out that most factors out there that people talk about sector factors, and the fact that making sector bets can be random, and there’s no reason to really try to tell two sectors, it’s also low Sharpe activity, fundamental managers actually tend to do that really well. So when we define idiosyncratic returns, Cory, we think about ultimately, the gross return of the security, we think about the fact that you probably don’t want to be making a bet on the market, you probably don’t want to be making most likely unless you are sort of a country type of a specialist bet on countries or currencies, those are not really the areas that you’re interested in. But when it comes to sectors, you might actually consider that that would be something interesting that you want to keep in as sort of the component of return. And then there’s sort of other components around style factors. And I think you sort of mentioned these ideas and what we call back the portfolios, I have momentum, I have value, I have growth, I have volatility, and how do I think about all of those sort of in that context? Well, typically, as we probably know, and some of you might really would have held barred external bills, they’re sort of factor models. All these factors get put into a multiple regression, but you can actually pull out of them something called a factor mimicking portfolio and a factor mimicking portfolios. Essentially, it’s a long short portfolio, right? equity market neutral where the portfolio itself where you’re tilting to essentially an exposure that gives you a unit exposure, literally a z score of one if we think about that to the underlying factor, ie momentum and zero simultaneously to all other factors. And that portfolio is obviously created through some I’m kind of an optimizer. And by doing that, then you can actually put in into your factor Zoo. As we’ve discussed, you can select 10 or 12 factors, put it in the ones that you believe represent all the major styles that quant sort of invest into, that you don’t want access to. And you can create essentially factor making portfolios across those sort of 12 factors. And those factor returns that you measure that on a given day through those portfolios would be subtracted from the securities return in order for you to isolate the return component that you really, really care about that does not have those style factors. And then that obviously, can be translated into a risk into a variance estimate into a covariance estimate across the board. And then you can essentially drive portfolio construction in that fashion.

Corey Hoffstein  30:41

So I can think of sort of persistent idiosyncratic risks for a given company and to your point, like a sector exposure for a given company would tend to be persistent and something you could model pretty easily and with a high degree of accuracy, but what about event driven idiosyncratic risks? So either like a takeover or the whole biotech sector comes to mind, for example, is one that is almost entirely event driven, idiosyncratic? How do you think about capturing that aspect of idiosyncratic risk?

Omer Cedar  31:11

That’s a great question. Actually, during my time at two sigma, we had to sort of think through those problems. And so when you think about an event driven context, it’s a very complicated problem, because you essentially have a component of the company that is trading on the basis of the rest of the market, like a peer group, you sort of look at it, you say, This is what the markets doing. This is what the peer group is doing. This is how I’m going to think about this company. And then there’s another component of the company that is trading based on these event driven investors, were saying, But wait a second, I think there’s going to be a very, very big event coming up in the next three months, six months, nine months, whatever it is, based on some FDA approval, or I think this is a takeover target, there’s been a lot of expressed interest there appears of being taken out, I think this is an interesting company, how do you balance those two essentially flows because you essentially have investors that are kind of pushing flows in one direction that are saying, this is a company that I would expect to trade with its peer group. And then you have another set of investors that think this is a company that should be trading based on a completely different set of factors. And ultimately, one way to manage it is through some kind of a probability function, that there’s a probability distribution between two states of the world, you have a state of the world where it’s purely idiosyncratic, and it’s running in this sort of event driven binary analysis. And so you end up having too much more of a tail risk function that you have to model. So it’s sort of what’s the upside, what’s the downside, and then you’ve got a much more sort of a traditional, what I would call normal distribution, where this is really where the factors come in, you have essentially a z score distribution, and you assume that its returns in the way that it sort of behave is going to look a lot like its peer group, and the law of large numbers will hold there. Now, ultimately, this is kind of where some of the art and the science comes in. As an expert in this area, you might assign a probability of the binary outcome to be 80%. And your peers might assign it to be 20%. But ultimately, that’s kind of where I don’t believe there has been consensus in the market about how to sort of charge a normal probability associated with it. Certainly, you can try to sample it based on the returns and see how much of the returns historically, the realized returns have been idiosyncratic versus being sort of more factor driven. But ultimately, the idea there is that that’s kind of what you need to do is you need to come into the world thinking I’ve got two models here. One is binary one is much more normalized, and I need to come up with a probability distribution between the two of them. So one of

Corey Hoffstein  33:25

the conversations I’ve been having a lot recently with folks is the idea of like, the difference between themes, factors, how you can get these thematic factors that emerge, the sort of example that keeps coming up is the COVID-19 factor, which seems to sort of have an overlap with quality junk with a focus on strong balance sheets and companies that aren’t as levered as others, but also seems to have this technology bits versus physical aspect to it. And the question that keeps coming up as well, is this a theme? Is it a factor? And to you, I would ask if a fundamental manager wants to take a bet, either on one side or the other of this factor to you, is that something that’s a factor? Is it idiosyncratic? Or is it entirely something different?

Omer Cedar  34:08

That’s a great question, because there’s actually a continuum in my mind here of this sort of trend or early emergence of a trend is typically picked up by fundamental managers, they typically understand that there is something sort of changing in the marketplace that is taking place that has not happened in the past, this sort of COVID-19 experience, fundamental managers are picking up on this obviously, in the January and February timeframe. And we’re starting to sort of either position their portfolios or find identifying candidates as sort of companies that would be exposed positively or negatively to the sort of this effect. Now, initially, that doesn’t feel much like a factor because you have a lot of idiosyncratic movements in the securities that are sort of representing managers beliefs, that certain companies are better positioned to this sort of exogenous if you will trend that’s happening, that might be sort of ephemeral. It might go away in a month. If they’re too, but over time, if this becomes more persistent, and people are starting to realize that this sort of trend is there to stay, and it’s becomes much more of a somatic type of a trend that people look at, then I would say that all of a sudden quants can start capturing that a lot more in the form of a factor. And why because all of a sudden, there’s a lot more sort of alignment between many securities that are starting to behave alongside this type of a factor. So when positive COVID news comes out, when things are getting better, certain companies rally, when negative COVID news comes out things people things are getting worse, certain companies rally. And if you can tie those sorts of type of movements, to a set of companies, large enough set of companies, we’re talking 500 to 1000 companies out there, then I think you can actually construct essentially a factor portfolio from that. And that’s obviously when we’ve seen the delta one desks the sell side, we’ll get into it by creating their own factor portfolios to try to represent the companies to allow you just sort of play if you will, the theme or the strategy behind that. And that’s really when the fundamental manager’s job becomes a little bit more complicated, because now they’ve identified all these names that are positively or negatively exposed, but they want to make sure that they still retain the idiosyncratic nature of their bets, and that they’re not getting swung around by the COVID, 19, or whatever, that thematic factor that’s now being invested by retail investors, and by you know, sovereign wealth funds and by quant funds. And ultimately, what they need to do. I mean, talking about the factor zoos TAKE THAT COVID-19 factor a factor portfolio, and essentially overlay it on top of their portfolio and see if the bets that they’re making in their names are highly correlated, and highly exposed to that COVID-19 factor portfolio. And if it is, then now they understand that a lot of the added value that they’re providing is already been baked into the market, a lot of it is already there, than they really need to either tilt their portfolio in a different directions or potentially exit their ideas altogether.

Corey Hoffstein  36:51

We spent quite a bit of time talking about maybe how fundamental managers can exploit or utilize the lessons that quantitative managers have learned over time. But is there a lesson that quantitative managers can be learning hear in ways that they can better be building their book things that they could learn from fundamental managers?

Omer Cedar  37:12

Quantitative managers are always studying investor behavior? I mean, this is one of the most core scientific principles behind quants, right is to study how investors behave. And to understand whether the new investor behavior is sort of a change to the environment or something completely diff, is this just a new form of value that people are looking at? or is this really truly a new trend? And I think actually one of the areas that quants can sort of learn from fundamental managers is how they study the business models and how they kind of understand that the changes in the drivers and the economy, zero rate environment, we’re sitting right now with a very strange sort of environment where people obviously working from home and are sort of doing almost everything online. And it’s creating a change, obviously, in lifestyles, changing consumer behavior, all of those things are things that fundamental managers are actually analyzing, and ultimately, how we quants Are you sort of going to learn from them and be able to invest in those sorts of types of trends will ultimately what we want to see are maybe companies that are taking sort of certain leadership positions in this sort of new market environment that are highly idiosyncratic, that are not being explained by a lot of the factors. And then we want to be able to try to tie them together, because there’s ultimately what we call as quants, X Factor correlation between those companies, there is a chance that those idiosyncratic behaviors that we see that we chalked to just ADL return are actually an actual factor. There is actually something out there that’s happening and it’s changing. One example, I would say that’s even more broad based to just COVID-19 is value. Everyone knows value hasn’t really worked. But why hasn’t it really worked? Value hasn’t worked in many cases, because the business models of companies that expose value are different today than they were 1015 years ago. And if you look at the top companies in value today, look at top companies in growth today, and you started looking at their business models, you understand that? Honestly, there’s a lot more robustness and the growth based business model that there is in the value based business model. And that wasn’t the same, there was a lot more of a commonality between value and growth in terms of the business model. It just happened to be more around valuation, again, pre 2008. So what I would say is, is that these are the kinds of things that fundamental managers publish about the study they write about. And as quants, what we’re trying to really understand is, is there a big change in behavior? Is there a real change in the community of investors and how they’re looking at securities, how they’re analyzing companies, and if that’s the case, then we’re really are looking for a new factor to describe that new environment.

Corey Hoffstein  39:34

So where is all this going? In your mind? We’ve seen a huge upswing in the adoption of quantitative and sort of quant light strategies, a lot of firms throwing in the towel on traditional active management and buying Smart beta sort of low costs, structured quant type products. We’ve also seen traditional discretionary long short managers seem to have really struggled over the last decade when you look out over the next five to seven years, what do you think the investment landscape is going to look like?

Omer Cedar  40:06

Always a scary question to try to make that prediction. But I firmly believe that in efficient markets, even though they’re constantly markets that are striving to become more efficient, people are going to find the winning strategy that is able to extract the most amount of alpha at the moment that alpha sort of goes away, and people are struggling to generate it, they’re going to look at other sources of alpha. And ultimately, I believe that again, going back to what was essentially a cottage industry back in 2008, when I started this two sigma, the combination, the deep combination of fundamental ideas with quantitative signals, factor construction, and ultimately portfolio construction and risk management will likely wite out over the long term. The question is, how long it will take and what will fundamental managers do in the interim? Will they kind of band together and create kind of a centralized quant strategy? Will they employ more quants to help them sort of turn the strategies into more quantitative, I think it’s going to be a combination of many things, right, there isn’t going to be one winning approach. Ultimately, tools will be used by certain ones that are going to want to involve more quantitative culture in their underlying strategy. And others are going to say, You know what, I’m a good stock picker. I’m a really good fundamental analyst, I think I have really strong ideas. Let me partner with a quant organization to put together a much better strategy altogether. But it will have to probably be the hybrid of the two in order to succeed in this marketplace.

Corey Hoffstein  41:27

So last question for you. 2020 has been a bit of a strange year and continues to be a strange year for a lot of people perhaps not the happy and uplifting year, we were all hoping it would be at the turn. But looking forward, what are you excited about either personally or professionally? What are you looking forward to?

Omer Cedar  41:45

Wow, I’m, this is kind of an interesting question. Because when it comes to mind is always simplicity, right? The ability to sort of take a nice, short, quiet hike somewhere and have sort of a relaxing time with the family is always comes to mind when it comes to these things. I guess what I’m looking forward to professionally at this stage is that there’s going to be more people sort of realizing that there needs to be a change in the way the act of management is done, and more people banding together. I think that this has been talked about for many years, I think it’s been tried in different ways. I mean, again, don’t let a crisis go to waste, use the fact that there really is a change in sort of the way that people think about the world to allow people to say professionally as a career, it is an investment manager, I really want to do something different. I’m not going to just be stuck doing the same thing I’ve done for the last 10 years and just continue to sort of do what I’ve done before, I’d like to look at another picture of what the future would look like. And by even asking that question, more and more people will start getting into this sort of mode of is there another way of investing? Is there another way of combining these two? Is there another way of thinking about it? And I hope that that conversation is really what drives a change in the way that this industry is managed and hopefully all for the better.

Corey Hoffstein  42:59

Homer cedar a fascinating conversation of definitely a very different perspective. I can’t thank you enough for joining me.

Omer Cedar  43:06

Absolutely. Cory, thank you so much for your time.