Recent Trends in High Frequency Trading (Christina Qi)

good morning everyone welcome to the time summit this is the fourth year bridge Alternatives has hosted the summit and each year we try to create an event that is timely thought-provoking and challenging and I think this year's content that Ryan Duncan brian walls and Chris Kennedy have put together does all the things that you will see over the next two days first a quick update on bridge alternatives we celebrated our three-year anniversary late last year and currently have offices in Chicago New York and Geneva Switzerland we recently expanded our team as many of you have noticed with the hiring of Molly deeds akin our Chicago office we've also continued to build out our counterparties who as many of you know are Wells Fargo EDF Mann Royal Bank of Canada and a recent addition of ADM which now gives us an outlet for FX prime brokerage if anyone has any questions regarding our clearing capabilities please feel free to talk to the team members now a brief tour of the space we'll be using over the next two days the one-on-one meetings will take place just down the hallway or the other side of this wall all the managers and sponsors have private booths and I've arranged them in alphabetical order do keep in mind the scheduling app is still open and you can schedule meetings as you meet people during the course of this event there are directions that will be occasionally up on the screen here lunch will be served out in the hallway just outside if you have any meetings during launch feel free to bring food into the one-on-one space we will have people kind of circling picking up plates so it's not gonna get a little messy tonight's cocktails and dinner will be here in this room and right outside directly after dinner we'll be having a special guest speaker Jaron Lanier who will be giving a presentation of the dawn of the new everything I hope everybody will be able to join us lastly thanks to this year's returning sponsors Chicago Mercantile Exchange EDF hedge facts and Barclays hedge and to our new sponsors dynamic capital coinbase bit go proof sweet and Omni acts and a special thank you to our strategic partner Wells Fargo who has graciously supported this event over the years and Alister mcbarnett managing director of Fargo who's going to help us open it the summit Thank You Keith good morning everyone some of you I know and some obviously I don't for those who don't know me my name is Alistair mcbarnett I'm the head of clearing cells within Wells Fargo's STM and I also run the execution business so as clearing child's head I face off to hedge fund CTAs asset managers and the investors into that space and as execution had we run both the voice services and the outdrawing execution services for the business my team here today I have John Hyde from the clearing sales team and I have Steve Chavez from my electronic execution team this is our fourth year sports in the conference and our partners at bridge alternatives asked me to make a few comments they actually asked my boss to make a few comments and then he ditched and then he asked me to make a few comments on the conference and why Wells participates in this and I thought I'd do that by asking and answering two questions myself the first question was why why be involved with a conference like this and the the answer sort of for me was kind of more related to you look at the current market conditions you look at a rising rate environment for the first time in a decade you look at political uncertainty in the world you look at the rise of crypto currencies replacing fiat currencies there is a lot of change and to understand change you need tools and the one of the best tools we could think of is education and so bridge and their partners do a very good job of educating in our space so Wells Fargo is very comfortable and very proactive in sponsoring events like this because we feel it helps make our industry better then the second question we always went into is why does Wells Fargo like hedge funds and CTS do you think of Wells Fargo with a big retail bank who also has a Marcus division and I thought a few stats would probably sort of help solidify that everything about wealth Channel we have two point one trillion in customer assets wealth channel our customers in that wealth channel allocate to alternative investments we have billions of dollars of our customers assets allocated into alternative investments some of which with people in this room we ourselves as an FCM have 480 customers and about 11 billion in assets of which quite a larger proportion is from hedge funds and CTAs you're an important customer base for Wells Fargo and your important customer base for us as an FCM in the space so the reason we're involved with hedge funds and CTA s is because you're important clients of Wells Fargo and we want to continue to grow our business and your businesses together so with that I will hand you back to Chris who I think is moderating the first session thank you good morning everyone so today's session concerns innovation and I think this is a fascinating topic it's something we've had a ton of fun with your bridge but the word innovation I think is it's a word that I think is so commonly associated with with like Silicon Valley right when I think of innovation I think of Elon Musk and Steve Jobs and these you know going to Mars these big visions of the future right and I don't hear those kinds of things coming from the hedge fund space these days you know and I kind of was wondering why right I think one of the reasons is we don't want to do this I I think clearly there's some long-term visions there that didn't come true and I think in the hedge fund space you know we're very careful predictors very careful predictors and it's just what we do is a trade and I think it affects really the way that we picture the future but I think there's still something worth doing here right I think talking about innovation and thinking in these long term grandiose visions helps answer this question of you know what's next right and I think now maybe more than ever in the hedge fund business we want to start answering questions about what's next so today for us you know we really want to unpack innovation in something that looks a little bit like this right I think innovation can be thought of along these dimensions Alfa certainly kind of the incentive for innovation we want to touch on talent and technology certainly mediums of innovation competition right this is the context of the innovation that we're doing it's why we're innovating we're thinking about our competition and then also secular change right these are the results of innovation - and today you know what you're looking at here in this chart is liquidations and launches in the hedge fund space and so again I think when I look at this chart I very quickly think what's next you know you clearly have a different environment today than we've ever had and I think innovation is again kind of the medium for which we can start telling the stories that give investors participants in the space and idea of maybe what is on the forefront right if you think about a venture capitalist they can tell you their view ten years out or whatever industry they think is useful I just don't think we talked about those kind of things as much in this corner of the world so to help us with this we've invited two completely different firms to kind of act as little case studies for us right to broaden our horizons take this predominantly systemic managed futures crowd and start to think about what's going on in all corners of the investment landscape the first is Christina Keyes she runs Domeier Capitol dome Aird is an investable hft firm in Russia gonti runs orthicon Partners which is a private assets firm these two groups operate on completely different ends of the liquidity spectrum christina typically looks for the world's most liquid markets and reefs she seeks to avoid markets altogether he'll explain what that means later but I think even though their investment universes are extremely different I think these groups are both trying to do to really answer what's next in their own individual ways and again I think they provide excellent case studies for impacting what we're doing along those different dimensions so with that I'd love to welcome Christina [Applause] awesome thank you Chris for the introduction that was really nice so basically I'm a you know it's a privilege to be here and I'm happy to start off this session by giving you guys an update on kind of fee the trends and the you know basically what's happening in the high frequency trading space I know that not all of you guys are allowed to even allocate into hft or or even to quant for that matter but I think it's interesting it's an interesting learning experience to kind of understand kind of what's happening in in our corner here and what's actually going on because you hear all kinds of articles and things coming out like saying stuff like hft is dead so I'll talk about that as well and talk about kind of what's going on from a practitioners standpoint so it's let's go on here so first a quick disclaimer you know don't don't buy and sell stuff based on what I'm saying and if you lose money it's not my fault that's basically basically what this says so we are not trying to sell you anything so as a briefing I'll start off with a brief introduction first and then I'll talk about the current hft environment what is actually happening to all these companies why is everyone merging talk about the evolving talent pool because that's the future generation of the workforce in in our space and then also talk about the effect that the articles and the news has on high frequency trading startups like us and then also talk about any other insights that we might have so I'm gonna keep this introduction very brief so I'm Christina Chee I think you can probably just either look us up or you can add me on LinkedIn if you'd like but you know I basically I started Dome yard out of a dorm room in college about six years ago now I think and I graduated about five years ago and then since then I've been working at the same company since then basically and we decided you know from the very beginning let's focus on high frequency trading and let's also become let's do a hedge fund structure from the onset and so those are some you know to kind of what do you call it when they kind of don't really belong but we did that anyway and ended up you know becoming you know one of the fastest-growing hft firms out there over the last couple of years or so so we've been very fortunate to to still be alive today and to be up and running as a team so yeah we were located in Boston the average age I believe is now 37 on the team people always think that were you know a bunch of kids running a fund but in reality you know we do have some gray hairs on the team as well investors are mainly family offices and high net worth individuals because it's where capacity constrained fund and we have some great you know of course some great people on the team so for me the biggest privilege has always been to work with some group of very smart people who are much smarter than myself on the team so I I'm gonna move on to the next slide I think the best way to kind of show what we do is through photos as well so lots of photos included here but we're very tech intensive you know we're in Boston but our servers you know that we trade from they are co-located and you know it's a caucus in Aurora and all the different you know where all the servers are supposed to be so anyway let's move on here so a lot of our investors will come up to us and they'll you know they'll show us these articles about hft you probably some of you have probably already read them about you know - zoo trading is dead it's gone someone wrote about you know there's what's happening to the flash boys you know are they gonna go away is you know Virtue's acquiring virtu acquired kcg right and that's those two of the largest firms out there now you know they're in there one company so there's a lot of acquisitions happening and and people always ask well how does that actually what's actually happening out there you know are we actually dead and how does it affect a small company sometimes I do wish that hft was kind of dead because that would make hiring really easily really easy for us right we could just hire the best talent out there all the heads of hft from all the companies but unfortunately that's not the case what we actually see today's the hiring is more competitive than ever they're being offered salaries that are you know these people are making millions of dollars per year working at these firms right out of college and there some of these companies they just got a new office you know these are signs of growth rather than signs of failure so we'll talk about that a little bit later but anyway I just wanted to basically you know this part of the presentation is sharing with you kind of some of the changes that we've actually seen in the past you know five or six years or so while from when we had started the company all the way until today so there's been even in the past five years like a ton of different changes so the first thing that I want to talk about is one of the challenges that we face starting off which was basically one of the you know challenges about - - trading in general is that the data sets are very large right so we're dealing with potentially petabytes of data every day let's say in the New York Stock Exchange alone right you get like over 300 million messages within just an hour and 300 million messages here that you know for a computer to process it takes a lot of computing power so you can't just fit that into the mainframe of a you know a regular computer you need some specialized hardware and software to kind of make that work and so basically you know in order to do that in order to process data coming in so quickly at a high speed you need high speed systems and thus high frequency trading is born that's the whole the whole thesis of kind of what we're doing is that you know in order to take advantage of all that data coming in every single day and to be able to trade it on you know on a microscopic level like let's say you know the data is coming in at nanosecond resolution here it's really really fast and in order to take advantage of all that data and and not overlook some really important data points you need to be fast you need to be able to process that within the same hour be able to generate some really good signals and then be able to trade that really quickly so that's why we we exist you know as a company now when we were starting off though you know we had to create our data feed handlers right so we had to be able to basically you know take all that data coming in and process that using our internal feed handlers and so first of course what we do is you know when you're starting off you look online you see what are what are the open source feed handlers that are out there can I use anything or maybe go you know go to another vendor and and buy a food handler vendor well so five years ago that wasn't quite possible because I think there are only two open-source you know fixed engines out there that were available to us at the time so we were pretty much you know we didn't have an option we were pretty much forced to create our entire technology stack internally in-house we had to hire a bunch of developers and create you know our order management system our feed handlers all the major tools of a trading ecosystem back then so today you know we could do I'll give you an example how that's changed the open-source community has developed a lot in recent years actually so you know we were able to hire we had a bunch of interns you know we work with these universities now which has been really nice for us because of the open source community you know these we worked with MIT last year and basically what happened was they asked us hey can can we put some interns you know into your company and they can work on some projects the only constraint is you know because these are these are current MIT you know PhDs and master's students they're not allowed to sign you can't send a non-compete you can't sign an IP agreement you know they whatever they use they can just take to wherever they go next and so that was a challenge for us because we couldn't share with them our codebase but what we could do with these students was we said okay well why don't you go online look at all the open-source tools out there and you can do a project where you detect you know anomalies and market data so what they would do is they would look for data they basically parse data from like 14 different exchanges and they look for anomalies like spoofing or spamming or other sources of like market manipulation and be able to detect write a strategy that kind of detects that within a shorter time frame and so these students you know obviously a lot of them they have an academic background so they've never seen an order book before in their lives they don't know they don't have any just like when I was starting we had no experience so they had absolutely no experience and within one month they were able to do they process literally a terabyte of data on you know on one cut alone one laptop with four of them and they were able to you know from 30 different exchanges all this you know a ton of data and then be able to basically complete this project and search for all the anomalies that they were looking for using this open source these open-source tools that they had available to them so you know this would not have been possible like four or five years ago but today you know there's something that because how fast the innovation is coming up and you know I guess I the second photo I have is just some of the open-source tools that we see online so you guys can always look that up too it's fascinating how quickly it's developed in recent years so the other thing that we needed when we were starting up was data obviously so back then we were looking around at you know you look at the data vendors that are available you go to like the FIA Expo and the trading show and all those other major Expos out there and and you talk to all the vendors and and what we noticed was you know the vendors would try and sell us you know millisecond resolution pretty lossy data which in in high-frequency trading you know we need data at at least like two one or two orders of magnitude greater you know basically if it's microsecond or nanosecond resolution you know that's great and so so the data wasn't that great and it was it was actually really funny because that was the first time I felt like a celebrity by the way was I think I think we were at the FIA Expo when you know I was I think I just graduated from college actually and a bunch of people you know people were lining up to talk to me because they knew we were in hft firm we were like the ideal client base you know for data and he was one of these I was like how did you hear about it you know we literally had just started out of college and yet you know that these data they're so there are more data vendors than high-frequency trading firms you know about that time at least and so so anyway that you know so that was really frustrating for us and then the other thing was that you know we so we decided okay well why don't we process the data internally and just grab data directly from the exchange because it's cheaper that way so we just literally get data directly from the exchange we process it ourselves and then you know trade from that and other companies then would start coming up to us being like hi can we buy her data from you and that was also really bizarre because we had just started off to begin with and so today it's completely different as well you know the data now you can get literally nanosecond resolution raw data from from a lot of different vendor and it's much higher quality than it used to be so so that's definitely one of the other changes we've seen is just the data has because of the big data movement actually back then you know no one knew what big data was and today it's a huge buzz word because of the interest and the new talent coming in and all of the and also of the open source and everything combined you know there's a lot more competition which means that the data coming in is a lot higher resolution a lot higher quality than it used to be now the other thing that was a you know that we've noticed was the kind of abundance of microwave towers that have popped up in recent years so I was out of farm in Massachusetts and I don't know how closely you can see in this photo there's a microwave tower there so you know like regular normal people take photos of like you know food and like shoes and I don't know cars I take photos at microwave towers whenever I see one so there's a oh you can barely see it on the on the right side there's like a little pixel on the cloud you know on the back of the cloud there's like a little pixel there that's the microwave tower right there it's really small so you don't usually don't notice it unless you really look for it but I was like just at a random farm and I saw one I was like wow you know because these things didn't exist like 10 years ago so you know 10 / 10 years ago actually I think it was Goldman Sachs and someone can correct me if I'm wrong I think it was Goldman Sachs that they spent like millions of dollars building the fastest microwave routes between you know New York and Chicago this cost of millions of dollars back then today we're very lucky in that a lot of the fastest not all the fastest routes but many of the fastest routes out there are owned by vendors rather than by by like a one you know by a virtue or a Casey you know by Citadel or whatever so that's really good for small companies like us and for anyone kind of starting up because all we need to do is we just pay a set fee you know let's say I don't know how much six thousand ten thousand dollars a month depending on you know which routes are buying and you basically you can compete against some of the largest firms out there in terms of speed right and people never could have imagined that you know you could do that and so so the stuff that they wrote in flash boys is a little bit you know it's become a little bit more outdated in that today any small company can just go out there and just literally go to a vendor and just buy a fast route so the other thing that people always talk about with us is you know how fast really how fast are you and also when is when is it too fast right like have we reached zero yet there's something people called a race to zero where people try to you know everyone's like oh you got to be as fast as possible your Layton sees need to be zero at the speed of light you know in order to to get the you know maybe be one nanosecond faster and so what we've noticed is that you know at a certain point in time it's just not worth it to be one nanosecond faster than your competitor now actually probably clarify actually you know faster is always better in any industry it doesn't matter almost anything's just so in finance you know in if that's a TN t--'s motto right faster is better so AT&T vs. Verizon if Verizon is a little bit faster you can bet that a lot of people are going to move over to the faster carrier right the same there's the same thing for finance actually faster you know of course if you get information at a faster rate you can take advantage of that in in certain ways that other competitors can't but there's there's a limit to that in that let's say if you have to spend an extra million dollars to be like a little fraction faster and you're not gonna earn that money back then then obviously it's not worth it to be you know a little bit faster and so for us we realized our limit in terms of you know how fast how fast is good and then at one point does the you know marginal benefit kind of decrease right and that actually the answer comes from the exchange it doesn't come from us you know because we could be as fast as we wanted to and that's great and other companies will race to try to beat us but the limit here is on the exchange side because the exchange is matching architecture hasn't changed at all well a little bit but it hasn't changed nearly as much as it should in the last you know five six years or so so I hear the graph right here is a late latency histogram of it's actually our trades during a day on the green the green basically on this and then the other end of the the histogram spectrum is the exchange the exchange is matching engine and this is this is on a log scale by the way it's not a linear scale to log scale so you can imagine how much slower the exchange is compared to you know basically the closer to the left you are the faster you are the closer to the right you are it means you're extremely slow so on the exchange side they're extremely slow you know when you send an order out to the exchange there's a lot of jitter that happens there's a lot of noise that happens basically and so you know once it reaches the exchange the exchange is so slow that it doesn't even matter if you're a man a second faster because there's so much variance also in terms of how fast your order actually gets matched so there's a lot of exterior lot of variants keep in mind it's a log scale so the tails of the you know the exchange side is really large a lot of fat tails on that end too so because of that because of how slow the exchanges are you know for us as a high-frequency trading firm it doesn't make sense to you know to spend all of our efforts trying to be a little bit faster than a competitor when it doesn't even matter you know too much once you're you know trade actually hits the exchange so I think our panel after this we're gonna discuss you know talk about the Big Data machine learning influence a little bit more because that's become those are big buzz words today I guess the one thing I wanted to just talk about right now is the data you know machine learnings been around for since the nineteenth century you know lynnie it's called linear regression back then now it's called machine learning which sounds cooler now you know in terms because of the big data influence though there's been a lot of improvements so one thing that I guess who'll be introduced oh you know people using the cloud these days right like AWS or other cloud providers I think back when we were starting we would talk to some people who be like oh we're we're a high-frequency trading company and and we're we're in the cloud or completely in the cloud and I would I'd be like no you're not you can't possibly be you know be a high from the Z trader in the cloud back then because it's so slow everything is extremely slow in the cloud you have to have proprietary servers and everything you know in the exchanges that are co-located you can't just be in the cloud but actually today surprisingly the you know there's been a lot of innovation on the cloud side of things and so today if we were to start up again we would actually seriously consider any you know implementations and use cases of actually being in the cloud so that's something that we're definitely we're definitely starting to you know look at so so then the other things are you know back I would say like 10 20 years ago when you know companies like Citadel started up in virtue they would you know they would basically focus on generalization rather than like specialization right so they would go out to all the different exchanges they would try to trade as many different products as they could it was a you know it's really fascinating basically what they're trying to do which is just pretty much everything under the Sun today as the industry is starting to mature right we're seeing companies acquiring each other they're figuring out where they're where to actually specialize in rather than just trying to do everything at once because people are starting to realize now you know we can start to see who the winners are in our industry now encrypt by the way in crypto currencies crypto was at that stage of hft maybe 10 years ago basically so it's at an early stage where we don't know who the winners are and you know there's I'm sure in the next couple of years we're gonna start seeing mergers and acquisitions in crypto as well and right now you know a lot of a lot of hedge funds too they're gonna just try to trade everything they can every product they can so it's like it's thoroughly like that's just the business cycle of any kind of innovative you know technological trading that we see so that's beginning to happen as well in terms of cost people will ask well you know isn't the cost isn't the barrier of entry really high in high-frequency trading and that is true it is very very high but I'll tell you about some of the costs that have increased and then also some costs that have decreased so in terms of increased costs in recent years one thing that has increased is it like the cost of data right like you we're seeing notice I can show you might have heard of that was acquired by I think was at Goldman Sachs that acquired them and then there's like Raven pack there's tons and tons of these data providers everywhere and that actually drives a competition up and it also it also means that there's a lot of different sources of data so it's not necessarily that the cost has gone up but rather that there's just so many different sources now so like if you don't know how to specialize as a startup fund then you're screwed you know if you want a kazoo there's literally people who are selling there so like satellite data now with parking lots so you can see you know how much business companies get there's there's also satellites that look at oil rigs you know I don't know if you've heard that story where they look at how high the oil rigs are and then during this certain part of day you know the Sun casts a shadow and based on the height of the oil rig you can see how much oil is in that country you know that's a huge you can see the actual imports and exports before it actually gets reported so there's there's all kinds of these crazy sources of data out there and and so it's just a matter of where you concentrate I feel like it's it's similar to like Airlines for instance like right wasn't it like back in the 80s I wasn't around back in the 80s by the way but I remember but people tell me back in the 80s there's like you know there's economy class and like first class maybe and now there's now when you're flying an airplane if you're flying internationally there's economy there's premium economy there's you know there's first class business class Premium business class there's like all these different classes you got to pay a little bit more and then if you want to be a group one board and go to pay a little bit more to you know be in the faster boarding lane that's kind of what it's like in data too so like every couple of years I think especially if you're in like the the cash FX space for instance if you're in certain parts of the industry you're trading certain things every couple of years these vendors they're gonna release like a new they're gonna read some more on granular source of data so it'll be a little bit more granular than what they released in the past and it's gonna cost you a premium and and so if you're driven by an informational advantage if you're fun like us where we rely on information to succeed you know faster data and things like that then and if all of our competitors are buying the more expensive feed then it's like oh crap we have to you know buy it - so it's just there's a lot of different ways to spend money on these different type of classes of products so we you know unfortunately we can't just stick with the economy class of data we gotta buy the premium you know the first class data out there in terms of things that are cheap read probably read it by now but um in terms of is technically cheaper in a sense by the way the cost of colocation has always been around the same I think but the what I mean by technically cheaper is that one server can now process a lot more than it used to so we can literally take up like little one little server space and process so much more so much more you know data computing power than than it used to so so technically in a sense it has become a little bit cheaper so anyway I want to talk about the talent pool a little bit which is a completely different topic because I feel like um you know in order to understand what's next in our industry you have to understand the the pipeline in terms of the future generation of the workforce in our space so what we ended up doing and we do this a lot is we analyze job applications like all the time so what we did here was we analyzed 20,000 you know quant research job applications for our our quant research role lose a little bit one job position here over the course of I believe it's over the course of one or two one and a half years or so and and so we'll see a little flick of some of the results basically of what happened so we look at things like keywords write on you know on things like this one is on MOOCs by the way so you guys know have you heard of Coursera or Khan Academy some of those other things where you know we're taking class online and you get like a you get a certificate by the end of the class so those rose in popularity and then quickly declined in popularity in recent years I don't know why but it's just something interesting data for us you know we were like wow you know it gues turns out you're you know the Harvard degree does maybe have a little bit more value than taking a class on the Internet these days so so we're looking at that and then we've seen you know there's buzzwords right like so obviously big data data science artificial intelligence you know those terms are on an uptrend they've been you know they're continuing to be it's the popular thing that people want to do and so we're starting to we're definitely seeing more and more of that in in recent times let me see what else is there we also looked at machine learning data science and statistics and we see a lot of you know everyone's studying machine learning and statistics data science not not as much in our space you may see more data scientists working for like Microsoft labs for Intel for IBM but not necessarily for our industry because the the traditional machine learning techniques that a data scientist might use may not be applicable to too high frequency trading actually you know I it's a completely different talk but I'm happy to you know talk about that if you guys have any questions afterwards so the next thing I want to talk about really fast before I end is just the effects on on - Quincy trading startups like how are how are we affected by all this stuff that we just talked about here so the first thing is that back then we were kind of forced to build everything ourselves which is great for us because now we have a ton of IP and we can kind of license that out to different companies and and so that's nice at the same time it was also very expensive very time consuming and to be honest I was living in hell I mean it was like you know we we had to raise and I'll talk about this during the the panel as well but we just something really weird which was we raised venture capital as well as so that we raised literally at GP round and then we raised an LP ground for the actual clients of the fund and that back then I think that was unheard of like no one does that but we had to because you know we were a technology firm on the inside and then on a business level we were a hedge fund so today you know it doesn't make any sense to reinvent the wheel right when there's so many different great tools that are available in the community out out there so so yeah that's definitely one thing that would change is we would probably buy versus like building the other thing is like especially I would say talking about like some of the older companies that are from some of the legacy competitors in our industry they would have literally 60 developers working on like just to you know support like a team of one little team of traders basically so the emphasis back then was on hiring a bunch of good devs to kind of create and build those tools while today because you can outsource because you can pay you know handful of other devs to do it you know you don't actually need to hire so many people so right now internally what we're doing at Dome yard is we're starting to increase the ratio of researchers to developer so that hopefully one day it'll be you know maybe a two-to-one ratio versus right now it's about a one to one ratio of researchers and developers so so the other thing is you know the industry is evolving so you know costs are becoming fairer there's a lot more transparency there's a lot more accessibility in our space than there used to be than ever before actually instead of focusing on how to differentiate ourselves on an infrastructure level so like instead of playing the the race to zero you know buying all the servers and and playing that hardware game that some companies still do today a lot of companies are starting to focus on research and business differentiation so how can we differ you know differ ourselves in the research pipeline how can we do more innovative strategies rather than just being just blunt force faster than everybody else right so those are things that we start to think about and then a couple of other things that I think we'll start to see in the next couple of years are things like better tools that'll speed up the research pipeline because right now the research pipeline is largely you hire someone you know as maybe a PhD and they'll spend a lot of time it's almost like running a research lab which is not a good thing by the way but they'll you know take some time to kind of come up with these ideas but if there are some tools to kind of automate that process I think that'd be really really interesting to see so those I think we'll start to see more of those in the next couple of years or so and then also a lot of those innovations we've seen in high-frequency trading are starting to trickle into the rest of the industry actually so you know I used to work on a discretionary trading desk even the discretionary traders that you know that I used to work with they're starting to use some of those tools like the data analysis tools where you know you don't have to know to code all you need to do is just use a software open it and run an analysis so you're starting to use a lot of those tools that that I'm like hey I used to you know I use that in high frequency trading and now they're starting to use that today so I think that's really cool how even the most discretionary traders are starting to adapt some of these technologies into their practice so I think that's it for my presentation you know thank you so much for doing this it's a huge privilege to be here I will be here for the rest of the days you're welcome to have any questions to come talk to me you can also reach out to me through LinkedIn if you go want to chat further about this so thank you guys so much it's a privilege to be here [Applause]

Channel: Bridge Alternatives

Views: 15,943

Rating: 4.7441077 out of 5

Keywords: Christina Qi, time Summit, Bridge Alternatives, High Frequency Trading

Id: JEE9ffnUz6w

Length: 37min 36sec (2256 seconds)

Published: Wed Jun 27 2018