WEBVTT - Paul Wilmott Has Some Feelings About Quantitative Models 0:00:02.200 --> 0:00:06.800 This is Masters in Business with Barry Ridholts on Bloomberg Radio. 0:00:09.080 --> 0:00:11.880 This week on the podcast, I have an extra special guest. 0:00:12.080 --> 0:00:15.640 His name is Paul Wilmot, and if you are remotely 0:00:15.840 --> 0:00:20.000 interested in anything having to do with quantitative finance, then 0:00:20.440 --> 0:00:23.959 this is the podcast for you. UH. He is not 0:00:24.079 --> 0:00:28.440 only a colleague in Pier of such esteemed quantz as 0:00:28.440 --> 0:00:31.680 a manual Derman and nasiem To lab who is really 0:00:31.720 --> 0:00:36.240 more of a pure mathematician, but he is the purveyor 0:00:36.440 --> 0:00:40.760 of the world's largest website on quantitative finance. He has 0:00:41.000 --> 0:00:47.239 written numerous books on UH, textbooks on quantitative finance. He 0:00:47.320 --> 0:00:53.320 has helped create the certificate UH for Finance UH quant Work. 0:00:53.520 --> 0:00:59.240 And he has just been absolutely on the forefront of 0:00:59.360 --> 0:01:05.080 identifying what's wrong with financial models, markets, derivatives, risk taking. 0:01:05.440 --> 0:01:08.440 There are there are a few people who understood why 0:01:09.400 --> 0:01:12.520 the financial crisis of oh an O nine was coming 0:01:13.360 --> 0:01:17.360 UH more specifically and earlier than he did. Eight years 0:01:17.360 --> 0:01:20.760 in advance. He was warning, Hey, these models are really 0:01:20.840 --> 0:01:24.960 problematic and and they're going to result in in big problems. Um, 0:01:25.000 --> 0:01:27.680 They're gonna result in real issues. And he turned out 0:01:27.680 --> 0:01:30.640 to be not only right generally, but what he was 0:01:30.920 --> 0:01:35.000 criticizing specifically, turned out to be a large part of 0:01:35.000 --> 0:01:39.440 of why markets and credit went went through its collapse. So, 0:01:39.560 --> 0:01:43.520 with no further ado, here is my conversation with Paul Wilmot. 0:01:47.880 --> 0:01:51.080 I have an extra special guest. His name is Paul Wilmot. 0:01:51.160 --> 0:01:54.880 How do I describe him? He is a expert in 0:01:55.000 --> 0:01:59.600 quantitative finance, a researcher and author, a consultant, the professor. 0:02:00.040 --> 0:02:04.040 He has written over a hundred research papers on mathematics 0:02:04.040 --> 0:02:09.800 and finance, several best selling and some would say groundbreaking textbooks. 0:02:09.919 --> 0:02:14.800 He runs what is the biggest quantitative analysis website in 0:02:14.840 --> 0:02:19.040 the world. Uh no, lesser a critic than seemed to 0:02:19.120 --> 0:02:23.400 Leb described him as the smartest quant in the world, 0:02:23.880 --> 0:02:27.720 the only one who truly understands what's going on, who 0:02:27.840 --> 0:02:31.639 uses his head and has a sense of ethics. Paul Wilmot, 0:02:31.680 --> 0:02:33.760 Welcome to Bloomberg. Hi, it's a place to be here. 0:02:33.800 --> 0:02:36.120 I have to correct you already. I have to correct you. 0:02:36.440 --> 0:02:39.520 I'm not, and never have been, a professor. I'm a 0:02:39.560 --> 0:02:42.840 mere doctor. I'm afraid were you a lecturer? I've been. 0:02:42.919 --> 0:02:45.840 I've had all sorts of research positions in universities, but 0:02:45.960 --> 0:02:48.880 never reached that exalted. I'll tell you why I drew 0:02:48.919 --> 0:02:52.400 that conclusion. There is a YouTube video of you describing 0:02:52.480 --> 0:02:55.400 some of the chapters of some of your textbooks, and 0:02:55.440 --> 0:02:58.760 you just have a natural professorial air. I have this thing, 0:02:59.560 --> 0:03:01.720 and I through the assumption that I perhaps should not 0:03:01.760 --> 0:03:06.560 have drawn. So let's talk a little bit about um, 0:03:06.680 --> 0:03:10.280 your research, your work you're writing. I have to start 0:03:10.320 --> 0:03:12.880 out with a quote of yours from one of your papers. 0:03:13.280 --> 0:03:17.280 The paper published in two thousand was the Use, Misuse 0:03:17.480 --> 0:03:21.760 and Abuse of Mathematics and Finance. That title alone is 0:03:22.080 --> 0:03:24.600 a good place to start. But here's the quote that 0:03:24.960 --> 0:03:28.560 was so prescient. It's clear that a major rethink is 0:03:28.639 --> 0:03:32.680 desperately required if the world is to avoid a mathematician 0:03:32.800 --> 0:03:38.440 led market meltdown. That is pretty much dead on. And 0:03:38.600 --> 0:03:41.160 seven years later we had the quant crash, and a 0:03:41.240 --> 0:03:45.080 year after that we had the full blown Great Financial Crisis. 0:03:45.720 --> 0:03:48.280 What were you looking at that gave you such a 0:03:48.320 --> 0:03:52.000 clear understanding of of the coming storm? Well, I also 0:03:52.240 --> 0:03:54.000 a few years after that, I wrote that I did 0:03:54.680 --> 0:03:56.880 specify that I thought it was going to be credit 0:03:56.920 --> 0:04:00.400 datives that might be the problem. So even more specific, yes, 0:04:00.440 --> 0:04:04.400 and that that's because of my background, I haven't taken 0:04:04.440 --> 0:04:09.280 the classical quant route, and I come with a lot 0:04:09.320 --> 0:04:13.160 of skepticism, and I'm not a great fan of following 0:04:13.200 --> 0:04:16.240 the herd. So whenever I look at mathematical models, I 0:04:16.279 --> 0:04:20.080 can pretty much tell you whether the based on any reality, 0:04:20.240 --> 0:04:23.719 or maybe whether there's any dangers involved with the the 0:04:23.760 --> 0:04:26.440 hedging or the valuation. And everywhere I looked, I just 0:04:26.480 --> 0:04:30.160 saw that there were dangers. It was not The models 0:04:30.160 --> 0:04:34.080 for equities are not too bad in quantitative fires. The 0:04:34.120 --> 0:04:38.039 models for interest rates are pretty bad, except interest rates 0:04:38.040 --> 0:04:41.719 don't really do much, certainly at the moment anyway. But 0:04:41.880 --> 0:04:45.479 credit the models are not only the models really really 0:04:46.080 --> 0:04:48.640 bad in terms of not matching reality and how they use, 0:04:48.960 --> 0:04:52.640 but also credit markets are absolute enormous, tremendous. In fact, 0:04:53.000 --> 0:04:56.880 that also from the same paper, the underlying assumptions of 0:04:56.960 --> 0:05:00.479 financial models, such as the importance of normal distribution, the 0:05:00.520 --> 0:05:06.520 elimination of risk, measurable correlations, etcetera, are all incorrect. And 0:05:06.520 --> 0:05:09.880 and when I read that, I was reminded of the 0:05:10.040 --> 0:05:13.400 famous George Box quote all models are wrong, but some 0:05:13.520 --> 0:05:16.960 are useful. And I know that's sort of a bastardization 0:05:17.080 --> 0:05:21.359 of of what he really was intended. But the natural 0:05:21.480 --> 0:05:24.800 quote next question is are all models wrong? And how 0:05:24.839 --> 0:05:27.440 can you tell the ones that are useful from the 0:05:27.440 --> 0:05:30.040 ones that are dangerous? Well, models in finance are going 0:05:30.080 --> 0:05:33.039 to be wrong, and it's a definition you can you 0:05:33.120 --> 0:05:37.200 can perfectly understand what's not told. Um. And as for 0:05:37.960 --> 0:05:40.000 the word useful, I think you have to ask what 0:05:40.160 --> 0:05:43.840 is meant by useful and the various ways of looking 0:05:43.839 --> 0:05:48.400 at this useful. Are they useful in helping you control risk? 0:05:48.600 --> 0:05:52.160 Are they helpful in allowing you to do more business? 0:05:52.680 --> 0:05:55.040 Or there are all sorts of different angles, and some 0:05:55.080 --> 0:05:58.159 of these are sort of conflict with what the man 0:05:58.160 --> 0:06:00.800 in the street might want. For example, a model might 0:06:00.839 --> 0:06:03.640 be useful because it's great for marketing, and it you know, 0:06:03.680 --> 0:06:07.240 it fools the regulators and it's not very good, but hey, 0:06:07.240 --> 0:06:09.160 it's it's useful if you want to try and start 0:06:09.200 --> 0:06:13.479 a really big hedge fund. Um. So it depends what 0:06:13.520 --> 0:06:15.760 you mean. I don't think that's what professor Box was 0:06:15.760 --> 0:06:20.440 referring exactly. However, exactly, they certainly have conserved useful purposes, 0:06:20.839 --> 0:06:25.800 albeit perhaps not accurately predicting what the universe may look 0:06:25.839 --> 0:06:28.160 like exactly. So you hope that when we talk about 0:06:28.320 --> 0:06:30.400 having a model. It's useful, you'd you'd like to think 0:06:30.400 --> 0:06:33.600 in terms of, oh, it allows us to manage our 0:06:33.720 --> 0:06:36.760 risk or hedge our risk something like that, or or 0:06:37.200 --> 0:06:43.040 produce important new products that are helpful in hedging your 0:06:43.080 --> 0:06:46.600 business or whatever. But I'm a bit too cynical to 0:06:47.160 --> 0:06:50.200 think of. So you said you didn't take the usual 0:06:50.279 --> 0:06:53.800 route into quantitative finance. How what was your route? How 0:06:53.800 --> 0:06:56.960 did you find your way into quant finance and how 0:06:57.440 --> 0:07:01.120 does that differ from the typical on Wall Street. Well, 0:07:01.600 --> 0:07:04.120 you have to go back to my my my childhood. Really, 0:07:04.839 --> 0:07:08.000 I've always from a very young age run my own 0:07:08.200 --> 0:07:13.240 businesses at various sizes, and even when I was a 0:07:13.280 --> 0:07:17.760 mathematician at university doing my doctorate or doing post doctoral work, 0:07:18.360 --> 0:07:22.280 I was always interacting with the real world, trying to 0:07:22.280 --> 0:07:26.080 do consultancy, etcetera. And it meant that I saw a 0:07:26.080 --> 0:07:31.640 lot of different problems from fields outside of finance. In fact, 0:07:32.200 --> 0:07:35.280 when I started in finance in the late eighties, before that, 0:07:35.840 --> 0:07:37.800 I didn't even know there was any mathematics in finance. 0:07:37.840 --> 0:07:42.600 But I'd work for are anautical companies, for steel companies, 0:07:42.680 --> 0:07:45.840 for all sorts of different businesses where you try and 0:07:46.720 --> 0:07:50.160 take a physical problem and turn into into something mathematical. 0:07:50.600 --> 0:07:54.200 And then a colleague introduced me to these things called options. 0:07:54.240 --> 0:07:57.360 This was in the there was actually just before the 0:07:57.680 --> 0:08:02.160 seven crash, and just looking into the literature, I found, hey, 0:08:02.200 --> 0:08:08.040 there's there's real mathematics involved in in in derivatives. And 0:08:08.160 --> 0:08:11.120 so I started just to apply the same principles of 0:08:11.160 --> 0:08:14.520 mathematical modeling too derivatives as I had to all these 0:08:14.520 --> 0:08:19.120 other real world physical processes that I had worked on. Actually, 0:08:19.160 --> 0:08:23.160 the when I first saw the famous Black Shoals equation, UM, 0:08:23.840 --> 0:08:25.320 I thought, this is this is great. This is just 0:08:25.400 --> 0:08:30.760 like second year undergraduate mathematics. UM. These days apparently you're 0:08:30.760 --> 0:08:33.040 supposed to have a PhD. But actually that's that's a 0:08:33.160 --> 0:08:36.439 that's a lot of nonsense. UM. Most quant finances just 0:08:36.480 --> 0:08:39.720 second year undergraduate maths. Some of the book is really 0:08:39.800 --> 0:08:43.000 quite fascinating, and you start out with a history of 0:08:43.200 --> 0:08:48.200 investment theory, and I thought, skipping from the south Sea 0:08:48.200 --> 0:08:52.080 Bubble to Adam Smith to the efficient market hypothesis, what 0:08:52.320 --> 0:08:56.760 is the connective tissue between all three of those major 0:08:56.840 --> 0:08:59.800 events in the history of finance. Well, when I think 0:09:00.040 --> 0:09:04.199 you start to see is some going from just people's 0:09:04.200 --> 0:09:08.200 personal experience through too trying to lay down some principles 0:09:08.360 --> 0:09:11.839 and then trying to quantify those principles. So up to 0:09:11.880 --> 0:09:14.839 the period now, the modern era, when pretty much everything 0:09:14.840 --> 0:09:17.920 in finance is all quantified, everything is in terms of 0:09:18.679 --> 0:09:22.280 expected returns and volatility and quantities like that, it could 0:09:22.320 --> 0:09:26.199 be reduced to mathematical models. Everything has become mathematics. Yes, right, 0:09:26.240 --> 0:09:29.360 So so then I always thought of the random walk 0:09:29.440 --> 0:09:34.840 theory as a sort of squishy psychological claim that hey, 0:09:34.880 --> 0:09:38.480 you know some maybe some people can beat the market, 0:09:38.960 --> 0:09:41.520 but you're probably not one of them, and you probably 0:09:41.559 --> 0:09:44.360 can't pick them. And then once we bring in taxes 0:09:44.400 --> 0:09:47.600 and costs and everything else, you're better off just indexing. 0:09:47.960 --> 0:09:51.880 But you relate the whole concept of random walk to 0:09:52.200 --> 0:09:56.000 probability theory. What ties these two together? Well, if you 0:09:56.040 --> 0:09:58.760 go back to the different types of analysis that people 0:09:58.840 --> 0:10:02.839 do in in finance, you've got fundamental analysis, which is 0:10:02.880 --> 0:10:06.480 about trying to figure out, by looking at the company report, 0:10:06.559 --> 0:10:10.800 company's reports and the directors, et cetera, the product, how 0:10:10.840 --> 0:10:14.400 much is this company really worth? Now, that's very difficult 0:10:14.440 --> 0:10:17.439 to do that, That's a lot of analysis is required. Um, 0:10:17.480 --> 0:10:19.640 I mean, can you read balance sheets? I can't read 0:10:19.679 --> 0:10:24.440 balance sheets for exactly exactly so, but the real problem 0:10:24.480 --> 0:10:27.080 is is that you're not trying to predict what the 0:10:27.080 --> 0:10:31.120 the the market value of the company should be, because 0:10:31.120 --> 0:10:33.200 you're trying to predict what other people the people are 0:10:33.240 --> 0:10:37.319 buying certain shares, think those famous beauty exactly. So that 0:10:37.520 --> 0:10:40.559 you move on to fundamental that sorry, move on from 0:10:40.600 --> 0:10:43.280 fundamental to technical analysis, which is exact opposite, which is 0:10:43.520 --> 0:10:46.679 really easy. You draw a chart to some trend lines 0:10:46.720 --> 0:10:50.600 and some patterns. Problem is that sort of thing, statistically speaking, 0:10:50.760 --> 0:10:54.640 doesn't work. The scientific evidence suggested the vast majority of 0:10:54.679 --> 0:10:57.480 that doesn't work. So the great break through the fifties 0:10:57.480 --> 0:11:00.680 six seventies was to say, let's throw that away. Let's 0:11:00.679 --> 0:11:04.240 just say we can't predict things. We've got the concept 0:11:04.240 --> 0:11:08.000 of the efficient market hypothesis as a background for this, 0:11:08.080 --> 0:11:10.600 but let's just say it's like tossing a coin. Now, 0:11:10.760 --> 0:11:13.400 is it a tossing a biased coin? What's the chance 0:11:13.440 --> 0:11:15.400 of heads, what's the chance of tail? So you put 0:11:15.480 --> 0:11:20.320 numbers to these probabilities and the sort of from that 0:11:20.679 --> 0:11:24.520 we get lots of theories about derivatives valuation. And once 0:11:24.559 --> 0:11:27.480 you've got derivatives, you can value derivatives, you can create 0:11:27.559 --> 0:11:30.640 new instruments, and the new instruments what might require new math, 0:11:30.800 --> 0:11:34.160 new mathematics. So you have this snowballing effect where you've 0:11:34.160 --> 0:11:37.320 got the mathematics increasing and the number and type of 0:11:37.320 --> 0:11:41.120 products similtarency increasing. So there's a quote in the book 0:11:41.160 --> 0:11:43.400 I really like, and I'm going to mangle this a 0:11:43.400 --> 0:11:47.959 little bit. No investment strategy based on mainstream finance theory 0:11:48.040 --> 0:11:53.600 can protect protect investors from market wide crashes. First, is 0:11:53.600 --> 0:11:56.800 anyone really suggesting, Hey, our strategy is going to protect you? 0:11:56.960 --> 0:12:00.600 Is that? I guess that's naive of me to uh 0:12:00.640 --> 0:12:04.160 to ask. But do people really believe, oh, here's the 0:12:04.200 --> 0:12:09.360 magic seven like portfolio insurance that will protect us. I'm 0:12:09.360 --> 0:12:11.680 sure there are some things you can do. Um, there 0:12:11.720 --> 0:12:14.400 are all sorts of nuances here. I remember Nassy, my 0:12:14.480 --> 0:12:18.080 dear pal, Nassim Talent, telling me about his um the 0:12:18.120 --> 0:12:21.800 way he pitches this this idea, and that is to say, 0:12:22.080 --> 0:12:25.080 it's insurance. But it's not a portfile, it's not an investment. 0:12:25.400 --> 0:12:27.800 It's insurance. You're gonna lose money. In fact, you hope 0:12:27.800 --> 0:12:29.720 you'll lose money because if you make money on this 0:12:29.800 --> 0:12:31.959 insurance is because your house has burnt down, which you 0:12:32.000 --> 0:12:36.640 don't want to have happened. So that's quite that's a 0:12:36.679 --> 0:12:40.160 psychological way of looking at this, getting the putting in place, 0:12:40.240 --> 0:12:43.880 the the the protection as a as a as a 0:12:43.960 --> 0:12:48.080 cost rather as an investment. Um. There are some things 0:12:48.080 --> 0:12:52.360 you can do. Of course, the famous portfolio dynamic portfoileon 0:12:52.360 --> 0:12:54.400 insurance of the of the eighties was it was a 0:12:54.440 --> 0:12:59.760 classic example of a feedback effect, the positive feedback effect, 0:12:59.760 --> 0:13:02.839 that the the very action of trying to protect your 0:13:02.840 --> 0:13:07.720 portfolio is what caused made which is a beautiful concept. 0:13:07.840 --> 0:13:10.319 What what I don't understand is who thought, oh, if 0:13:10.320 --> 0:13:12.200 the market is crashing, I'll go out and buy some 0:13:12.240 --> 0:13:15.720 pots and that will protect me. Summing it up that way, 0:13:15.920 --> 0:13:19.160 how did anybody think that that model was valid? And 0:13:19.440 --> 0:13:22.880 granted there's a little hindsight bias here, but still it 0:13:23.080 --> 0:13:28.040 sounds so obvious after the fact that that would not work, right. 0:13:28.280 --> 0:13:30.719 I don't know. I don't know the I think the 0:13:30.880 --> 0:13:33.400 I think in all of this that a key thing 0:13:33.440 --> 0:13:35.560 you've always got to remember is that the psychology of 0:13:35.600 --> 0:13:39.920 the markets and that whatever the mathematical models say, somehow 0:13:40.240 --> 0:13:43.800 human beings in this particular field do try to do 0:13:43.960 --> 0:13:49.360 manage to mess up the models. So here's another mangled quote. 0:13:49.800 --> 0:13:55.200 Bankers offering complex financial products don't always understand the risks, 0:13:55.640 --> 0:13:59.200 and that when a bank goes bust, stock markets collapse, 0:13:59.280 --> 0:14:02.800 house prices tumble, it's your bank account, your shares, and 0:14:02.840 --> 0:14:07.439 your home equity that suffers. Right, Well, it's certainly true 0:14:07.520 --> 0:14:11.840 in my experience that bankers and especially regulators, I have 0:14:11.920 --> 0:14:15.400 to say, I do not know as much as they 0:14:15.440 --> 0:14:19.760 ought to. Um, if they were doctors and they didn't know, 0:14:19.960 --> 0:14:23.800 you know, bleachers do a little bleeding exactly, or you 0:14:23.800 --> 0:14:25.760 know which side is the heart? Aren't you know that 0:14:25.840 --> 0:14:29.600 sort of thing, then they be lawsuits flying. But bankers, 0:14:29.880 --> 0:14:33.240 in my experience, there's a lot of herd mentality within 0:14:33.320 --> 0:14:36.880 bankers and that works for them. That's how that's how 0:14:36.920 --> 0:14:40.480 they make money is from from pretending there are no 0:14:40.640 --> 0:14:44.360 risks and although they may deep down they may realize 0:14:44.360 --> 0:14:48.520 there are risks. And regulators I haven't met that many regulators. 0:14:48.520 --> 0:14:51.520 I don't seem to bump into them very often, maybe 0:14:51.480 --> 0:14:53.480 because the things I've said about them in the past, 0:14:53.800 --> 0:14:57.800 but my experience is often they're very good at book learning. 0:14:57.880 --> 0:15:00.280 You know, they've they've got that they've read the math books. 0:15:00.320 --> 0:15:03.600 They know the models, but they're sorely lacking in street 0:15:03.640 --> 0:15:06.520 smarts in my experience, so they don't really understand how 0:15:06.520 --> 0:15:10.240 the world works. So you have in the past suggested 0:15:10.280 --> 0:15:15.160 a solution to the problem of both bankers that don't 0:15:15.240 --> 0:15:18.600 understand risk and regulators that don't know how to regulate 0:15:18.640 --> 0:15:23.200 it is a tobin tax on each transaction, a point 0:15:23.400 --> 0:15:28.080 zero zero eight percent tax on any buy sell, any 0:15:28.120 --> 0:15:31.520 trade whatsoever. And how would that work and how would 0:15:31.520 --> 0:15:35.320 that protect markets? You're very good with your numbers. The 0:15:35.320 --> 0:15:41.080 the idea of the Tobin tax is to stop very 0:15:41.160 --> 0:15:45.600 high frequency trading because, uh, you then lose with the 0:15:45.680 --> 0:15:47.640 high frequency. The problem with one of the problems with 0:15:47.680 --> 0:15:51.440 high frequency trading is that you lose the connection between 0:15:52.200 --> 0:15:56.360 the value of a share company and the price. It 0:15:56.480 --> 0:16:00.760 just becomes about, you know, the these these some time 0:16:00.840 --> 0:16:03.280 series of numbers, and it could be anything. It doesn't matter. 0:16:03.960 --> 0:16:06.160 But there's a company here and there are people whose 0:16:06.840 --> 0:16:09.720 jobs on the line. Let's talk a little bit about 0:16:10.160 --> 0:16:14.960 the time you spent at Kayasa Capital Capital two people 0:16:15.080 --> 0:16:17.680 separated by a common language, So you were a founding 0:16:17.760 --> 0:16:22.760 partner of the volatility arbitrage hedge funds, which was running 0:16:23.240 --> 0:16:26.880 about two million dollars just on the two million and so, 0:16:26.880 --> 0:16:30.840 so tell us a little bit about volatility arbitrage, which 0:16:30.880 --> 0:16:35.600 is sort of ironic in these days of almost no volatility. 0:16:36.000 --> 0:16:39.160 It's interesting that that it's very difficult I find I 0:16:39.560 --> 0:16:42.840 found to predict the direction a stock is going to 0:16:42.920 --> 0:16:46.680 go up put down. But it seems to be easier 0:16:46.800 --> 0:16:49.080 and there are statistical reasons for this, actually, but it 0:16:49.120 --> 0:16:52.720 seems to be easier to to forecast volatility. How much 0:16:53.040 --> 0:16:55.800 noise there in there is in a stock. So you 0:16:55.880 --> 0:16:59.400 put a counterintuitive though, that it's harder to find the 0:16:59.440 --> 0:17:01.880 signal and to use you the noise. Yes, I can 0:17:01.920 --> 0:17:04.800 see that, but at the same time, you can see 0:17:04.800 --> 0:17:08.960 what the market is sort of thinking volatility is because options, 0:17:09.040 --> 0:17:14.119 the valuation of options hinges upon estimation of volatility. So 0:17:14.160 --> 0:17:17.359 you can say, oh, here's a core option, what volatility 0:17:17.400 --> 0:17:19.639 are they plugging into? The famous black shows formed to 0:17:19.680 --> 0:17:21.919 give that that price that you see in the market, 0:17:22.280 --> 0:17:23.800 and so you can you can back out a thing 0:17:23.800 --> 0:17:26.160 called implied volatility, which is sort of a bit like 0:17:26.200 --> 0:17:30.760 what the market thinks. So you look around for when 0:17:30.960 --> 0:17:35.560 is the implied volatility different sufficiently different? From your forecast 0:17:35.640 --> 0:17:38.600 volatility and if they're different, and if you just happen 0:17:38.680 --> 0:17:41.800 to be right, then you can make some money. So 0:17:42.000 --> 0:17:48.040 that complex it's it's the purest sort of volatility arbitrage. 0:17:48.600 --> 0:17:54.800 I'm betting my forecasts against the markets forecasts essentially is 0:17:54.840 --> 0:17:56.600 what is what it amounts to? So in not so 0:17:56.720 --> 0:17:59.680 efficient market, not so efficient, not so efficient. I don't 0:17:59.680 --> 0:18:02.439 think believes in efficient markets. Who's listening to this program 0:18:02.480 --> 0:18:06.359 to UM? I think there are a decent number. It's 0:18:06.359 --> 0:18:09.240 it's a really interesting point because I think there are 0:18:09.280 --> 0:18:13.960 a number of people who say, um. Look, cliff as Nests, 0:18:14.080 --> 0:18:17.960 who runs a q R Gene Fama was his PhD 0:18:18.200 --> 0:18:22.119 dissertation advisor and Fama's credit. He said, hey, if you 0:18:22.160 --> 0:18:25.240 want to go out and prove factors work, I'll I'm 0:18:25.240 --> 0:18:28.919 happy with that. So we know the market is kind 0:18:28.920 --> 0:18:32.960 of a sort of mostly eventually efficient. Is that? Is 0:18:32.960 --> 0:18:35.680 that a fair way to describe it? Not necessarily. It 0:18:35.920 --> 0:18:40.359 depends where there's any mechanism for for removing any efficiency. 0:18:40.400 --> 0:18:42.440 People just say oh that, you know, people say things 0:18:42.440 --> 0:18:45.160 like the markets always right, and well, we know that's 0:18:45.280 --> 0:18:49.560 certainly not How how could the market be twenty three 0:18:49.680 --> 0:18:52.560 percent lower the day after the crash. Was it right 0:18:52.600 --> 0:18:54.520 the day before? Was it right the day after? Is 0:18:54.520 --> 0:18:57.120 it wrong both times? It's kind of hard to say 0:18:57.119 --> 0:18:59.800 the market is always right right. Well. One then example 0:18:59.840 --> 0:19:03.199 I I sometimes give my students is about uses the 0:19:03.240 --> 0:19:05.879 idea of car insurance. For example, suppose you've got a 0:19:05.920 --> 0:19:09.119 twenty thousand dollar car and your annual car insurance is 0:19:09.200 --> 0:19:13.200 let's new simple numbers a thousand dollars. A quant would say, oh, 0:19:13.280 --> 0:19:15.080 it's a twenty thousand dollar car. It's a thousand dollar 0:19:15.119 --> 0:19:16.680 car in shoans are that means must be a five 0:19:16.680 --> 0:19:19.800 percent chance of crashing. And because they're missing out from 0:19:19.880 --> 0:19:22.400 that five fence the fact that the insurance company needs 0:19:22.400 --> 0:19:25.480 to make a profit, I'm not doing things all sorts 0:19:25.480 --> 0:19:28.919 of things. So so the way your typical quant or 0:19:29.800 --> 0:19:34.720 um Chicago finance professor looks at things is in a 0:19:34.800 --> 0:19:37.040 very pure way that sort of misses all the fun. 0:19:37.400 --> 0:19:40.639 What I think, and as I've said before, I've been 0:19:40.720 --> 0:19:44.439 running businesses since I was a child, virtually, and the 0:19:44.480 --> 0:19:47.480 business side of things is what makes things interesting. And 0:19:48.080 --> 0:19:51.240 so when we approach this sort of going back to 0:19:51.280 --> 0:19:56.479 volatility arbitrage, you'd be amazed at if you look at 0:19:56.480 --> 0:19:59.960 the literature. The there must be tens of thousands of 0:20:00.040 --> 0:20:07.359 papers which talk about how to how to back out 0:20:07.720 --> 0:20:10.879 from from market prices, what the implied vault inty is, 0:20:10.880 --> 0:20:13.600 and how they assume that the implied volatility is right, 0:20:13.680 --> 0:20:16.320 that the market knows, the market has a crystal ball, 0:20:16.359 --> 0:20:18.560 it knows what vaults is going to be. Must be 0:20:18.640 --> 0:20:21.720 tens of thousands. How many papers are there which say, well, 0:20:21.800 --> 0:20:24.800 you know what, what if the market is wrong and 0:20:24.880 --> 0:20:28.000 the volatility, the real volatility is different from implied volatility, 0:20:28.400 --> 0:20:32.919 how can you make money? Then that's the underlying strategy, 0:20:30.760 --> 0:20:36.560 that said Catholic. That must must be half a dozen papers, 0:20:36.720 --> 0:20:39.399 and surely that's the most important thing. We've got an option. 0:20:39.720 --> 0:20:42.119 The question is what am I going to do with this? 0:20:42.320 --> 0:20:44.000 I could hedge with it, or I could make money 0:20:44.040 --> 0:20:47.320 from it, and then just half a dozen papers on 0:20:47.359 --> 0:20:50.000 how to make money. So so, how did the hedge 0:20:50.040 --> 0:20:54.280 fund do? The hedge fund was did did wonderfully for 0:20:54.400 --> 0:20:57.959 it's its lifespan of three years, which I believe is 0:20:57.960 --> 0:21:01.000 is the average lifespan of a hedge fund. And and 0:21:01.040 --> 0:21:03.040 so at the end of the three years, you basically 0:21:03.080 --> 0:21:06.560 tapped out and said I'm going to go back to writing. Well, 0:21:06.600 --> 0:21:08.720 I I learned a lot from that that hedge fund, 0:21:08.920 --> 0:21:11.679 a lot of things about not just about hedge funds, 0:21:11.720 --> 0:21:16.120 but also about about myself. I found that very interesting 0:21:16.160 --> 0:21:19.800 because the Hedgeman was based in New York um and 0:21:19.800 --> 0:21:22.199 where were you located was located in London, so there 0:21:22.280 --> 0:21:25.040 was this five hour time difference, and just as things 0:21:25.040 --> 0:21:28.320 were getting interested New York was when I was sitting 0:21:28.359 --> 0:21:30.080 down to dinner or onting to go to bed or 0:21:30.119 --> 0:21:32.960 something like that. So that was kind of frustrating. Let's 0:21:32.960 --> 0:21:37.159 talk about the Wilmot business model. So you run the website, 0:21:37.480 --> 0:21:41.399 you have a print magazine, you have a Paul and 0:21:41.440 --> 0:21:45.040 Dominic quant recruitment, the book, which we all know is 0:21:45.119 --> 0:21:48.639 books or giant money makers. What is your business model 0:21:49.080 --> 0:21:53.280 as quant expert. Well that the Poul and Dominics doesn't 0:21:53.280 --> 0:21:56.760 exist anymore. That's but the and also there was the 0:21:57.640 --> 0:22:00.160 we talked about the world's largest quant website. It's also 0:22:00.160 --> 0:22:03.600 apparently the world's most expensive magazine. I've been told. I 0:22:03.640 --> 0:22:07.280 don't know that's true or what's the magazine costs. It's 0:22:07.280 --> 0:22:10.200 not six hundred dollars for six issues. I don't think 0:22:10.240 --> 0:22:13.119 that's terribly expensive, and I looked it up. It's not 0:22:13.160 --> 0:22:14.760 actually this And if you look at some of the 0:22:14.840 --> 0:22:17.399 like the medical journals and stuff, these are thousands of 0:22:17.920 --> 0:22:19.760 In fact, there was a big article not too long 0:22:19.800 --> 0:22:23.200 ago about what used to be academic research now sells 0:22:23.240 --> 0:22:25.960 for thousands of dollars and there's two companies that have 0:22:26.080 --> 0:22:29.359 sucked up all these formally free publication. And that that 0:22:29.600 --> 0:22:31.720 was a quote from Esquire magazine. So I'd just like 0:22:31.800 --> 0:22:35.760 to repeat that it's not very useful in marketing, though 0:22:35.920 --> 0:22:38.320 there was until recently, well there still is. The Certificate 0:22:38.320 --> 0:22:41.560 and Quantitative Finance that that is is the world's largest 0:22:41.640 --> 0:22:45.120 high level quant education with a company called seven City. 0:22:45.320 --> 0:22:47.920 This CQF was was founded with a company called seven 0:22:47.920 --> 0:22:51.960 City Learning two thousands three was it. But it was 0:22:52.000 --> 0:22:55.159 sold up to Fitch about three or four years. So, 0:22:56.000 --> 0:22:58.280 but what is the business model? The business model, Well, 0:22:58.320 --> 0:23:01.439 my whole business model all my life has been do 0:23:01.600 --> 0:23:05.119 something which is fun and then and then I accidentally 0:23:05.240 --> 0:23:07.520 or I always have this urge to turn things into 0:23:07.680 --> 0:23:10.080 into businesses. It's not a it's not a greed thing. 0:23:10.119 --> 0:23:13.200 It's an enthusiasm. Think. It's almost like I've got some hobby. 0:23:13.280 --> 0:23:16.400 I'd like to make it public. It becomes more long 0:23:16.520 --> 0:23:20.240 last thing and sustainable if there's a revenue stream behind it. Yeah, 0:23:20.280 --> 0:23:21.680 I don't think of it like that, you know, it's 0:23:21.680 --> 0:23:24.080 just sort of is something in my DNA that says, 0:23:24.280 --> 0:23:27.800 you know, for example, I've started to learn the ukulele 0:23:27.880 --> 0:23:30.040 a few years ago, and I just know sometime in 0:23:30.080 --> 0:23:32.080 my life there will be a time when I'm on 0:23:32.080 --> 0:23:34.080 a stage and people are paying money to listen to 0:23:34.080 --> 0:23:35.560 me play the ukule I'm going to take the other 0:23:35.560 --> 0:23:37.879 side of that trade. You should hedge that bet because 0:23:37.880 --> 0:23:39.640 I don't see that happen. No, No, like, yeah, I'm 0:23:39.640 --> 0:23:41.840 pretty good, I'm getting I'm getting that, I'm getting. No, 0:23:41.920 --> 0:23:44.000 it's not. It has to do with your skill set. 0:23:44.320 --> 0:23:47.000 It has to do with the demand ukulele. Before you'd 0:23:47.000 --> 0:23:50.040 be surprised. I would be surprised. It is the instrument. 0:23:50.080 --> 0:23:52.520 All the schools are taking up ukule because it's I'm 0:23:52.520 --> 0:23:55.200