YouSaid · the spoken record

Michael Robbins

lines on the record: 95
first: 2023-07-13
most recent: 2023-07-13
sittings or episodes: 1
sources: podcast

Every line below is reproduced as it was said and linked to the record it came from. Nothing here is summarised or generated. Directory · Search · Corrections

“Yeah, so we're an investment advisory. So we're a low frequency firm that provides different portfolios for different clients based on their risk talencies and their preferences. And we do use quantitative methods to design those portfolios and to engineer them so that they're as efficient as possible so we can maybe beat other portfolios or less efficient.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“The large cap stock market, right? So that much is true right now. And I think it will be true in the future for things that are sufficiently good at what they're doing. It just might be that the time decay on those trades might be very, very fast. And computers may do a basis trade for this microsecond and do another pairs trade for another microsecond and do an oil gold arbitrage for another one. And computers might just be doing these things so quickly we can't keep up.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Well, so to kind of deconstruct that sentence, you're talking about averages, first of all, right? So yeah, the average investor will not beat the market, but I'm sure there are quite a few exceptional investors who might not be consistently year in and out, year out, but on average beat the market. So there are some people with skill. And we don't want to discount that. Another thing is the market is not necessarily a monolithic thing. So for my first 15 years, I was a proprietary trader. And the people I worked with made money every year. They were really good at trading. They weren't trying to beat the S&P 500. They were doing, say, a basis arbitrage or something like that. Everybody had their own trade specialty. But you can consistently beat a market, maybe not.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“But no matter what, you still have that dynamic where the market has a beta of one, the market's return is the market's return. There are folks who outperform that, you know, net of fees. There are people who underperform that net of fees. And the people who outperform benefit at the expense of people who underperform it's kind of a zero-sum game. And that's true now when it's human investors and traders. If there's a future where 95, 99% of all trading is computers, that will still be the case, right? just compositionally.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“So it'll be a complicated cat and mouse game. It'll be really interesting in the future. And I don't know at what point we reached that singularity where human input just won't have that much advantage. But we're not there yet. There's still a lot we can do ourselves”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“And you want to get out of the way of that, you don't want to do high frequency trading with your mouse. But there are also things that computers are still not very good at that you can identify other weaknesses and expose them to. But they'll also gradually go away. I think I read a week or two ago that a person beat a machine playing the game of Go, you know, that little game with the rocks, but they did it by asking another computer what that algorithms weaknesses were. So he used a computer to find the weaknesses of another computer. He didn't beat it just by outthinking it.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“In maybe a closed source model, but also to train it to come up with a proper conclusions has a big part to do with how these models will be able to fight themselves in the future, to fight each other, and to come up with trades and to beat each other. There'll also be misinformation and you see in markets. And there'll be, as you see in some more legal markets like distress stead, there'll be ways to try to get around regulations and to provide misinformation without running afoul of the law and things like that. So things I think will get really interesting and complicated and fast and what can be done now about it is now people can outsmart regimes for the most part, I think. There's a lot of things that machines are really good at.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Moving parts, you could have a better algorithm, you could have a better algorithm trainer. A lot of people are talking about prompt engineering, how to use models. For a long time thought that one of the really important careers in the future is not to be the guy who makes the prompt, but to be the guy who trains the model, who feeds it the data, and builds its network internally by exposing it to the right information in the right way. And right now, that's the biggest problem with these large language models is that they have a bunch of nonsense fit to it as information, right? They have like the whole internet full of misinformation, and they sometimes respond in ways not only don't they make sense because their logic doesn't force them to make sense, but also because they were trained on some data that might not be accurate. So to feed it correct data.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Expenditures, right? High frequency trading is a lot about having the fastest technology and making the fastest decisions and putting enough money in that technology so you're not behind the curve and your competitors don't have better technology, at least when it first started. High frequency trading had very little to do with good decisions. It had everything to do with finding an opportunity that would only last a microsecond or less and just arbitraging it. you know buying gold on one one part of the world and selling it in another part of the world right the exact same asset and um so i think technology won't be quite like that with ai because it is very differentiated there's lots of different models and lots of different ways to tune them and feed them different information so that will certainly be part of it and there will be a lot”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Computers to make good decisions, but more likely than not, most people are going to use computers to make a lot more bad decisions because they're easier to make in a confusing way. But in terms of computers fighting computers, I had a short career in electronic intelligence, in electronic warfare, and it reminds me a lot of that it also reminds me a lot of high frequency trading where people are basically just trying to be faster than each other and they're doing all these crazy things to be as unbelievably fast as possible cutting down forests so they can have a direct microwave beam and you know things like that and so I think the lesson to learn from high frequency trading in this regard is that it really transformed the business of trading from being predictive to building a business that's efficient and requires capital”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, so you brought up a few really interesting points. And I think one of the things that's changed recently over the past 10 years is not so much the sophistication of the algorithms, but the access to them. So tools like MATLAB and Python that are free or not very expensive are accessible to everyone and they can use these really sophisticated techniques. And most recently with large language models, they've been really popular in the media and they're certainly really accessible by people who don't even know how to program. So I think that's the most recent big change. In terms of computer versus computer, First of all, people using computers versus computers is really a question.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“The computer. And no one, not everyone can beat the market because everyone is the market. You know what I'm saying? So, what's kind of the endgame if you have this kind of machine versus machine endgame where this genius computer beats the market, but then if everyone uses that computer, it stops working.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Might be a factor, predicting whether to buy or sell the two-year note or five-year note or 10-year note is something entirely different.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“But the fives or the tens are rich or cheap compared to other current issues that are off the curve or on the curve. And you can move those different valuations based on the specific assets and take that yield curve abstraction, that signal of twos, tens or fives tens, and then modify it to come up with a valuation for individual bonds and then say, should I buy or sell those individual bonds based on my two tense factor. So for instance, your two's tens might be rich or cheap. It might say you should buy the curve or sell the curve, but then when you go to say, should I buy the two year note or the 10-year note, it might say, no, don't do either one of those things. Those are so rich or cheap to the curve. You should buy something off the run. So predicting the 2's 10s abstractly, not the actual.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Well, yeah, there's lots of research papers describing hundreds and hundreds of factors and any little anomaly might be considered a factor, any dislocation from the yield curve or from a quality sentiment, anything you could think of that's maybe an abstract thing can be considered a factor and used to make predictions that are not subject to the high frequency noise of the underlying assets. So for instance, in terms of the yield curve, you might use a yield curve smoothing technique, right? And you can predict the yield curve in the absence of supply and demand for individual issuances, right? And you could say, well, the 5's tens curves should have this certain slope based on my economic model.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Then you can work on the second stage of figuring out which assets that fall in that category are the ones to buy and sell. But it's, I think, too complicated to try to predict the actual assets without removing that noise away and treating them as factors. Does that make sense?”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Well, to me, a factor is anything that can abstract away from the actual assets. Some people even think of asset classes as factors. I think that's a little too close to the actual investable assets. But if you want to think about even like large cap equities, that could be a factor. You could predict large cap equities without specifically predicting any particular stock or any particular index. So if you can abstract that, then you can lower the noise in the signal by focusing on the mechanism. So what makes large cap stocks go up relative to small cap stocks where quality stocks relative to growth stocks in the abstract sense, what makes those factors work? Then if you can focus on that and predict that.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“And for the small order execution system, which had a small market, so when the market maker got nervous and they wanted to change their price, they change it in the Nasdaq system and then they changed it in the subsystem as well as a delay. And so there were some smaller traders who glued themselves to the screen, watched the Nasdaq screen. And as soon as a big pulled, they'd hit the bid on the SO screen, right? And they took advantage of that small time difference and arbitrage it away. And these esoteric factors are like that too.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Well, yeah, they're the five most common factors, but a lot of what I talk about in the book is that there are lots of factors, and people often It doesn't matter if a factor persists for a microsecond as long as you can make money off of that. And so I try to emphasize factors that are a little more esoteric, that have a little more juice in them. When you talk about something like quality and growth and momentum and things like that, everybody's watching it. Everybody's your competitor, including people who are a lot smarter and faster than you. And that really draws your skill away into the luck region. Whereas if you can identify a really specific factor, then you can take advantage of it and maybe you won't have so much competition. So you may remember the So's Bandit arbitrages back in the 90s where there was usually one market maker for the Nasdaq level two, which had a big market.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“There's a lot of things about shorting and liquidity in general that are really interesting. And you can certainly try to predict them on shorter time spans that are relevant to your investments. But over longer time spans, just like the two stance curve, it really has a lot to do with the Fed and what happens on a political level.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, and then you have the problem also if you're too right and the company gets delisted, then what do you do about your short?”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“And often, you know, if someone who doesn't have a position in a particular thing, think this stop a little fundamental, this stock is so overvalued. It's going down. It's a short, they're right. And if you could short it for free, it would be an easy trade all day. But the bank's going to charge you 500% to short it. And if you want to buy a put option, it's implied volatility of many hundreds of all points. So the market knows that it's an easy short. And that's why it's hard to short.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, right. But even so, you may not be allowed to. Even when you do say total return swaps, a lot of times when I short, I short through total return swaps. But even then, the bank can pull the underlier. It also happens.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Where really affects investments is when you're shorting, right? If you have an arrangement where you can get your shorts pulled, then you might lose your most profitable shorts. And by profitable, I mean the ones that will go down because they'll get pulled from you and you'll no longer be short them. And that's happened to me frequently in real life. And I've actually had to build into my back tests that if my shorts are too good, they have to get stopped out. Because somebody else will stop me out, you know.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, the liquidity is kind of oxymoronic in that there's a lot of really good quantitative measures, like looking at options and things like that. But at the end, it really is highly manipulated. You can remember quantitative easing and the Fed could be raising rates and injecting liquidity. It's really manipulated. What's more useful for my investing to go back to kind of the blocking and tackling where we focus is how liquidity affects your actual investments.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“How do you do quantitative measures think about liquidity? I mean, there are some models that are somewhat simplified that, oh, liquidity goes up when the Fed balance sheets goes up. It goes down when the Treasure General account goes up or when the reverse repo facility goes up. And it's kind of that tripartite analysis. It's obviously a lot more complicated. What quantitative successes have there been to sort of track liquidity, not just track liquidity and measure it, but track the correlation between, oh, liquidity is actually going up, so risk assets should go up. And I know there are people who post my liquidity model is actually very correlated with the S&P 500, but there are a lot of challenges that quantitative folks can find of actually saying, is this kind of overfitting auto correlation stuff like that? So it's totally about the quantitative efforts to kind of measure and measure liquidity and see how impactful is it in predicting.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Well, that's what those tests are for. That's what those inventions do. They test to make sure the model is accurate. They prove that the model actually works. Now, of course, you've got to design your tests properly. You can't just be thoughtless about it. But there's a whole science behind testing these models and mathematically proving that they're consistent and also monitoring them to make sure they continue to be consistent. So for instance, maybe the 2 tense curve might be one of the arrows in your global economic model. And that 2's tense curve could flash recession. But the other branches in your model might indicate, no, it's not a recession. They contraindicate that branch. And the causal map could say, well, ignore the 2's tens curve for now because we're getting other confirmation from other branches.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, it's a really fascinating theory called do logic, DO logic. And the idea is you can create these experiments to test your theory, to prove to yourself that it's true mathematically and you create what's called interventions. And you don't have to rely on what's called natural experiments, which is seeing things happen in the market and saying this confirms or disconfirms my theory because it happened and it just adds a little more information than I had before. With causal theories, you can test your theories and prove that they're true so long as your map is right. So it's a whole fascinating branch of science. And there are companies that are using this to invest.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“On themselves, so the technical term for that is a directed acyclic graph. The arrows have a direction and they can't create a cycle. Now there's no reason why in the future the theory can't expand to adjust for those things, but right now there's limitations. But if you can”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Exactly. But there is a theory of causation where you can actually define those things and develop a mechanism that has a causal effect. And what's really great about that theory is that it can also tell you when things aren't working the way it thought and say, well, we had a cause of model and we could rely on it, but we can't rely on it any longer because our theory has been violated. So an example of how that works lately in the current state of the theory is you can create what's called a graph, which is just a bunch of arrows between events and other events. And the arrows are the mechanisms between those events. And currently, unless I miss some recent research, it only works if you can draw the graphs. so that the arrows go in a single direction and don't loop back.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Like a distressed debt, right? Or mergers and acquisitions. These are things that have much more defined bets. There's risk in them for sure, but there are mechanisms as well, whereas the stock market is much more divorced from the drivers. It's a lot harder to say, well, for every basis point, interest rates go up, stock prices are going to go down by this beta. It's just not true. And it's a lot fuzzier. What's really interesting lately is this branch of AI called causal AI. And that seeks to really tighten up these mechanisms. So the idea behind causal AI is to eliminate the idea of tendency that's behind statistics. So we all know when you draw a regression, there's no real reason why that regression should hold just because the dots produced a line. There's no reason why the next stock shouldn't be way off that line. Correlation is”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Well, so we focus mostly on the things that don't rely on signals like diversification. There are some signals that are stronger than others, but what's really important to understand when those signals work and when they don't and to be really thoughtful about it. And that's where value has its advantage because it's a really thoughtful theory, right? And there is some mechanism tying book value to how much people should pay for a company in things other than stocks like bankruptcy cases those mechanisms are a lot more sure. And a big part of what I write about in the book is that the world is a lot bigger than large cap stocks. There are many different things you can invest in. Things that have much sure signals and much sure mechanisms that you can rely on, at least more than a sentiment driving the price up or down.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Lower prices directly. And so that's why these simple things, they're really attractive, they're easy to talk about, they make good headlines, but they're not always the best bets. Even if they happen six out of six times before.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, and that's also another good example of why things might not apply. The traditional way of looking at banks is that deposits are really sticky. And obviously that's not always true because of what happened with Silicon Valley Bank and several others. So the idea that those that supply of money will persist was an assumption that was made and a reason why you might think the yield curve might affect banks in a certain way, which proved not to be precisely true. So the yield curve is a gross indicator. It's not really very precise. It doesn't directly address the true mechanisms that make prices go up and down. And it may have a lagged effect. It may have an effect that doesn't see its way into high prices.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“GP Morgan actually is in a relatively catbird seat, whatever you want to put it. And they actually hedged a lot of that risk out. And unlike Bank of America, who has over $100 billion of unrealized losses of their securities, they didn't buy a lot relatively. They did not buy a lot of long-term bonds. But you're absolutely right.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Where they invest in long term bonds and they get their money from short-term deposits, but they're still earning a positive spread. So, yeah, their earnings are lower, but they have no risk of going out of business. And there's no reason why they shouldn't stop making money.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, absolutely. And in terms of the mechanism that causes the yield curve to slow the economy, there's a lot of steps in between. And yeah, it affects the discount rate, but the discount rate is not a mechanism that's enforced. But if there's a lot of liquidity in the market, if companies, their products are selling, all these things boost the value of stocks just because it might be more expensive to borrow doesn't mean that stocks are going to go down. And if they are, it might take a while, right? If sales are robust, maybe you can survive without really low financing for a while. Maybe your spreads are enough so you can survive indefinitely. There was an article recently about, I think, JP Morgan where it said, yeah, they're kind of in that trap that banks are in.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Absolutely. And also, people when the yield curve averted about a year ago, people said recession imminent right now, buy bonds, sell stocks, short stocks, buy bonds. That has worked out horribly. So being early is the same thing as being wrong. You are absolutely right.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Well, I don't think it's random. The question is whether it applies to our current situation and whether your timing's right. Timing is very important. And as a lot of people saw it in, say, the 2001 tech crisis, a lot of people knew it was going to happen, but they didn't time it right and they lost a lot of money. Absolutely.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“And I'll stay the data doesn't go back that far, but if you look in the 1800s Yield curve is very frequently inverted. And I mean, you had a lot of recessions, but it was inverted all the time when you had booms too. So it's there are mechanical reasons. But it seems convincing. I'm not saying people should devote their entire strategy to it, but if it is random, it's pretty odd that it's this, right?”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“210 spread, and there are many different types of the curve that can be. Two years higher than the 10 year. And it typically goes a few years, months before a recession, which is in a shaded bar. One, two, three, four, five, six, oh, six. Okay, six out of six is not nearly as good as 10 out of 10.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“So, even the track record of 10 out of 10, it's not 10 out of 14. It's 10 out of 10. That's, I mean, obviously it could be better. We don't know what the yield curve was like in Mesopotamia. But explain just how that it could be random evidence. Because for me, if I see someone flip, if you flip a coin and 10 out of 10 times its head, I'm thinking, hmm, like, well, either there's a, you know, one over two to the 10th power or whatever that is 512. I don't know, that's a coin. But the coin might have two heads, you know? Maybe. I mean, at some point, you're approaching the probability that something strange is going on here and there could be a relationship. So how do you know, I mean, I guess the answer is we don't know, but in what way are you sort of not convinced by 10 out of 10? And let me pull up the Fred, I mean, because it's like we could put the chart up of the...”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Tied to the mechanisms in the market at this current point. That's a useful tool for study rather than just saying, well, nine times out of ten, it worked, or 10 times out of 10 it worked. Well, that's only 10 observations. That's not enough to really bet a lot of money, not for me. And it could be spurious too, right? If you only have 10 observations, it's hard to say that it's significant. If you have a thousand observations and it worked 51% of the time, that might be enough, but 100% of 10 observations might not be.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Fundamentally, theoretically, if you believe in the dividend discount model, then the interest rates will affect the discount rate. And if you have a higher duration on a stock, then certainly it will affect it more. Whether or not that narrative is true, that growth stocks are growing because people expect long-term growth, or if it's just like a meme stock that has a high momentum, that's a question. Also, the very simple rules that people like to talk about and hang on to because they're easy to verbalize, like the Yo-Curve slope. That certainly is affected by rates, but whether it has an effect on us today or whether this market is going to pay attention to it. That's a whole different story. And identifying when that might work and when it's kind of out of whack and not really”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“One, Savor Cow in particular is not only that higher interest rates, so an increase in bond yields are not good for stocks, but that they're worse for growth stocks than they are for value stocks because growth stocks are of a longer duration, Tesla, all the money that Tesla is going to make or Apple is going to is far in the future, whereas like a coal stock that has a price to earnings ratio of four, you know, a lot of the money it's ever going to make in its lifetime is this year or in 2024. So therefore, a rise in the discount rate hurts growth stocks more than it does value stocks, cheaper stocks. That narrative makes such intuitive sense. And, you know, on a one-year time basis, three-year time basis, I mean, you could always, you know, especially anecdotally say, oh, well, bond yields exploded higher today. So no surprise to Nasdaq is down 3%. But on a long-term basis using the rigorous quantitative tools we've been discussing here.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“When they underperform, maybe it's in rising markets, maybe it's in rising economies, maybe sector rotation strategies work when the economy is growing and not in recession or vice versa, right? And by being really precise in your thoughts, by putting numbers on them, or at least rankings, it helps you refine your thought process. And I think it helps you be a better investor.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“It's especially true of, say, ESG investing, but it's also true about inflation protection or dividend investing. All those things, they work sometimes, they don't work other times. It's just a fact that people used to accept and now have the tools to investigate and at least know when to stay out of it, right? You might not be able to find the solution to how to make it work when it doesn't work, but you may be able to identify those periods or those investments where it's not appropriate and just step out of the way for that period. Or with AI, you can create a meta model, a model that picks which strategies to use and when. Kind of like an asset allocator picks hedge fund managers. And it can identify using different indicators like the VIX when some strategies perform and when they”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“So, in terms of growth value and other things, when you see that growth may be trending in certain periods and not mean reverting, if you're quantitative, you can try to ask, well, why is that true? What periods does that happen in? And should maybe we do quality growth, right? Or is there some way we can adapt our strategy incrementally so that it works when it wouldn't work in the grosser manner? And I think it's true for almost every strategy. Obviously, I have my own biases towards quantum investing, but for me, if I can't verbalize something in a quantitative way, then I really don't understand what I'm talking about. And it helps a lot to just examine my thoughts and say, what did I really mean by that? And that's true of any kind of balancing.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“Yeah, and I guess if I'm understanding you right, so the thing is, oh, when growth has gone up, we'll sell some growth and we'll buy value to always be balanced. But if, let's say, on an index from 1 to 100, when the index is at 100 growth is ridiculously expensive relative to value and when it's at one, growth is relatively cheap to value. If the index is at 70, you'd say, oh, I'm going to sell some growth to buy some value or something like that. The index could go to 71. It could go to 72. It could go to 78, 88, 98, 99, 100. There's nothing just because 50 is balanced. There's nothing that says that it can't continue to increase. And I mean, that's really interesting. And so you could further elaborate on that. Or what other sort of sacred cows can we slay of investing? If one of those sacred cows is, oh, I'm going to rebalance my.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT
“I did a backtest for that a few years ago, and in the particular way we were doing that, and I don't mean it'll necessarily be true for all versions of this trade, it just didn't make money. There was a lot of momentum in momentum trades where if it was going up, you'd leave money on the table if you rotated back into value, right? With value, there was a period where it just didn't perform very well for a long period. putting money, shoveling money into a strategy that continues to not work is not a good thing. So it's time dependent. It depends on the regime you're in. It depends on your precise strategy. And that's also a real benefit to systematic methods because it forces you to question exactly what you mean. It's easy to say, well, I want to balance growth and value.”
2023-07-13 · Forward Guidance · The Laws Of Quantitative Investing | Michael Robbins · IDENTIFIED FROM THE TRANSCRIPT