Kent Daniel: Price Momentum

uh it's a pleasure to be here thanks everybody for coming I'm glad to see everybody so excited about uh about this so what I'm going to talk about today is um is a particular quantitative strategy that is often employed by hedge funds and in fact it's employed with a lot of Leverage this is going to tie in very very well um with the topics that professor sundas and Professor Wong are talking about and what I want to do is uh introduce the topic of quantitative investing tell you a little bit about that and then dive in particular into this one strategy and this is related to some research that I'm doing on trying to understand what it is that's responsible for Price moment in other words why do you see it in markets because I'm going to make the argument that the market isn't completely efficient and in fact my view is that the reason we see this particular phenomena is because investors don't process information perfectly so that's the reason that markets aren't particularly aren't perfectly efficient but I want to tell you uh more today just about the characterization of momentum and give you a few thoughts on uh what's underlying this so it turns out as um as I was walking through Penn Station about a week ago I saw this on the cover of uh of the Bloomer magazine you can see this is the November issue and it turns out this is perfect because this is a guy whose name is cliff asnis and there's a great story actually in in this magazine about his particular firm which had a very rough time the last couple years and it's how they're fighting back okay turns out they had a rough time because of momentum okay and so I'm going to tell you why that is now um uh let's see it turns out they haven't given up on momentum and in fact this is from inside the uh the magazine in the article since it's Bloomberg magazine they tell you how you can go onto a Bloomberg and look at the performance of aqr's funds they it turns out they've recently launched a set of mutual funds so in addition in addition to to playing in the hedge fund Universe which they've done for some time now um they're also launching some mutual funds and if you can see this probably a little bit small but you can see one particular fund that's highlighted here is the aqr momentum fund okay so what aqr does both with their mutual funds and with their hedge funds is they use a set of mathematical rules to generate to put together large well Diversified portfolios that have a certain tilt and it turns out for momentum basically what they do is they bet on winners so they buy winners and they sell losers okay and what we're going to do is dive in and see exactly what that means now it turns out um to introduce the topic this is something from a paper that's written by three guys uh asnis moscowitz and peton in fact asnes is the same guy whose picture you just saw the other two guys are professors at NYU and uh in Chicago respectively and what this particular particular paper does is it looks at the presence of momentum in different asset classes okay so this page I'm not going to spend much time on these but the green lines here represent the cumulative returns to Value momentum and combination strategies I.E if you combine value and momentum and what's pretty interesting here is you can see momentum at least historically and in fact I should mention leading up to the date at which this paper was written which was 2008 early 2008 um momentum did very well in the US in the UK in Continental Europe and not so well in Japan which is actually another interesting topic this chart shows what if you go to different source of strategies this is um let's see this is an equity country selection so in other words make bets on whether the us is going to go up relative to the UK relative to Continental Europe relative to um to Asia based on how well those firms how well those countries have done recently okay this is Bond country selection so the same thing for government debt for the different firms momentum doesn't work as well here but it's still still positive this is foreign currency okay so in other words do you bet on the Dollar on the Euro on the Yen on um you know various currencies around the globe based on how well they've done recently turns out momentum Works quite well for currencies and then finally this one is I believe for Commodities though the headings seem to have fallen off here and in fact commodity momentum is another very successful strategy okay so what is it that's driving momentum well just let me just give you a quick um uh one minute guess or theory on what it is its driving momentum uh I actually had did some early work on this and um the thing that appears to be the best I think the best supported Theory as to what's driving momentum is that investors ERS get information okay they receive information and this can be from earnings announcements for individual firms it can be news about the uh particular country and the economic prospects of that country investors we know have a status quo bias they tend to you know think things are going to stay roughly the same and um when this new information when they see it they think okay it's going to make a difference and the price will move in response to that new information but in general it won't move enough for lots of Behavioral reasons and in fact there's some uh I had lunch yesterday with uh Professor elki Weber and I know she and Eric Johnson are working on some really interesting behavioral theories as to what drives this okay so there's a lot of neat work going on around Colombia on this specific topic but anyway the information isn't fully incorporated into prices investors don't immediately realize they don't immediately understand the full impact of this information so the price moves a little bit and then it takes a long time a year or so for the information to be fully reflected in prices okay so if you as an arbitrager as somebody who's kind of making bets in financial markets you see a price move and it turns out interestingly in equity markets this is something we're not going to talk about but if you see a price move and you can link that to information then it's likely that that price move is going to continue and of course momentum in physics means you know something that is moving along in a certain direction okay at a certain speed is going to keep moving at that speed unless something something stops it right that's exactly the phenomena in PRI in financial markets when prices start moving they keep moving for a while okay so that's the idea of momentum now how do you build a momentum strategy so I thought it'd be useful here to talk about what if you want to put together a simple momentum strategy you want to do with what cliff asnas and aqr do on your own how would you do it so this is a a chart that I put together that shows how to build a momentum strategy okay and I should somehow I thought when I put this chart together this color would look really good I can see I was wrong there but we're going to have to go with it okay so let's say you're sitting at the beginning of April 2009 okay so it's what's 30 March 31st I guess okay and what you're doing is you're building your portfolio that you're going to hold over the next month over April 2009 okay well one particular momentum portfolio that you could build would be a 122 moment portfolio what does that mean well what you'll do is you'll take every single stock let's do this in the US just to keep it simple okay so in the US what we're going to do is we're going to look at the monthly returns to every stock that's traded on the nysse on AMX that's listed on NASDAQ and we're going to look at their returns from 12 months ago to one month ago so that's going to be if you label the return for April month T that's the portfolio we're building right we want to take advantage of any price movements that occur in month T in April what we're going to do is we're going to use all of the Returns on these individual Securities from one year ago up until one month up until one month ago okay we're going to sum those up for each of the individual Securities and then all we're going to do is we're going to rank these Securities we're going to say okay which one had the highest return over the last 12 months minus the last one month and which ones had the lowest return okay now what we're then going to do is we're going to say okay let's take the top 10% of the firms so there roughly how many firms are there that are listed say across the NYC AMX and NASDAQ about 5,000 yeah so you're going to take the top 500 firms okay those that had the highest return over the last 12 months and you're going to put those in a portfolio you're going to do what's called valuate this portfolio what that means is you'll put more you'll buy more of the large market cap firms and you'll buy less of the small market capitalization firms okay so the top 10% of the firms we're going to call the winner portfolio and the bottom 10% of the firms we're going to call the loser portfolio Okay now again with this idea that there's momentum the winter portfolio If there really is momentum in returns that's going to keep going up right because of momentum it's been rising it's going to keep Rising the loser portfolio is going to keep going going down Okay so we're going to look at these two portfolios we're also going to look at another portfolio which is a long short portfolio Okay some of you have probably heard this term it's it relatively straightforward basically with a long short portfolio what you do is you short a dollar's worth of the loser portfolio again this is based on the idea that the losers are going to keep losing right so what we'll do is we'll borrow those shares and then we'll sell them okay and then of course at some point in the future we got to pay back whoever we borrowed the shares from and the way we're going to do that is we're going to buy back the shares and if the price really went down we'll be able to do that at a lower price than what we sold them at and therefore we'll make money on our short position okay so our goal here is going to be to you know buy these winners hopefully they'll keep going up sell the losers hopefully they'll keep going down okay now for this particular strategy it turns out that what we're going to do is every month at the beginning of the month we'll rebalance the portfolio so what that means is we'll re do our ranking period returns okay our formation period returns and basically for example when we get to the beginning of May 2009 what we do is we' you know take off this April return and add on the March return and then we'd have a new 11-month return and we're going to rank on that do the whole thing again take the top 10% the bottom 10% and when one of the things that you can see is with this particular strategy there's not going to be that much turnover in the portfolio right because the returns over this period are going to be pretty close to equal to the returns over this period but obviously once you go out 12 months the returns are not going to be very closely linked because those returns uh one year apart are likely to be to be quite different okay all right I was going to ask any questions so far but I can't do that right all right so let's look at how well this strategy works so I'm going to look at the period here from 1949 so basically post World War II up to 2007 so which you may say well why didn't you go to 2010 I'll tell you in a minute okay so this shows first what we want to see is how well we would have done with investments in different assets so this first shows what if You' put your money in t- bills okay and specifically and I did things in logs here so my apologies for those of you who don't understand this but basically 10 to the zero is one everybody remember this kind of yeah okay 10 to the zero is one 10 to the one is 10 this is 100 a th000 10,000 so this shows if you'd invested $1 in TB bills turns out by the time you get to the beginning of 2007 you're up to $15.73 pretty good right okay what if you'd put your money instead in the market so in basically like the S&P 500 this is the valuated crisp index it turns out but you can see how stocks have done really well typically historically relative to T bills and that's reflected here so with an investment in Common Stocks you would have had $741 197 Okay so what if you'd bought instead of buying the market You' bought that portfolio of losers remember the worst 10% of the stocks over the last year well it turns out instead of being up $741 197 you'd be up to from $1 in 1950 or 1949 you'd be up to a188 okay so the returns of the losers are abysmal okay they're basically they're less than 1% per year okay and quite a bit less than t- bills okay so that's pretty impressive there does seem to be this momentum basically the things that have been going down tend to keep going down now let's see what happens with the winners turns out we see the same thing with the winners okay so there's a in fact here just a remarkable difference the winners are actually up to instead of $741 you're up to $ 4,131 by the time you get to 2007 from a $1 invest perfect okay um so what do we see well now let's look at the excess returns and basically what what we saw in the last chart is reflected here basically the momentum has a mean return of 16 A5 per. so this is the difference between the winners and the losers the difference is actually over this period 16 a half% per year which is a huge number right the mean Market return over and above T bills is 7.7% okay um the volatility of the momentum portfolio is a little higher but not much the sharp ratio is quite a bit bigger okay and moreover the beta of the momentum portfolio is actually negative so it's negatively correlated with the market okay what this means is if you put them together into a portfolio you combine the market and this momentum portfolio you're up to a sharp ratio of 1.02 compared to the market of 0.53 so if you build a portfolio that has the same volatility as holding the market you would have gotten a return that's just about double okay so it's remarkably successful okay now let me show you what's happened recently with momentum okay because you may have wondered before why didn't he show us the returns from 2007 on so this here you see really really good returns again right except for one problem this is actually the past losers okay not the winners these are the winners down here so our winners and losers have reversed okay so what happened well so this I actually started this chart at a very opportune date March 8th 2009 anybody know what's special about March 8th 2009 Market reversed okay so it turns out the market fell from roughly 2007 the end of 2007 early 2008 the market started falling and it fell a lot it fell by over 60% okay and actually the day on which the Dow got the lowest over that period was actually March 8th okay so starting March 9th it started turning around and it turned out when it turned around basically those losers started doing really well now what kind of firms were there in this loser portfolio lots of financials okay so it turns out city was in there okay um there were also a bunch of Highly levered firms it turns out there's a firm called International Paper that was in there okay International Paper has a lot of Leverage and it was looking really really shaky in the middle of the financial crisis and people were afraid that it might go under well it turns out once the market started rebounding International Paper did really well the winter portfolio had a lot of firms in it one particular firm was a firm called AutoZone does anybody know what AutoZone is they sell like used used car parts things like that right do those do firms like that do pretty well in a recession yeah so it turns out AutoZone had done really really well as the market fell but when the market rebounded it did okay but it didn't do great so there was a big difference between the losers and the winners so here if you had a momentum bet on you got crushed absolutely crushed okay so when you see on the cover of the Bloomberg magazine say where it says aqr fights back this is what they're fighting back from okay now let's look back and see do we see anything similar in other historical periods okay so here I picked another propitious uh period which is June 1932 to December 1945 okay so what's what's why did I pick June 1932 any guesses what happened in like 29 through 32 did the market do well yeah not that well actually so it turns out the market over that period anybody know how how far down it was turns out about 90% so the Dow in September of 1929 was at 381 at its peak and it fell to 44 okay so if you think the crash of you know 2008 2009 was bad 32 was a lot worse so I started here in July okay or in June I guess and what you see is we see the same pattern so basically this line that goes way up it's not the winners it's the losers and the losers you see just a phenomenal rebound okay and in contrast the winners and by the way this is a long period right this is a long long period this is about what 13 years and the losers do much much better than the winners okay so here's some numbers okay so this is the 32 period basically let's look at just the two months following the market bottom the market was up by 82% over those two months the losers outperformed the winners by 206% so the losers gained 2 136% the winners only gained 30 in our more recent period the three-month period from March through May the market was up by 30% so nowhere near as much as this and by the way when the market Falls 90% And then goes 80 up 82% is it only down 8% okay good good to see yeah so the market was still way way down from the peak right but um it turns out when it bounced back the dramatic gains were all for those losers all for the stocks that had really been pushed down that's what we saw more recently in 2009 too okay um all right now ah I was supposed to ask something here but I guess I'm not allowed to ask questions anyway so let me tell you something one of the things you should when you're looking at this you should be thinking is well wait a minute if the Market's going up by 82% these losers are going up by 236 per. and here the Market's going up by 29% and the losers are going up by 149% what's one possible explanation of this yeah oh wait I'm not supposed to ask questions I keep forgetting this I'm sorry okay so one possible explan I think somebody said it there is they've got a high beta right so let's look at this I'm G to skip that slide and here what I do is I show I've plotted I've calculated using a set of rolling regressions okay the beta is for the losers and for the winners so all I do is I run a regression probably most of you have seen this where you regress the returns of this portfolio the excess returns of this portfolio on on the market and what you see is particularly in these periods following the big Market down moves the betas of those losers get really really high what's driving that well it turns out it's two things one if the Market's gone way way down and you've got a bunch of firms who have gone way way down probably one good reason why they've gone down is their High beta stocks to start out with okay the other thing is that as these firms fall okay um it's if they're losing that much money one of the things that Professor sundon is going to talk about is leverage well firms have leverage too right they have a lot of debt and as their market capitalization Falls and falls and Falls they become more levered and their beta shoot up dramatically okay so it turns out that the betas of the momentum portfolio become very very negative why well because to build a momentum portfolio you're buying these winners and you're selling these losers you're selling the very high beta stocks so what this means is for the momentum portfolio when the market starts shooting up it's going to really shoot down those losers are going to do well okay and since you're shorting those losers that's a bad thing okay so one of the things we could oh and by the way this is the same uh similar chart but for the more recent period here you can see the effect is if anything even more dramatic here the betas of these loser portfolios and by the way this is after the collapse of the tech bubble the betas are up to like three three and a half and in the more recent financial crisis roughly the same level one of the things we can potentially do is hedge out this Market risk so let's see what happens if we do that so this plot shows now this unhedged line and by the way here I'm I'm making a little bit of a jump these are the cumulative log returns for the unhedged momentum portfolio that's the green line and for the hedged momentum portfolio which is the blue line so basically what I'm going to do I know this thing has a negative beta with the market okay I know that when the market goes up it's going to go down so one of the things I can do is I can hedge that out how do I do that well I add a little bit of Market exposure to my momentum portfolio right and if I do that just right what that'll mean is the movements of this portfolio are now insensitive to the market okay so I'm sorry I'm gonna running out of time here so I'm going to jump over this quickly but you can see you can see but what happens is this actually makes a big difference okay it turns out the cumulative return here is uh is about minus 50% for this unhedged portfolio by the time you get to 49 and this you can see is actually slightly positive this is up about 30% so it actually makes a big difference and um if you look over the longer period we don't get quite as much benefit from the hedging and the more recent financial crisis but there's still something there okay now one of the things that's interesting and this is what actually I'm working on now which unfortunately I don't have a whole lot of completed research to to show you but it turns out that even after you hedge this momentum portfolio there's still a lot of predictability and basically momentum does lousy when you're in a bad financial time when people are scared basically when the market has gone down so what's one other strategy you could try well one would be to actually reverse your bets on momentum when the market has been down okay so what I do now is I look at a strategy where if the Market's been down for two years you reverse your momentum portfolio Okay and it turns out if you do this what you see is this is the dynamic strategy and you can see it does a lot better now it by the way it doesn't look like it's that much better right because you you look at this and it's only this much higher that can't be very much well it turns out the profits from this strategy if you put in a dollar here this is the standard momentum strategy you're up to 387 pretty good by the time you get to here if you'd hedged it you're up to $1,421 and if you'd gotten out and in fact reversed your momentum bets when the market over the last two years has been down it turns out you're up to $36,000 by the time you get here so a dramatic difference okay so um what's causing this well it seems like this idea that people get fixated on their views and therefore don't respond to new information particularly bad information about bad firms that have done poorly doesn't seem to work when times have been really really bad things seem to reverse okay so anyway I'm doing more work on that and once that work's completed hopefully I'll have a paper pretty soon uh I will post that and uh you can all see it okay thank [Applause] you

Channel: Columbia Business School

Views: 53,083

Keywords: Price Momentum, Kent Daniel, Columbia Business School, Asset Management

Id: 2w1OYDeTfsk

Length: 24min 57sec (1497 seconds)

Published: Fri Jul 15 2011