tag:blogger.com,1999:blog-17232051.post3809256051398203074..comments2024-03-28T03:16:14.104-04:00Comments on Noahpinion: A finance industry insider on equilibrium asset-pricing modelsNoah Smithhttp://www.blogger.com/profile/09093917601641588575noreply@blogger.comBlogger17125tag:blogger.com,1999:blog-17232051.post-50324708418354163112015-07-23T02:13:00.105-04:002015-07-23T02:13:00.105-04:00I don't even think factor models help you hedg...I don't even think factor models help you hedge much.Factor models are of help in performance attribution ex-post, resulting in a better understanding of positioning and allocation at large. Trading is a tough business exposed to a very complex market<br />system. Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-76038602546836946832015-07-19T02:10:28.617-04:002015-07-19T02:10:28.617-04:00Shouldn't they be useful for hedging risks, th...Shouldn't they be useful for hedging risks, though? <br /><br />Factor models are also supposed to be useless for trading, other than helping you hedge out risks.Noah Smithhttps://www.blogger.com/profile/09093917601641588575noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-81580342317643638832015-07-18T13:30:47.614-04:002015-07-18T13:30:47.614-04:00Equilibrium models are designed to explain why mar...Equilibrium models are designed to explain why markets move with the economy; why stocks fell in the 2008 recession, and why risk premiums rose at that time, for example. They are not designed to help you trade. They are designed to explain what markets look like after all the trading opportunities have vanished. They are designed to be useless for trading, so it is no surprise that they are useless for trading. John H. Cochranehttps://www.blogger.com/profile/04842601651429471525noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-43581484281172559382015-07-17T12:44:54.713-04:002015-07-17T12:44:54.713-04:00"Equilibrium models are not newfangled gadget..."Equilibrium models are not newfangled gadgets that haven't been tried yet; they are something that has been around for a good long while and - unlike their cousins, the factor models - have not yet passed the rigorous test of industrial applicability."<br /><br />How many more decades will it take before academics understand that? <br /><br />I guess in their tenured positions they don't have to worry.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-59715182292988476362015-07-17T09:39:57.677-04:002015-07-17T09:39:57.677-04:00All models are only one point of view of reality All models are only one point of view of reality Martinhttp://oferty-kredytow.com/ranking-kredytow-mieszkaniowych/noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-17662985573471140812015-07-17T03:39:51.563-04:002015-07-17T03:39:51.563-04:00Your post shows you really have no useful thoughts...Your post shows you really have no useful thoughts. In the future you should keep them to yourself.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-39926452843869959772015-07-16T18:32:26.058-04:002015-07-16T18:32:26.058-04:00As always, the situation is much simpler. Any mode...As always, the situation is much simpler. Any model should have somen useful information to contribute. Some might do it only on very large time scales, which makes them useless for any decisions. However, if there's any info present, there's a trade to be had. If there's no edge you can produce from those models, that implies that they do not have any useful ( incremental) info.<br /><br />If any academic thinks that the academic models are unfairly maligned, the solution is very simple. Show us the money.Krzyshttps://www.blogger.com/profile/15794655390770135247noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-88639434356736256102015-07-16T16:38:05.754-04:002015-07-16T16:38:05.754-04:00BTW, of course that long run does not have much in...BTW, of course that long run does not have much influence on any active trading, so the basic practical irrelevance of equilibrium analysis in financial asset markets is pretty much correct.rosserjb@jmu.eduhttps://www.blogger.com/profile/09300046915843554101noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-20681173078609431892015-07-16T15:21:11.734-04:002015-07-16T15:21:11.734-04:00A few years ago I had a conversation with a trader...A few years ago I had a conversation with a trader who had a physics background. He argued that there are three relevant time horizons for traders. The shortest term is high frequency trading, really at its peak within a day and really far shorter than that, which essentially uses econophysics sorts of methods or variations thereof. The intermediate time period, which is at its best at maybe a half year interval or so, but clearly dominates from maybe more than a day out to several years, is behaviorally based. The fancy version now used by lots of algorithmic traders is to use rolling and declining abnormalities identified by massive data crunching. Put them in and then let them gradualy slide out with declining weights. Finally, in the long run, many years, neoclassical equilibrium holds, basically random walk.<br /><br />Barkley Rosserrosserjb@jmu.eduhttps://www.blogger.com/profile/09300046915843554101noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-42955550061844318252015-07-15T19:49:14.669-04:002015-07-15T19:49:14.669-04:00You may not know it, but I am a dumbass and I don&...You may not know it, but I am a dumbass and I don't always remember to do stuff... ;-)Noah Smithhttps://www.blogger.com/profile/09093917601641588575noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-43642605816614066592015-07-15T17:56:13.094-04:002015-07-15T17:56:13.094-04:00quote
A couple posts ago I wrote about equilibrium...quote<br />A couple posts ago I wrote about equilibrium asset-pricing models,<br />unquote<br /><br />you may not know it, but there is this technology called hypertext markup that lets you put a link to the old column in..Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-20399327881026038382015-07-15T15:23:58.598-04:002015-07-15T15:23:58.598-04:00It's really the other way around. You might be...It's really the other way around. You might be thinking of Cochrane, but he's the exception.Noah Smithhttps://www.blogger.com/profile/09093917601641588575noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-26272079274699117322015-07-15T07:21:29.551-04:002015-07-15T07:21:29.551-04:00Very nice. Its even inspired me to write a blog po...Very nice. Its even inspired me to write a blog post of my own. http://qoppac.blogspot.co.uk/2015/07/equilibrium-asset-pricing-models.htmlRob Carverhttps://www.blogger.com/profile/10175885372013572770noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-4683473798184858182015-07-15T06:55:54.086-04:002015-07-15T06:55:54.086-04:00On the equity side, my finance went unused. In fa...On the equity side, my finance went unused. In fact, algebra went unused, except for a couple two variable systems that showed up. Logs were useful for a while, but they devolved to rules of thumb. PV calculations have been consistently useful. <br /><br />These fixed income guys, though, they use terrific leverage and hedging. Both force a short orientation, as does the compensation cycle. It does seem that there should exist long oriented fixed income shops working equilibrium models, no leverage, sufficient net worth to not worry about compensation. <br /><br />I think it was 05-06 when Warren Buffett basically said, okay all you hotshots, I'll beat you with one stock, and it was an industrial REIT, close to a bond. The equilibrium analysis behind that was along the lines of "these valuations cannot go on forever, therefore they will stop", a la Herb Stein. That's a long term insight, but does not require much math. <br /><br />johnhttps://www.blogger.com/profile/01628063779348336042noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-17684597860050617442015-07-15T06:03:19.431-04:002015-07-15T06:03:19.431-04:00Sad. The finance economists don't even succeed...Sad. The finance economists don't even succeed within their core field, but that doesn't stop them from infringing on macroeconomics.Alexander Sebastian Schulzhttps://www.blogger.com/profile/15135338616598357444noreply@blogger.comtag:blogger.com,1999:blog-17232051.post-1026218593845398422015-07-14T23:50:49.713-04:002015-07-14T23:50:49.713-04:00When you're working with a complex system, any...When you're working with a complex system, any model that is good in some respects is going to be bad in others. Two modelers who wish to put their models to different uses will weigh those plusses and minuses differently. A simple conjecture is that academic economists want models that generate understanding, while industry economists want models that predict the future. Other things being equal, both would like to do better on each of these dimensions (and countless other dimensions). But when push comes to shove, a rough and ready regression often beats a theoretical model when it comes to making predictions, and a model built on optimizing agents with defined constraints often beats a regression when it comes to generating understanding.Ramnoreply@blogger.comtag:blogger.com,1999:blog-17232051.post-60422508918736971042015-07-14T22:31:32.913-04:002015-07-14T22:31:32.913-04:00Good one, thanks. In truth the industry people who...Good one, thanks. In truth the industry people who "just took a look at equilibrium models and said 'Hahaha no way!'" outnumbered those who "understood how the models work, tried to use them, thought deeply about the content of the models, and also thought deeply about the risks of using them." But the industry does hire no small number of people steeped in econ and/or finance theory and those people do try to make as much use out of it in finance as they can.Tom Warnerhttps://www.blogger.com/profile/11247836188106712069noreply@blogger.com