Babak wrote: Babak, I got a really good laugh at this. Thanks. Makes me feel a little inadequate that I haven't written anything which debunks EMH. Richard
simple, the 80% and 20% are constantly in flux. it's not the same individuals that make up the 80% losers and 20% winners. they change, crossover, and drop out constantly on both sides. surfer
Yes, the paper's main point is that retail position trader's performance is persistent -- a clear violation of EMH. What I thought was interesting was that winning traders earned about 12 to 15 basis points of excess return, while losing traders lost only 11 to 12 basis points of excess loss. Given the belief that few traders win, and most lose, I would have expected much worse loss rates for the losing traders (a left-skewed distribution). Instead the distribution appears to be right-skewed with more big winners, than big losers. Curious...... Best wishes, Traden4Alpha
Got ya. Sorry, I was a bit sleep deprived and slow . You do have a point. I'll try to look at that paper later this week if I find time. It could be a function of the way they measured returns and/or the sample they had. I'm working on something else right now that's really exciting and it's hard to find time for other things until I'm done with that one. Seems promising so far... I'll PM you if I find what I think I might.
I take your hand and lead you to the mountain side and show you the fossil in the layers of earth. I then ask you "If the earth is only 6000 years old, how do you explain this?" And you say, "To confuse heretics like you!" [sigh] So be it. For those that have not had their common sense beaten out of them by the blogeon of academia I offer the following: The January Effect: 1] My thread in the "Strategy Forum" 2] KCBT's strategy -- long Value Line Arith/short SPX [for more see www.kcbt.com or www.mrci.com] Why is it a fossil? Because .... 1] It persists year after year 2] It offers risk adjusted positive returns 3] Those returns are WAY above commisison/slippage costs Anyone who objectively and scientifically looks at those two examples of the January Effect and describes them as 'dents' in the EMH.....well...I dare say they themselves have a few dents in their skull. And also, I wrote a paper for my Hons. Bachelor of Business degree which also proved 'stickiness' or hot hands. The error that most academics make in surveying this topic, specificaly, is that their time horizons are either too short or too long. Since returns are mean reverting after some time, they then proclaim EMH is true. Got an A+ on the paper and the supervising prof offered to co-author it with me for submission to a finance journal.
You still don't get it. Yes, there are ANOMALIES, BUT, as long as you don't offer a unifying model that explains them AND the general behavior of returns cross-sectionally and time-wise, showing anomalies is like showing a pothole in a road and saying "whoever says the road surface is smooth has dents in their heads." By the way, if indeed the Janurally effect was so easy to capitalize on, it would be exploited away. The fact that it's not shows it ain't so freaking easy. Also, the beauty of EHM is that (besides the fact that it makes perfect sense, unlike the alternatives the likes of you offer (or abscence thereof)), is that it's self-healing. You find smth that really works, you start trading on it and it will go away. What I have found and am trading on is already showing signs of going away and I'm only moving about 10 mil. And that's in large cap stocks. By the way, the January effect has been shown to weaken lately as it has become more known. As for the project you wrote for your "Hons. Bachelor of Business degree", it's a JOKE! Submitting it to a journal? You gotta be freaking kidding me. That prof either needed to have his head examined or was just flattering you, most likely. I have seen my share of projects written even by the best honor's kids. The are just that, honor's projects. You have no idea what standards have to be met to even think about submitting it to even a decent B journal. A bad C maybe. The error in time horizons? What are you talking about? There are volumes of papers enough to fill the room you are sitting in examining hot hands in as many horizons as you could imagine and more. Naturally, if you mine the data long enough, you'll find some "not too long, but not too short" whatever horizon that seems to show persistence in positive returns. If you dig a 100 diff. such horizons, by type 1 error you are bound to find 5% of them will show what you are looking for at the 5% confidence level. FYI, what you said makes no sense. How is "mean reverting after some time" related to the fact that they supposedly skipped some very important horizon you were so brilliant in digging up? The only way to truly examine hot hands is with account level data. Is that what you did? I doubt it. The absence of logic in your "convincing" arguments only goes on to show what a joke those honors reports are. So you should really keep that story to yourself man. It isn't as cool as you think.
Whatever. Sounds like you never went beyond that bachelor's "degree." I thus rest my case. Dude, even after I got my MBA with perfect grades etc etc I still felt humbled by how freaking little I knew. I still feel the same way being a few months away from dissertation defence, and knowing hundred times more than I did back then. How someone with a bachelor's can act like he's the next Einstein knowitall (and "I can prove it all, b/c I did this amazing project for my honor's thesis that accounts for what those "scholastic" phd dudes were dumb enough to omit"...), will always puzzle me. Talk about overconfident traders. My contributions to this thread cease here.
vladiator, its interesting that your post is over the top boasting about your accomplishments and knowledge and you belittle others accomplishments. Yet, you judge my paper without even having read it. You simply dismiss it. I didn't act like an Einstein know it all. That is what you are doing. I simply stated a fact. That I had written a paper which proved stickiness....something which the EMH can't explain. I also provided two links, one for a strategy I outlined in the "Strategy Forum" which takes advantage of the Jan Effect as well as the KCBT's take on it. Did you take a look at them? No, rather you dismissed it without even investigating. Now that is truly the scientific process!! You dismiss the January Effect with the circular logic of "if it was real it would be arbed away". That kind of tautology may go over well in academia but not in the real world. It is as assenine as the old joke about the economist stopping his friend from picking up the $20 bill saying "If it was really there, someone would have already picked it up". I truly doubt that anyone who has access to and read the many articles and papers re EMH will come away calling them 'dents'. To each his own, I suppose. If you want to believe that the fossils are there because the 'devil' planted them to confuse us then so be it. Others are not so blind. For those that are interested to read these papers that I talk about a good starting place is "The New Finance" by Bob Haugen. Not only does it succinctly prove that EMH is baloney, but it cites a boat load of papers. And it is only 100 or so pages so you can finish it in one weekend.