I agree with your second point, but not your first .... I'll come back to changing systems later ... as for 'going with the herd' ... isn't that momentum trading, the one that tests OK over long periods of time, as per the study papers posted?
Going with the herd is OK, if they are heading for free coffee and donuts, but not good if they are jumping off a cliff? Re changing systems: I am actually going the other way and pursuing the HolyGrail of systems .... the one that never fails, but I will come back to that. Howards choice of the term 'market inefficiencies' constrains our thinking somewhat, IMO. In fact it carries the seed of it's own destruction in it's meaning, so if we accept the terminology we accept that traders can trade the inffeciency back to efficiency i.e. make it disappear. Efficiency implies that something that has been done has not been done in the best possible manner ... in trading I take that to mean that the activities of market participants leave behind arbitrage opportunities. If they do actually exist then yes, they will be traded back to financial entropy (surely those with the lowest commissions and the fastest computers will win this game?) However, this is not the only game in town and not the game that the majority of 'traders' are playing because the majority of traders don't have the fastest computers and cheapest commissions. Another opportunity for trading could come about if significant non-random patterns exist in market prices. It is a rather spurious argument to claim that these represent 'inefficiencies' .. they are just a market going about its daily business ... tough luck for the guys,or girls, who wrote their thesis on the Efficient Market Theory ... they can call it inefficient if they want but do traders really care ... to us it is just todays market ... they are in the (discredited) minority who say that the stockmarket Gods decreed that the market 'shalt be efficient'? As per your comments, if traders operate in and around those patterns, with volume, they will simply create new ones ... the markets are closed financial systems == a zero sum game (one good argument for those who say we should apply discretion when applying our trading rules?) ... so these patterns are not traded to effeciency, they are just traded to another morphology, which is definitely non-random because it is caused by the action of traders. Another class of trades is time based trades (if in fact they do exist) e.g. say, a commodity goes up every year in the autumn because of some seasonal pattern that affects crops ... trading a time cyclic pattern, like that, will also erode the trade but once again are we justified in calling the cylic pattern an inefficiency in the first place ... the price behaviour is caused by people going about their normal business and traders simply take advantage of any opportunities that are provided to them. There is another class of trades,if they do exist, that will never be traded away .. these is are behaviours that are endemic to the markets e.g. volatility around the open ... if a trader can design a system to take advantage of something that occurs every single day they can be confident that the opening volatility will still be there, to take advantage of, in the next decade or two. IMO momentum, begetting momentum, is endemic to market behaviour, as is the 'endings' of momentum and that is the reason why it defies the axiom that 'trading the edge erodes the edge' ... trading momentum simply leads to more momentum! --- In [email protected], "Jerry Gress" <pleasenospample...@...> wrote: > > Hello, > > This thread reminds me of a private tour I had many years ago on the Pacific > Stock Exchange (San Franscico), near the pit I pointed to a 'kid' on a > computer and the guide said he works for one of the big players tracking the > latest indicator so that the big player knows when the 'heard' is going to > jump in so he can jump out! > > Also, my claim to system fame is "selling" a TradeStation system, putting in > as part of the code an expiration date 1 year down the road. Of the two > customers one called back one year for an upgrade, gave him another two > years, and never heard back from both:) Moral is people will change a system > over time or just go on to the next one thus all systems fail. > > Regards, > > Jerry > > -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf > Of brian_z111 > Sent: Tuesday, June 02, 2009 9:16 AM > To: [email protected] > Subject: [amibroker] Re: Do all trading systems stop working? - Howard > Bandy's book > > Thanks for the feedback. > > Samantha's question fired off my search algorithms for sure and they are now > zeroing in on the US institutions... I think that is where the answer lies > to the question of 'eroding the edge' e.g. if I am trading the SPY, and buy > on a system entry at 0930 US time, there is so much volume going through > that instrument, at the same time that I am buying, what difference does my > little bit make and who knows what is motivating every single dollar? > > My assumption is that the vast majority of money going through the US > markets is controlled by institutions and that in most cases they would not > be the least bit interested in my trading systems, or theories for one > simple reason, .... they are restricted officially by due diligence etc and > unofficially by their clients tolerance for risk .... I am far to extreme > for their tastes. > > No matter how well a manager invests the money they can't stop a run on > their funds when the fund, or the indexes are dipping markedly, so they are > forced to sell 'under their clients orders'.... I guess that played a big > part in last years debacle. > > The only interest in 'trading systems' would come from boutique funds (hedge > funds etc) so it depends what % of the total US market they are controlling. > Even then there is so much literature out there I doubt very much that they > would ever stumble on my posts let alone take them seriously. > > You are quite right though ... I am going to do some more homework on the US > funds ... I have been underestimating them and some of them are far more > interesting than what I thought, up until now. > > I dare say a lot of the boutique funds got their butt kicked last year too > because of the mistaken believe that they were diversified into > non-correlated markets/systems.... I will have to look into that also. > > I have to say, though, that I am more afraid of my broker getting in front > of me, which would definitely clip my take, than I am of sharing my systems > with a bunch of investors/traders/fund managers. > > brian_z > > --- In [email protected], "Ed Hoopes" <reefbreak_sd@> wrote: > > > > There are publicly traded funds organized around various trading systems. > Below are a few for comparison: > > > > NFO - Insider Info > > STH - Stealth > > XRO - Sector Rotation > > PIQ - Magni Quant > > PSP - Private Equity > > FVI - ValueLine 100 Stocks > > BWV - Covered Calls > > CSD - Spin Off Companies > > DEF - Defensive Stocks > > EZY - Low PE Ratio Stocks > > > > Now take each one of the above and do a relative performance to the > overall market - like VTI Vanguards Total Market ETF - and you can see how > well they work. > > > > NFO, PSP, EZY top the list with a modest out performance using my ranking > algorithm. The majority equal the market or underperform. > > > > For me the most disappointing is FVI only as good as the broad market - so > much for $650.00/yr fundamental/technical analysis newsletter. XRO - is the > worst. > > > > ReefBreak > > > > > > > > --- In [email protected], "brian_z111" <brian_z111@> wrote: > > > > > > Hello Samanatha, > > > > > > Thanks for your post ... a good topic and thanks also to D and PS for > additional leads and others for the discussion. > > > > > > <snip> .... all trading systems will stop working forever at some point > (because the inefficiency in the market they exploit will be killed by > everybody jumping on board).<snip> > > > > > > This point of view isn't shared by all traders. > > > There are at least two grounds for objection: > > > > > > - the massive number of possible permutations, at any point in time in > the market, make the chance that two traders are doing the same thing with > significant amounts of money are unlikely e.g. Aronson puts forward this > idea in his book, "Evidence Based Technical Analysis". > > > > > > - based on the behaviour of market participants it is also unlikely that > a significant number of traders will trade exactly the same trade even if it > is "published in the Washington Post" e.g. one of the Wizards interviewed in > one of Schwagers book's argues along those lines when he is asked if he is > reluctant to talk about his trading methods. > > > > > > Take this topic for example ... how many people read the topic ... read > it carefully ... read the links ... thought about it ... did some homework > ... go on to study the system ... put it into practice (without changing > anything) and then go onto to trade it in the same market, same instruments, > same timeframe etc with significant amounts of money. > > > > > > I consider myself to be a trend trader but my definition of a trend is > unlikely to be used by more than a handful of people ... the chance that > others are watching the same trend, in the same instrument and the same > timeframe is almost zilch. > > > > > > The caveat is if and when large institutional traders are systemic > traders and/or algorithmic traders .... perhaps large players can mop up > systems if they are interested enough to do so. > > > > > > There has been little discussion, on this board, about systematic > trading by institutional players. > > > > > > Siddhartha did say he didn't observe that the practice was widespread in > his time in the industry. On the other hand I recall reading an article that > said Goldman Sachs were into algorithmic trading in a big way. > > > > > > As an aside ... I thought that the axiom "We will miss most of the > growth if we miss the 10% biggest gain dayss in the market (ditto for a > weekly/monthly/yearly basis etc) was basic (same for missing most of the > losses if we avoid the worst ten%). > > > > > > Looking at any index chart, with hindsight, it seems obvious that there > are several points where any number of indicators could have told us to get > out and we would have been better off ... the trade off is the cost of exit > and re-entry. > > > > > > I put a lot of effort into investigating that payoff/versus cost when > deciding how often to trade (buy and hold versus, say, short term or day > trading). > > > > > > I was surprized last year when so many in this forum (of all places) > seem to be hurting. > > > > > > > > > > > > Re Momentum trading: > > > > > > There are two articles here on trend trading (scroll down to 3.1a and b. > > > > > > > > > http://zboard.wordpress.com/library/miscellaneous-articles/ > > > > > > Michael Covel appears to be the current king of trend trading (I like > his book but not his videos). > > > > > > www.TrendFollowing.com > > > > > > > > > How do we know when a system is failing? > > > > > > We can't get a math measurement to tell us when that momement has > arrived ... all models assume stationarity and as soon as it is broken we > are in unknown territory .... classically a shift in the average value or > the dispersion (of the trade series) signifies non-stationarity, although > random data series contain a good deal of variance and it is hard to > distinguish random variance from a system breakdown. However IMO most > traders are trend traders and almost anything will work while we are on the > right side of the trend .... so in the real world a system is broken when > our assumptions about the underlying trend are incorrect. > > > > > > > > > --- In [email protected], "samu_trading" <samu_trading@> wrote: > > > > > > > > All, > > > > > > > > In his really good book Quantitative Trading Systems, Howard states > that all trading systems will stop working forever at some point (because > the inefficiency in the market they exploit will be killed by everybody > jumping on board). > > > > > > > > On the other hand you have momentum / ROC based systems working > forever now, same for trend following MA crossover systems like The one > propagated by Mebane Faber. Momentum and MA rossover trendfollowing does > seem to work "forever". > > > > > > > > Any comments from the gurus here? > > > > > > > > Thanks, Samantha > > > > > > > > > > > > > > ------------------------------------ > > **** IMPORTANT PLEASE READ **** > This group is for the discussion between users only. > This is *NOT* technical support channel. > > TO GET TECHNICAL SUPPORT send an e-mail directly to > SUPPORT {at} amibroker.com > > TO SUBMIT SUGGESTIONS please use FEEDBACK CENTER at > http://www.amibroker.com/feedback/ > (submissions sent via other channels won't be considered) > > For NEW RELEASE ANNOUNCEMENTS and other news always check DEVLOG: > http://www.amibroker.com/devlog/ > > Yahoo! Groups Links >
