Steve, is this what you are referring to on the last bit of your post? http://www.gummy-stuff.org/FunnyPics/nosex.jpg
Yep, thats one of them. From my point of view, to make money, one has to get way past this "debate" and move on to the next stage. I learned so much from importing intraday data into excel and changing it from absolute price into price jumps. Once I started to do that, I noticed patterns (size of jumps) and runs (persistence of jumps of a certain size). Then as I did it for a longer period, I started to see what I would characterize as "imbalances". Then when you refer back to the charts, you can see the price points where buyers and sellers decide to lift the offers, and hit the bids. Once that happens, all you have to do, is wait there in the tall weeds for that price to come around again. Basically that is what I do every day. For me trading is a matter of having the discipline to wait for the setup. Good luck to you Steve
steve, having reread your posts, in the spirit of mutual self improvement, i see that we are in more agreement than disagreement---- it's primarily definitions and slight interpretative views were the difference lays. good trading to you ! surfer
I often see reports that a great trend following guru living in this resort of another just can look at a chart from across the room and determine if its going up or down, and then the better it looks , the more he buys or sells. These pleasant reflections on reports were brought home again by the note on this list that a 9 y ear old child can see the trends on charts and tell you that they exist. Such youtfhful and resort oriented observations must be vetted against the fundamantal property of random walks, i.e. that the variance of their price changes is a linear function of the number of time intervals that are observed. In other words the variance of a 100 day change will be 100 times the variance of a 1 day change for random walks, or in elementary terms the variance of a sum is equal to the sum of the variances. When one looks at a chart one sees big changes over long time periods, and unless one has the trained sensibilities coming from working with random number generators, reading of the random charts generated by Roberts or Working in their papers on this subject, or has read a statistics book recently such as Snedecor that elicits this common tendency of humans to believe that trends exist in random phenomena because of this tendency, observations of kids and gurus living in resort areas, and anecdotal observations about one's profits, or the lack thereof of those who dont explicitly believe in trends without testing of such standard measures as those like serial correlation coefficients, and r uns tests ( of which I have provided quantitative estimates for stock indexes in my previous posts, in contrast to the completely qualitative and ad hominem attacks that are all too frequent here), should at best be considered as working hyotheses and self reported anecdotes subject to testing and refutation. proturf
Good question proturf. But why bother with all this. It is silly trying to apply concepts of stochastic processes to areas where they clearly don't make sense. Brownian motion is a phenomenon that can be observed and measured. Market behavior? Nobody seems to have any clear idea at all about WHAT exactly is supposed to jump around like "random walk". One issue like IBM, GM? The DJI or S&P500? The 30yr bond futures? EOD, 1hr bars, tick bars, ticks? It's a cheap gimmick used by academic savvy entrepreneurs to bamboozle credulous investors, making these believe the former understood and know something they don't - "Scientism" at its best . You could as well maintain that profits arising from the advice of such quacks, academic or not, truly follow "THE LAWS OF RANDOM WALK". Indeed, 9 year old kids or even monkeys will probably perform at par with these random wizards. nononsense