Fooled by Randomness

Therefore the safest approach is to assume that XIV is a random walk and that technical analysis may not work for trading – excess returns may be illusory.

Thank for sharing the Vix strategy. :) Still, the author developed profitable rules (moving average) based on the random walk.

Posted by max.shi
 
Here's more information I stumbled upon recently on this topic ....quite an interesting blog the guy has, too....

Why So Many Overestimate Their Technical Analysis Skills

Posted on March 19, 2017 by
Many traders overestimate their technical analysis skills and this mainly involves those using old and outdated methods. However, in some cases technical analysis is used as a cover to promote wishful thinking.

After following closely technical analysis articles and social media posts in the last 8 years after the March 2009 bottom, I have come to the conclusion that the bears of the early stages of the stock market recovery are now the vivid bulls and the old bulls are now in the best case skeptical if not bears.
There were many articles from 2010 and 2011 by angry bears after two failed death crosses and expectations of a major top and return to 2009 lows. The bears believed the high profile economists who thought quantitative easing will lead to hyperinflation. Any technical analysis they used was in an effort to confirm this bias. I remember a friend in fund management was paying thousands every month to some analysis firm for crappy forecasts of a surge in inflation and the demise of the stock market. He now regrets he did not invest the money in the market.

But why do so many use simple technical analysis that no longer works? The answer is maybe simple: the analysis tools are easy to understand, use and are accessible by anyone. The result is noise. Most use technical analysis to justify their own biases and are the victims of selective perception, which is the dangerous tendency to perceive what one wants to, when analyzing information. There are always exceptions to this.

source: http://www.priceactionlab.com/Blog/2017/03/technical-analysis-skills/

He did an interesting Forbes interview ( here: https://www.forbes.com/sites/johnna...n-interview-with-michael-harris/#7602190c4733 )
 
To go back to the original post, if you are interested in random generation of plausible-looking stock price data, do have a look at mathematician Benoit Mandelbrot's fascinating book, "The (Mis)behaviour of Markets".

If the name sounds familiar to you, it's probably because he's better known as the father of fractal geometry; the famous Mandelbrot Set is named after him. However, he also did a great deal of work at IBM on stock analysis, likely before most of us were born.

In the book I cited he claims that simply generating random up/down movements isn't enough: that better results can be had by generating sequences which are "fractal in price, fractal in time". Sadly he doesn't provide any algorithms or code, but I guess that is, as they say, left as an exercise for the reader.

Posted by Martin Janzen
 
Top
Contact Us