Search results

I should have been more precise, "randomness" is a term too broad and I wasn't clear enough. So here it is: Markets sum up to swarm behaviour. Ninna

You are terribly wrong. Your (very bad) example might apply under very strict conditions to extremely illiquid markets. In the bigger picture, markets involve so many participants with so many conflictual interests that their behaviour pretty much sums up to randomness. Which is a great...

There are several positive expectancy strategies for RW's. But as I said, you're not quite ready to grasp it yet. I can now see more clearly where the problem comes from, you are mixing up randomness and normal distribution. I invite you to work on your knowledge of probabilities. I can't...

You stated: And I said that as a prerequisite for this discussion you should understand how edges exist in "pure randomness". Ninna

This discussion is over your head if you think that "edge" and "randomness" are mutually exclusive. Ninna

Well well, Maybe you ought to do the tests by yourself as I suggested.. See attached, auto-correlation as a function of the lag. Note in particuliar the 25% correlation for a lag of 1. I used SPX minute data if you want to reproduce it. Ninna

Test it. Most markets have a tendency to trend more than coin flips. Test it and you will see. Ninna

Based on normal distribution. http://en.wikipedia.org/wiki/Trend_estimation Ninna

Most markets have a tendency to trend more than normal. So yes, there is a (small) edge in trend-following. However there's a bigger edge in the bullish market bias in the long run. Ninna

What is your trading system based on? All reasonable trading strategies, when you strip them down, have an underlying distribution that you bet on. Let me give you an example. Let's assume that I have a RTM strategy. I compute a spread between security A and B and when the spread is...

There you go: http://en.wikipedia.org/wiki/Stationary_process Ninna

The absolute mean deviation ratio: d=sqrt(2*N/pi). Where N is the number of throws. Ninna

How do you know? Did you test it? I did and I disagree with your statement. :) How can you justify noise on some timeframes but not on longer ones? What about the fractal symetry? Ninna

My points were 1. Securities actually behave very much like coins most of the time. Test it and see for yourself. The fact that millions of heterogenous groups play the same game (and the structure of the feedback loops) precisely make the market behave like that. 2. Even though coin...

This is curve-fitting disguised. It works for some problems but not for the stock-market due to the non-stationarity and non-normality of the price distribution. You might stumble against a stable distribution but the chance is extremely small. And if you do, will you be able to identify it...

Nevertheless, even for a coin flip RW with p=0.5 you can compute the envelope that bounds the RW. And you will see that, on average, the throws fluctuate around that envelope. And this is where it gets interesting... (regression to the mean, anyone?) :) Ninna

What is your point? Throw a coin and you also see trends... Ninna

If the underlying distribution that you use for placing bets isn't stationary your system is doomed to fail. Therefore unless you can guarantee stationarity max DD is total ruin because the fat tail will always get you in the long run. (And the long run might be in 25 years as well as...

No. Ninna

The difference between the market skew and normal is basically your profit. You need to find a way to measure that difference. Ninna