Market phenomena - page 42
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
But I wonder: don't you classify your poking around with Eview as cleverness? In that sense, you are a shining example, too! Maybe even better than Gaidar - he didn't have Eview, and Hedrick and Prescott weren't Nobel laureates at the time.
;)))
EViews is the bottom of the market educated, but automatons with automatic theories, PR nonsense should be kicked out of economic forums so they don't get clever.
That's very Gaidar-like of you!!!
:))))))
OK, I will not deliberately jump to any fundamental conclusions just yet. I suggest coming back later.
if interested, I think here are worthy goals:
(1) to find a more coherent classification of the separation of the 'mix', i.e. the underlying quote. I.e. a more proper "tailing off"
(2) Understand the properties of the "betta" process, if it is predictable, then it is cool, said - mildly
Could it be that alpha is the speculative component (generally predictable, volumes, bulls/bears sentiment, etc.) and beta is force majeure (unpredictable, a nuclear plant blows up somewhere, tsunami, etc.)?
Where is the proof that there are thick tails in your sample?
Please.
Here is the distribution of minutes for EURCHF for 2009-2010 Alpari:
In the left figure the columns with circles show this distribution, in blue the normal distribution fitted to the original points by the least squares method. You can clearly see Fat Tails out of Gauss on the edges of the distribution. On the right is the same but on a logarithmic scale to make the effect clearer.
Have I proved the presence of TX in this sample?
Here is the Alpha and Omega for this sample:
I think we can say there is a significant negative mutual correlation between the two processes. The separation was done at the 5/2 sigma level.
Same, but for the Eurobucks. The separation was carried out at 3 sigma level:
For the Yen. The separation was at 2 sigma:
It seems that Alpha and Omega should be separated according to the condition of minimum sum of pair correlation of processes with the price series. In this case maximum information capacity is reached for these processes concerning prediction of the price series behavior (this assumption requires a proof).
So, Gray, what's next?
I have laid out the processed data on three different instruments (one, even, at the moment of crisis) and now we need to decide on the targets. You take a look at the results and summarise the intermediate results.
You are welcome.
Have I proved the presence of TX in a given sample?Thank you for your serious answer. Now I have no doubt that you are looking for what is there, just by a different method.
Another question remains: why? There are meaningful doubts:
1) the market has inertia (memory) from any news. By looking at quotes over 2 years you are trying to use such events that have caused inertia over 2 years?
2) The time horizon of the investment. The thick tails you found show that it was dangerous to invest because of the volatility of the variance. What about the future? What predictive power do your findings have?
3) the theoretical profit % of which we are going to capitalise. I looked at the quotes (I do not understand the end or beginning of 2010): the beginning 1.1600, the end 1.1500 - 100 pips. The drawdown to 1.2700=2300 (3300) pips.
Conclusion: all the theorizing outside of the future trading system is meaningless.
For trading, the right side of the quotient is interesting.
And the first question: how many previous bars are needed to predict the next bar with some probability?
Second: what analysis do you need to do and on how many bars to substantiate this probability?
Of course, the pro-rector of Higher School of Economics is not interested in such a down-to-earth view and as a PhD who has not lost a penny on the stock market, he can teach TS and DS processes in earnest to the virgin souls, but what about us? Either we predict or we don't. And 2 year analysis is no help to us in this. The inertia we can ride lasts for days, weeks. Months are out of reach for us because of catastrophic drawdowns. Here are the limits.
God, I'm so sick of these smart people from the Universities! Since Soviet times, they've been clever and clever. The brightest representative is Gaidar.
It's just an interesting approach, the book is purely mathematical, mainly applied engineering problems, and of course the author tested his methods on some indices and quotes, but the processes are very interesting. No selling and no claims like "I'm going to break Forex" and "I'm selling my Expert Advisor". Everything is OK.
Why do I mention it? It's very simple, the classification that I use for this "phenomenon" is not very correct, and there is an interesting method, as I see it.
to IronBird
If the assumption is correct (and omitting conclusions for now), then the market (in the sense of the areas in which the price should hit) is probably "led", although anything is possible. You may be right. I think the speculative component in forex is 90%, at a guess.
By the way... By the way, taking deltas of minutes and counting for them all seems not quite correct - there is a strong dependence of each delta on the time of day (trading session), in fact, I'll tell you a secret - from every minute, not even from an hour ... So. We should first calculate the RMS for each minute of the day for several years, and then normalize all the deltas under study by these values. That would probably be more correct, wouldn't it?
Interesting idea, in our community only a few people investigated these dependencies, it's Yurixx and someone else. I don't even remember anymore. But we need to think about it, maybe it's an additional, important factor.
to Neutron
Итак, Серый, что дальше?
Wait, now I have to make sense of it... I'll get my astrolabe together, check something out and post the result. (Probably not until Monday).