Why is the normal distribution not normal? - page 24
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
Honestly, a lot of nonsense from educated people. The apparatus was given, not taught how to apply it.
What educated people say seems like nonsense to you for the simple reason that the filter you have in your head has too narrow a bandwidth. It's a fixable thing, work on yourself. Then there will be less arrogance too.
When timeframes are considered, it's a "cling" to time.
It's not personalities, I've tried and spent several pages to prove to forum users (without names), who know MathCad, their gross errors in reasoning in the same math package. I was faced with stubbornness which I managed to overcome only with numerous posts, substantiated with MathCad results. I don't want to waste any more time on this.
Certain results seem more meaningful to you when you cite screenshots from mate. packages and use mate. terminology.
They tell you the case and you call it nonsense. You should at least have some doubts.
And I do it because I want to see sensible ideas in market analysis. Which may be, if you go the right way.
Now, smartass, quote me when I said "I'm calling bullshit". You're the one who was responding to my post. And if you can't, apologize for the nonsense you said.
One more thing. You, alas, failed to understand my personal attitude to the use of time. Although I've made myself quite clear. So, for the especially gifted, I personally gave up using time about 4 years ago. But, mind you, I have not started to poke others that they are all fools. On the contrary, I often find sensible ideas on this forum. Like what Neutron and Avals tried to show you, for example. If you think that no one around you understands anything, it is your problem. A big one.
What the educated say is nonsense to you for the simple reason that the filter you have in your head has too narrow a bandwidth. It's a fixable problem, so work on yourself. Then there will be less arrogance.
Now, smartass, cite my post where I "call bullshit". You're the one who was responding to my post. And if you can't, then apologize for the nonsense you said.
One more thing. You, alas, failed to understand my personal attitude to the use of time. Although I've made myself quite clear. So, for the especially gifted, I personally gave up using time about 4 years ago. But, mind you, I have not started to poke others that they are all fools. On the contrary, I often find sensible ideas on this forum. Like what Neutron and Avals tried to show you, for example. If you think that no one around you understands anything, it is your problem. A big one.
If you measure pussy (profits absolute and relative), 99% of the time I have more... Except I don't judge a person and their knowledge by the size of their pussy.
Specifically on the case:
Yurixx 04.12.2009 22:52
And no one is clinging to time here, including Neutron. You'd better, darling, strain a bit and figure out what he's writing.
Neutron 04.12.2009 11:04
Getch wrote(a) >>Good point! After all, market time is a measure of change in fin. volume. I don't understand, really, what is a(n) in your reasoning?
It's opening price which is formed from the minutes. For analysis I needed to show the dependence of volatility of an instrument on the known parameter - Time Frame. Having a number of minutes, I can generate any TF. For example, I can get TF10 by considering opening prices of one-minute bars in 10-bar steps. Thus, it turns out that the difference between opening prices of bars in TF10 is the difference of bars' prices in TF1 - a(i*10)-a((i+1)*10), where i takes values 1, 2...
Lots of bukafs.
... These questions have long been chewed up and chewed over. And no one is clinging to time here, including Neutron.
While they still cling, here is a histogram with time lag of one day on M15 (by the way, distribution of any sample is very similar to each other and differs only in absolute values) one column - one sample MO.
The red line is the RMS taken from the corresponding MOs within samples.
In my graph, the horizontal axis is TF derived from minutiae using the following algorithm: TF1 a(n)-a(n+1), TF2 a(n)-a(n+2),...,TFk a(n)-a(n+k). So, colleague, we are doing exactly what you advise.
By definition, the pairwise correlation coefficient between adjacent readings in RPM of a price series, is the probability of correctly predicting the colour of the next candle, or in other words, it's the MTS efficiency. To convert percentages to points we need to know the volatility of the instrument in the selected timeframe. By multiplying the efficiency by the volatility, we obtain an estimate for the profitability of the TS as the average value of points per transaction. This must be compared with the brokerage company commission (spread). If the profitability on any TF exceeds the spread, profitable trading is possible.
I like this branch. I don't even know why. :)
Neutron, it seems you have built a simplified analogue of ACF and proved that noise exceeds a useful signal on a minute.
Have you tried looking at the larger time scales? It's possible that there will be a slightly different result.
Honestly, a lot of nonsense from educated people. The apparatus was given, not taught how to use it. No one here is engaging in self-aggrandisement.
When timeframes are considered, it's a "cling" to time.
I've tried it without persons and for several pages I was proving to forum users (without names), who know MathCad, their blunders in mathematical equation. I was faced with stubbornness which I managed to overcome only with numerous posts, substantiated with MathCad results. I don't want to waste any more time on this.
Certain results seem more meaningful to you when you cite screenshots from mate. packages and use mate. terminology.
They tell you the case and you call it nonsense. You should at least doubt it a little.
And I do it because I want to see sensible ideas in market analysis. Which may be, if you go the right way.
Gettingch time is just another row of data. You may or may not use it, it all depends on the trading idea. In some cases it is necessary and its use is logical, in others it is useless and can render a system meaningless. For example, most of the common ways to use indicators with a fixed calculation period. Mindless use of time should be avoided, as well as any other data.
Neutron, you seem to have built a simplified ACF analogue and proved that noise exceeds the useful signal at minutes.
Have you tried looking at larger time scales? It is possible that there will be a slightly different result.
Tried it.
If I find a long enough series of minutiae (several years) I will show the result. In a nutshell, it is as follows: As the TF increases, the pairwise correlation coefficients between adjacent samples of RPR fall almost exponentially fast, and the product of CC on the instrument volatility steadily falls as the TF increases, despite the root growth of volatility (the exponent decreases faster than any power function). That is why after the TF 100-500 there is nothing to catch, before the TF100m there is nothing to catch. As a rule, the optimum is at 4h, but DTs strictly cover the maximum return in this scenario at all pairs.
Here is the data for the DOI index:
The M1 series from 2006 to the present is on the left in the figure. By the way, the failure of the World Crisis is clearly visible. The second figure shows the distribution of increments... How do they quote it! On the right is the profitability of the "ideal" TS not using time in the analysis (only price changes).
>> I will keep looking.
P.S. Especially for getch. I never use time when analysing price series. I work in terms of "BP counts" and do not analyze time intervals between counts at all. The fact that I operate with the term TF in this post is purely demonstrative and is intended to reflect the dependence of one of BP parameters on standard time without trying to apply it to TS in practice.
Am I the last fool around here who still uses time sometimes?!
Am I the last fool around here who still uses time sometimes?!
))) I suppose it's about different sample lengths and, in general, about selecting them for analysis to build discretes (like new bars), on which all these mathematical perversions already make sense.