again on the tester - page 8
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
Blaming us for not collecting ticks on a Saturday when there are no quotes is less than incorrect.
I can see you have a history. You stored it in 2007.11.28 and now it's 2008.12.20. It means it is important for you (history) and you keep it a year has passed. But we do not need it. Why - it's noise. But excuse me, if the tick data doesn't make sense, then it doesn't exist in minutes and so on... because all bars are built from ticks (or will you deny that too?).
If you had read the history of the forum in depth you would have seen that this question is raised all the time and I personally answer it all the time. That's where the saved story from 2007 from the terminal is from when I was once again proving the quality of the generator. Just found this piece (no other data on Saturday) on my personal computer at home and re-posted the report.
You need to work further to understand:
- the distribution of ticks by time within a minute bar does not matter
- The shape of the bid trajectory within a bar does matter - we model it quite accurately, in the vast majority of cases absolutely precisely
- very rare divergences of 1-2 pips are absolutely normal and acceptable
I will now check your reports and present my analysis.There is absolutely no information on teak history in the above link.
You must have meant this link.
There is indeed a teak story there.
But how many people need it? ;)
I think that such a rigid emphasis on the indispensable accuracy of history at the tick level is out of thin air - simply because such an "objective history" at such a shallow level doesn't exist. It turns out to be something analogous to the Heisenberg principle. Accurate history is a myth, if only because it is inherently statistical. In reality, there are too many factors that could change it quite significantly.
P.S. In the same thread I saw Renat's suggestion:
That's very interesting. And have the models already been developed accordingly, Renat?
Absolutely right - it's the conclusion "don't lay on the noise within the spread" that you come to in practice.
All brokers have dynamic and adaptive filters (and more than one) which reduce noise at entry. Knowing the principles of price formation, you have to be completely naive to analyse noise plus or minus a pip. I've been repeating for years "don't fall for the deceptive ease of piping in noise" (no adjustments - I take 2 spreads - it's not piping! What is piping? I'm a strategist, not a piper! etc.).
We did not add requotes - we will probably do it in MT5.
You must have meant this link.
There really is a teak story there.
But how many people need it? ;)
Yes, there is from April 2007. But that's disastrously little for tests and few people can apply it.
And we have a really free, clickable minute history from 1999. And our tester does an excellent job of modelling the development of the bars on this one-minute history.
Regarding the comparisons of real and tester quotes you posted.
And made a comparison.
1. I checked if the tester works in the same way as you do. Yes, it is exactly the same. Generated 267 ticks. And it's a perfect match. Here's the figure.
Here is the formula.
It checks if price and time match. If yes, then 1. And they are added together. There are 267 coincidences, i.e. all ticks coincided and this can be seen in the figure too.
0 coincidences, i.e. the tester created by you has never reproduced exactly one tick !!! And it didn't even match the number of ticks in real they are 333, generated 267 !!!! Where is the 20% of ticks ?
What kind of modelling adequacy are you talking about ? And what do you mean by this word ?
Regards Privalov.
Dear Privalov,
I looked through your archive - there is my data and your run. All the data are congruent, except that you are making blatant conclusions based on a mindless comparison of ticks.
Have you forgotten about normalisation and time alignment in minutes? And I have not forgotten and have attached a clear report in Excel with a graph. If there are any complaints, justify them on any of my lines.
My graph with coincidences is the real proof, and your Matkad file (which cannot be opened - did you think about people who will check it?) with the conclusion "0% coincidence" is complete nonsense.
You can play maths, but you can't be so embarrassing with practical conclusions.
That's how it happens every time - you get people out to do real work with facts (tables, samples, graphs, excel) and immediately there is a merging of theorists in a society of practitioners. No offence.
Victor, a couple of words in defence of the modelling algorithm: .
Alexander, even you don't want to understand me...
I never expected that my innocent remark that "ticks in the tester should not be trusted" would bring down an avalanche of such fierce discussion.
What was meant was my personal lamer point of view "I don't understand and don't get involved", on the basis of which the decision to work on bars was made.
I don't need tick history, the quality of tick simulation is of no practical importance to me, and I'm very sorry for having given
>> reason for mutual accusations. I'll be more careful in the future.
P.S. Look into your personal communication
I can't write. A message pops up with a very large text. Renat, please see the attached file, I sincerely want to help.
I can't write. A message pops up with a very large text. Renat, please see the attached file, I sincerely want to help.
Dismissed - none of my lines in the modelling analysis have been refuted.
I don't think I'm interested in wasting my time and repeating practical conclusions for the tenth time to yet another novice theoretical trader. Waste your time, please.
You should reach everything you can based on your own practical experience. >> Good luck.
Take off the rose-coloured glasses.
First line Unix Time Time matches and price Close matches. So, the tick is equal to the real tick (1), second line - price is equal, time is not (0), third line - neither time nor price is equal (0). When comparing the real with the tester there is no 1, something does not coincide + the total number of ticks also - the model has 267, while the real time is 333. All this is easy to check, not necessarily matcad.
Nice maths, you had 333 ticks, the tester gave you 267. That's good. That's two identical numbers. What's all the fuss about?
Personally, I don't give a damn how you simulate. I don't care if you put a white noise generator and integrate it.
A few pages ago stringo asked me to prove that simulated ticks are worse than real ones. But it was not me who asked, but I decided to help and show the difference. I did it as much as I could. You may accept this information and try to correct it or leave it as it is. I was just trying to show that if you made your storage history as
Unix Time - GBPUSD - Bid - Ask
would be better, no, so no. You wear rose-coloured glasses. But don't give them to those who know and understand what adequate modeling is. They will not work.
Take off the rose-coloured glasses.
First line Unix Time Time matches and price Close matches. So, the tick is equal to the real tick (1), second line - price is equal, time is not (0), third line - neither time nor price is equal (0). When comparing the real with the tester there is no 1, something does not coincide + the total number of ticks also - the model has 267, while the real time is 333. All this is easy to check, not necessarily matcad.
Nice maths, you had 333 ticks, the tester gave you 267. That's good. That's two identical numbers. What's all the fuss about?
Personally, I don't give a damn how you simulate. I don't care if you put a white noise generator and integrate it.
A few pages ago stringo asked me to prove that simulated ticks are worse than real ones. But it was not me who asked, but I decided to help and show the difference. I did it as much as I could. You may accept this information and try to correct it or leave it as it is. I was just trying to show that if you made your storage history as
Unix Time - GBPUSD - Bid - Ask
would be better, no, so no. You wear rose-coloured glasses. But don't give them to those who know and understand what adequate modeling is. They won't work.
Not allowed to ask intimate questions about your TS, but at least give me a hint - at what sane TS such accuracy in ticks is needed, not to mention the fact that historical ticks would be different in different dts,
and the difference may be greater than with modelled ones.