Building a trading system using digital low-pass filters - page 31
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
Hello to you too, Sergei, hello.
No, no, God forbid, I did not publish in The Currency Speculator. I'm talking about questions about statistics and fat tails from diakin. Recognised a familiar sentence from his own article on sandwiches, he quoted it...
diakin, sorry to refer to you in the third person.
I see.
What do you think of the article posted? The authors unfortunately didn't suggest a way to implement the discussed ideal LPF, I wish they could come up with a function to minimise it...
Yes, yes there is such a thing :-)
Was the mad scientist able to defeat the Scarecrow, or is history silent?
As far as I remember, we have jointly implemented a quick averaging using the WPI code given in the article, and posted the graphs - the code there seemed to be not complicated.
The curves turned out to be, to put it mildly, far from smooth, but resembled some sort of averaging with a very small window. He started wagging his tail and started referring to the fact that the algorithm on WPI wasn't made by him, as he himself doesn't know anything about coding, but by a grad student (I think his almost literal words were something like "what can I take from her, a grad student").
Then we took apart his algorithm described and illustrated by schemes in the article. It turned out that despite the algorithmic complexity of the circuit itself, it is in fact the simplest linear filter with constant coefficients. And an ordinary linear filter does not exactly achieve the same smoothness and the same fast response time as Zhmurikov's one. He began to tell us that we have implemented this scheme incorrectly, showing at the same time a very fuzzy "correct" picture from the same article with low resolution (the article is from "Sun" in pdf format), where you can't see anything in particular. In short, there was no common language here either.
Finally we got back to the code, which we had made in the footsteps of his WPI code, and matched up an EMA or WMA with a period, which almost perfectly matched his revolutionary filter. That was the end of the discussion, and Zhmurik was never put to shame.
P.S. I remembered why Zhmurik is not approximated by any linear filter with constant coefficients. Zhmurik, according to its author himself, is sort of an adaptive filter... At the same time Smirnov still waved the flag and shouted that "give me a few dozens of consecutive samples of this filter, and I'll easily re-engineer it".
P.P.S. LeoV, it seems you tried to contact Smirnov. Did you get any results?
PATTALOM ))
It's called an artist can offend anyone, but an artist can offend everyone at once.
Kashpirovsky was misunderstood, he went to the USA and Smirnov to the same place.)
jartmailru tell me what kind of a frequency response meter can be made for filter tuning
Dear cabluk,
.
what do you mean by the word GIRL? :-) Do you - GOOD? (the place of the typo is your choice).
and - besides - where is the meter itself?
.
I'd have to figure out how to compare frequencies from different timeframes... I have not got it...
If it's a good idea - to somehow build smart digital filters - for example,
match the filtered signal to the input sinusoids, equalise the sampling frequencies
on the data from the different timeframes - and bump the signal itself...
.
and I would - but I'm lazy :-(
Dear cabluk,
.
what do you mean by the word GIRL? :-) Do you - GOOD? (the place of the typo is your choice).
and - besides - where is the meter itself?
.
I'd have to figure out how to compare frequencies from different timeframes... I have not got it...
If it's a good idea - to somehow build smart digital filters - for example,
match the filtered signal to the input sinusoids, equalise the sampling frequencies
on the data from the different timeframes - and bump the signal itself...
.
and i would do it, but i'm too lazy to do it :-(
an Ahh meter is easy to build... feed sine waves of different lengths in the wavelength range of interest and measure the amplitudes
ZS I wanted to write marvel but at the last moment my brain snapped ))