# From theory to practice - page 1711

Maxim Dmitrievsky:

better a bigger one. What TF is this series associated with?

Then we will have to wait a little longer...

This series is obtained from tick series with some time thinning. It obtains the series with the variance = const in any sample of data. I am currently working with such series on the real. You may see the result in my reports. But... Not many deals... Boring. You need an unbridled grail.

Alexander_K:

Then it will have to wait some more...

This series is obtained from ticks with some time thinning. You get a series with variance = practically const for any sample of data. I am currently working with such series on the real. You may see the result in my reports. But... Not many deals... Boring. I need an unrestrained grail.

Can you translate it into TF? It would not require effort to translate it into ticks, minutes and hours ))

I can test it for different timeframes and see, yeah (my trades will be done on some timeframe, it makes no sense to use ticks)

into the one-minute TF, then yourself

Alexander_K:. Few deals... Boring. Need an unbridled grail.

Multiply the lot, it would be a lot more fun. )

Maxim Dmitrievsky:

Can you translate it into a timeframe? So I don't have to bother with ticks, don't translate it into minutes and hours ))

I can test it for different timeframes and see, yeah (the trades will be on some timeframe, it makes no sense on ticks)

in a minute timeframe, then do it yourself

1 2 3 4 5 6 7 8 9 10
1 na 3 na na 6 na 8 na na

Vizard_:

1 2 3 4 5 6 7 8 9 10
1 na 3 na 6 na 8 na

cha cha one two three

there are converted tics, how do I know what matches what?

Maxim Dmitrievsky:

Describe the data format (row/column) you want. I'll try to do it.

If you're interested, of course.

Alexander_K:

Describe the data format (row/column) you want. I'll try to do it.

Well, if you're interested, of course.

Any, by column you can. The initial and the transformed ones should be synchronized.

The initial one for trading, the transformed one as an indicator for the NS

Alexander_K:

I would now be interested in the following experiment.

1. you take a BP on OPEN M1 or CLOSE M1 of some currency pair agreed upon between us. We agree on which time period of this data we will look at the results.

2. We look at the test results of your best neural network model.

3. For the same pair, for the same time period, look at the results on tick data.

4. Same for BP, which I will give you.

5. Compare, draw conclusions.

If you're interested, let me know.

OPEN is exactly one tick behind CLOSE !

OPEN of a current tick === CLOSE of a previous tick, theoretically ... ;)

Sash, don't work with OPEN, that's for sure !

Maxim Dmitrievsky:

any, a column is fine. Preferably, the original series and the transformed one should be synchronized

the original one for trade, the transformed one as an indicator for the NS

А! Do you think that my transformed series is the same as the one we made with Lambert?

No! Just a cleverly thinned conventional tick price series. Figuratively, it was 100,000 ticks and now it's 10,000 ticks.

My series corresponds approximately to the S12 TF. Its exceptional feature is the constancy of dispersion. I.e., I manage to obtain a quasi-stationary series from the nonstationary tick BP by "sifting".

Alexander_K:

А! Do you think my converted range is the same as the one Lambert and I fiddled with?

No! Just a cleverly thinned conventional tick price range. Figuratively, it was 100,000 ticks and now it's 10,000 ticks.

My series corresponds approximately to the S12 TF. Its exceptional feature is the constancy of dispersion. I.e., from a nonstationary tick VR I obtain a quasi-stationary series by "sifting".

Well, you may not trade on it anyway, may I open positions on the initial one?