Confused - Stochastic Question? | Page 2 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Let's talk about trends I'm confused 16 replies

Stochastic divergence indicator with stochastic cross? 0 replies

a very noob question so confused 4 replies

stochastic histogram question 0 replies

Stochastic question from newbie. 2 replies

Attachments: Confused - Stochastic Question?
Exit Attachments

Confused - Stochastic Question?

  • #21
  • Dec 6, 2021 6:30am Dec 6, 2021 6:30am
Quoting robots4me
{quote} Dude -- what do you think 'Close' is? It is a bar's last tick. 'Open' is a bar's first tick. 'High' is the highest tick within a bar. 'Low' is the lowest tick within a bar. Again -- all there is are ticks. Your broker does not send bar data to your MT4 session, they send tick data. The MT4 software then combines the tick data to form bars or candles. For example, in the case of an H1 candle, then MT4 accumulates ticks for 1 hour and uses them to construct an H1 bar. For the current bar (i.e. Bar[0]) that is still open, the Open price is...
To draw stochastic Mt4/5 use neither first tick nor last tick nor lowest tick. To draw your stochastic oscillator it use bar open, bar close, bar low and bar high.
You can code custom one to use ticks data, but the default one works as I said only. There are no such a setting to use ticks instead of bars.

And I think if topic started don't know how to extrapolate stoch settings to the higher TF it is barely impossible that he is talking and what to know about custom indicator features. So you better use your useless knowledge for your repainting indicators thread
Observer effect
  • #22
  • Edited 7:17am Dec 7, 2021 1:57am | Edited 7:17am
Oh boy... it's war zone out here. Can we please stop going back and forth correcting each others' posts?

As for the original question: there is no direct way to represent the stoch of a higher timeframe using the lower timeframe with scaled settings. This has to do with the calculation of the stoch which uses the highest, lowest and last price during a time interval. You can not mimic the time interval on a lower timeframe. If you scale up the settings of a stoch, it simply takes *more* time intervals, not larger time intervals.

The suggestions by Kafin (changing it in code) and jsf (using a MTF indicator) are best. If you really want to represent the stoch of a higer timeframe, you will see a "staircase" shape. The M15 stoch value changes once every 15 minutes, so when viewed on an M5 chart it will remain the same value for 3 bars before changing.

If you rather have more fluent lines, then you can indeed try to approximate higher timeframes, but the values will not be the same. You mention EMA as an example. You could do the same for certain other indicators too, but sometimes you also need to scale the indicator value itself (for example, ATR scales roughly with the square root of the time interval). But be aware that the values are not the same as the actual higher timeframe indicator.

So in short: best to download a MTF variant of an indicator and use that!
  • #23
  • Dec 7, 2021 3:20am Dec 7, 2021 3:20am
Quoting yoriz
So in short: best to download a MTF variant of an indicator and use that!
No. Best is just multiply on the difference between timeframes if you want to reproduce higher to lower or divide if you want to reproduce lower to higher. Using MTF indicator is worth, because it use higher TF data, and it will not show current data (or incorrectly extrapolated) until the referring candle will be closed.
Observer effect
  • #24
  • Edited 5:02am Dec 7, 2021 4:37am | Edited 5:02am
Quoting ryuryu
Best is just multiply on the difference between timeframes
The way I understood the question from OP ("that reflect the [...] higher timeframes"), is that OP wants to see the values of the higher timeframe indicator on a lower timeframe chart. The only way to achieve that for a Stoch indicator is to use an MTF variant. There is no way to do that using the regular Stoch calculation on the lower timeframe bar data.

What you propose, ryuryu, gives something else. Perhaps having a slow Stoch on a lower timeframe can also be useful, but it is not equal to the regular Stoch indicator on the higher timeframe. These give different values.
  • #25
  • Edited 7:00am Dec 7, 2021 4:47am | Edited 7:00am
Quoting yoriz
{quote} The way I understood the question from OP ("that reflect the [...] higher timeframes"), is that OP wants to see the values of the higher timeframe indicator on a lower timeframe chart. The only way to achieve that for a Stoch indicator is to use an MTF variant. There is no way to do that using the regular Stoch calculation on the lower timeframe bar data. What you propose, ryuryu, gives something else. Perhaps having a slow Stoch on a lower timeframe can also be useful, but it has no relation to the regular Stoch indicator on the higher...
No. To reflect higher TF Stoch on lower TF you just need to get the difference, then multiply the value. That's all.
I don't propose anything. It is just basic math.

Example: 14 stoch (rsi, whatever) on 1h TF is 840 on 1m, 56 on 15m, 28 on 30m, 84 on 5m.

Problem of MTF is that it will use data from higher TF and only once the candle will be closed. So it you are trading at 13:15 you will have data only from 13:00 candle. If you trade at 13:59 - the same. So this will be not accurate and useless. Using simple math is much better, because it will show you the data with 1 min lag maximum (since 1m is the lower TF available). Also keep in mind that lower TF data is more accurate because candles are smaller.

Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 27 KB

This blue line is 200 EMA from 1h TF on 1m tf.
You will never see last hour data using MTF. But as you see there are such a data. Because it is just EMA 12000. There is 1m lag only. Not 1-59 min lag.
Observer effect
  • #26
  • Edited 7:25am Dec 7, 2021 5:37am | Edited 7:25am
Quoting ryuryu
To reflect higher TF Stoch on lower TF you just need to get the difference, then multiply the value. That's all.
The Stoch(14 H1) is not the same as the Stoch(840 M1). The values are different. Even at the close of the H1 bar. You can check that yourself at the whole hour having an M1 chart and H1 chart open side by side. Try it!

Quoting ryuryu
This blue line is 100 EMA from 1h TF on 1m tf.
Also here: the EMA(100 H1) is not the same as the EMA(6000 M1). Compare the values at the whole hour on two side by side charts to verify!

Quoting ryuryu
To reflect higher TF Stoch on lower TF you just need to get the difference, then multiply the value. That's all.
I don't propose anything. It is just basic math.
False, basic math tells you it is a totally different calculation. Why do you think it would give the same values?

As a simple example, consider the EMA(2) versus EMA(4) calculations:

Inserted Code
EMA(2) = previous EMA(2) * 0.333 + Close * 0.667
EMA(4) = previous EMA(4) * 0.600 + Close * 0.400

Put that in a spreadsheet and create some artificial M15 and M30 bars, and do the calculations above:
Attached Image

As expected, doing a different calculation gives different values. They might be somewhat similar, but they are not the same.

Sorry, ryuryu, your idea that you can simply scale a lower timeframe indicator to get the exact same values as the higher timeframe indicator is incorrect. That is the reason MTF indicators exist. Obviously, if using indicators on a low timeframe with large/slow settings are useful to you, then go ahead! Just pointing out that they are different things and probably not what the OP was asking for.
  • #27
  • Dec 7, 2021 6:58am Dec 7, 2021 6:58am
Quoting yoriz
{quote} The Stoch(14 H1) is not the same as the Stoch(840 M1). The values are different. Even at the close of the H1 bar. You can check that yourself at the whole hour having an M1 chart and H1 chart open side by side. Try it! {quote} Also here: the EMA(100 H1) is not the same as the EMA(6000 M1). Compare the values at the whole hour on two side by side charts to verify! {quote} False. Basic math tells you it is a totally different calculation, thus giving different values. ;-) Again, probably having slow Stoch and EMA indicators on a lower timeframe...
I'm not interested in teaching you basic math. If you think that 2+2 is not 4 this is your problem. If you can't check and understand stochastic formula since 2006 this is your problem again.
I'm tired of people who post and post forex fairy tales day by day. Topic starter asked how to reproduce higher TF stoch to the lower tf, I answered and provided the formula. So my mission complete. Now you can use magic, feelings, believes, luck and whatever you want. 2+2=4 and global math rules don't care about your nonsense.
Observer effect
  • #28
  • Dec 7, 2021 7:53am Dec 7, 2021 7:53am
Quoting ryuryu
global math rules don't care about your nonsense.
Ok, man. Whatever. Since you mentioned yourself it was just basic math, I took 5 minutes of my time to create a simple spreadsheet for you to show some example numbers. Often people learn best by example. I was not trying to patronize you, only trying to open your mind.

Nice personality, ryuryu. It would have been more mature to just admit you made an error and learned something new.

Anyway, @OP: as suggested by several people in this thread: probably best to use MTF indicators.
  • #29
  • Dec 7, 2021 7:58am Dec 7, 2021 7:58am
Quoting ryuryu
{quote} I'm not interested in teaching you basic math. If you think that 2+2 is not 4 this is your problem. If you can't check and understand stochastic formula since 2006 this is your problem again. I'm tired of people who post and post forex fairy tales day by day. Topic starter asked how to reproduce higher TF stoch to the lower tf, I answered and provided the formula. So my mission complete. Now you can use magic, feelings, believes, luck and whatever you want. 2+2=4 and global math rules don't care about your nonsense.
He was right about his math ..

Why are you guys making it this complicated
Was using stochastic since 2016 following simple rules why complicate simple things
For me my best setup is Daily on a 4H timeframe
Wealthy
  • #30
  • Dec 7, 2021 8:42am Dec 7, 2021 8:42am
Quoting yoriz
Oh boy... it's war zone out here. Can we please stop going back and forth correcting each others' posts? As for the original question: there is no direct way to represent the stoch of a higher timeframe using the lower timeframe with scaled settings. This has to do with the calculation of the stoch which uses the highest, lowest and last price during a time interval. You can not mimic the time interval on a lower timeframe. If you scale up the settings of a stoch, it simply takes *more* time intervals, not larger time intervals. The suggestions...
@yoriz -- set aside Stochastic and instead consider Moving Average. Stochastic has 3 settings (KPeriod, DPeriod, Slowing), so scaling just the KPeriod won't work. But in the case of a Moving Average then scaling certainly does work.

The "stair case" shape you mention has ***nothing*** to do with trading -- it's due to a limitation in the way MT4 charts are drawn. The price is the price -- regardless if you are looking at an M1 chart or a D1 chart. @yoriz -- as much as it pains me to say it, in this case @ryuryu is right. @yoriz -- you confuse MT4's drawing of historical OHLC data versus trading price data in real-time.
  • #31
  • Dec 7, 2021 8:59am Dec 7, 2021 8:59am
The best way to get your own idea, compare two graphs with two differents timeframe .

For example :5 Minutes and 1 Hour.

Attached Image (click to enlarge)
Click to Enlarge Name: stochastic.PNG Size: 54 KB
KISS&V :Keep it simple safe & visual
  • #32
  • Dec 7, 2021 9:06am Dec 7, 2021 9:06am
Quoting robots4me
But in the case of a Moving Average then scaling certainly does work.
My example spreadsheet suggests otherwise. The values might be similar, so it could be a good approximation.

Quoting robots4me
The "stair case" shape you mention has ***nothing*** to do with trading -- it's due to a limitation in the way MT4 charts are drawn.
What I meant is that the lower timeframe should show the value of the higher timeframe. So it takes three M5 bars to one M15 bar. Strictly speaking, the MTF indicator will show the same value three times. Of course you can do some interpolation, or display the incomplete partial current bar.
  • #33
  • Dec 7, 2021 9:45am Dec 7, 2021 9:45am
Quoting yoriz
{quote} My example spreadsheet suggests otherwise. The values might be similar, so it could be a good approximation. {quote} What I meant is that the lower timeframe should show the value of the higher timeframe. So it takes three M5 bars to one M15 bar. Strictly speaking, the MTF indicator will show the same value three times. Of course you can do some interpolation, or display the incomplete partial current bar.
The point I am trying to make -- and obviously failing at -- is that looking at a chart is not the same thing as trading. A chart displays historical, OHLC data. When you trade (in real time) you are trading ticks. There is a big, big difference.

All there is are ticks. MT4 (and other platforms) simply aggregate ticks into candles and bars for our viewing pleasure. If you are just into colorful Christmas tree charts that's fine -- attach all the indis you want. But if you want to be a trader, then understand the relationship between ticks (the real data) and bars (the abstract constructions).
  • #34
  • Dec 7, 2021 9:52am Dec 7, 2021 9:52am
Quoting robots4me
All there is are ticks. MT4 (and other platforms) simply aggregate ticks into candles and bars for our viewing pleasure.
Yes, absolutely. In fact, everything above H1 is sensitive to the timezone your broker is using! Crazy that my D1 bars can be different from your D1 bars!
  • #35
  • Dec 7, 2021 9:53am Dec 7, 2021 9:53am
Quoting Sohocool
The best way to get your own idea, compare two graphs with two differents timeframe.
Thanks, that clearly shows the difference.
  • #36
  • Dec 7, 2021 10:09am Dec 7, 2021 10:09am
Quoting Sohocool
The best way to get your own idea, compare two graphs with two differents timeframe . For example :5 Minutes and 1 Hour. {image}
As mentioned a few times already -- Stochastics doesn't lend itself to rescaling because it has 3 settings that use a sliding window, so fiddling with one setting is not enough. Plus, the D Period and Slowing parameters are not moving averages.

Below is a comparision of MovingAverage on M15 and M5 charts. The period of the MovingAverage on the M15 chart is 20, whereas the period of the MovingAverage on the M5 chart is 60. An M15 bar contains 3 M5 bars and, hence, M15 contains 3x more ticks. For equivalent algebraic results the M5 chart needs a 3x larger period.

The visual comparison isn't perfect because I couldn't perfectly align their X-axes, but you get the picture, right?

Attached Image (click to enlarge)
Click to Enlarge Name: Rescaling-MovingAverage.png Size: 76 KB



MTF is completely unrelated -- whoever thinks that MTF is a solution please stop spreading misinformation. MTF is simply a drawing solution -- it does not rescale. A MovingAverage on M15 with period of 20 will trade completely differently than a MovingAverage on M5 with a period of 20. Just because you can draw them both on the same chart doesn't mean they will trade the same.
  • #37
  • Dec 7, 2021 10:21am Dec 7, 2021 10:21am
For those MTF believers, consider the chart below. I used a MTF Moving Average indicator two times on an M15 chart -- one set of blue/red lines is Moving Average (20). Another set of blue/red lines is an M5 Moving Average (20) redrawn on the M15 chart. They both use the same period -- 20. Do they look the same to you? Do you think they would trade the same? If so, well, I wish you luck -- because you are going to need a lot of it.

Attached Image (click to enlarge)
Click to Enlarge Name: MTF-vs-Rescaling.png Size: 70 KB
  • #38
  • Dec 7, 2021 11:08am Dec 7, 2021 11:08am
Quoting djahidait
{quote} He was right about his math .. Why are you guys making it this complicated Was using stochastic since 2016 following simple rules why complicate simple things For me my best setup is Daily on a 4H timeframe
Could you tell us your settings?
Thanks
"Great Men are not born great, They Grow Great."
  • #39
  • Dec 7, 2021 11:09am Dec 7, 2021 11:09am
ryuryu and robots4me are right, its basic math.


Here is a chart with aligned period separators.
The H1 chart uses stoch 5, 3, 3
Ma 25, 13 and 8
Ma high 2, Ma low 2

The M15 chart uses stoch 20, 3, 3
Ma 100, 50, 30
Ma high 8, Ma low 8

The movement is faster on the M15 than on the H1, but they line up more or less because of the scaling of the candles.
The most important thing is the to line up the period separation with each other if you want to compare the 2 charts.

Attached Image (click to enlarge)
Click to Enlarge Name: screenshot.png Size: 23 KB
  • #40
  • Dec 7, 2021 11:59am Dec 7, 2021 11:59am
Quoting OneandOnly66
they line up more or less because of the scaling of the candles. The most important thing is the to line up the period separation with each other if you want to compare the 2 charts.
Yes, please do compare the values in both charts. Use the "Data Window" (Ctrl-D) tool and read out the values in both charts. You will see that the values might be similar, but not identical. The calculations are different, hence the output is different.

Quoting robots4me
I used a MTF Moving Average indicator two times on an M15 chart -- [...] They both use the same period -- 20.
In my understanding that is not how you use an MTF indicator. In the MTF indicator you specify the timeframe and the period (expressed in that other timeframe) that you want to show. If you use two different timeframes with the same period, then obviously you get different things.

Also, you are trying to display a lower timeframe (M5) onto a higher timeframe (M15). Usually you would want to go the other way.
0 traders viewing now