NinjaTrader Support Forum Standard Deviations with different Average

 General Development General NinjaScript development questions.

06-12-2018, 01:10 AM   #1
Junior Member

Join Date: Oct 2014
Posts: 29
Thanks: 9
Thanked 1 time in 1 post
Standard Deviations with different Average

Hello,

I have developed custom Standard Deviation based on different average like the:

1. Geometrical average

2. Harmonic average

As shown in the attached pic, i would like to understand why, for the same Period:

1. The Regular Stdev is almost equal to the Harmonic StdDev
2. The Geom Stdev is almost equal to the Quadratic StdDev

It's more like a maths problem but i would be interested to have a feedback...

Thanks
Attached Images
 StandardDeviations.JPG (109.9 KB, 14 views)
Last edited by cbadr; 06-12-2018 at 01:12 AM.

06-12-2018, 04:31 AM   #2
koganam
Senior Member

Join Date: Feb 2008
Location: omegasigmaindicators.com
Posts: 7,235
Thanks: 56
Thanked 3,012 times in 2,437 posts

Quote:
 Originally Posted by cbadr Hello, I have developed custom Standard Deviation based on different average like the: 1. Geometrical average 2. Harmonic average 3. Quadratic average As shown in the attached pic, i would like to understand why, for the same Period: 1. The Regular Stdev is almost equal to the Harmonic StdDev 2. The Geom Stdev is almost equal to the Quadratic StdDev It's more like a maths problem but i would be interested to have a feedback... Thanks
That would be impossible to tell with certainty without seeing the code. Offhand, it would most likely be because your code refactors to identical code between the matching methods.

 06-12-2018, 05:28 AM #3 cbadr Junior Member   Join Date: Oct 2014 Posts: 29 Thanks: 9 Thanked 1 time in 1 post Thanks for the answer. Here the code i used for the Geometrical Average: But after investigations on google, i realized i couldn't used the same Std calculation once the Mean is calculated. I'll have to update the formula accordingly. Any thoughts on this? Code: ```protected override void OnBarUpdate() { if (CurrentBar < 1) { Value.Set(0); sumSeries.Set(Input[0]); } else { sumSeries.Set(Input[0] * sumSeries[1] * (CurrentBar >= Period ? 1/Input[Period]: 1)); double avg = Math.Pow(sumSeries[0],1/Math.Min(CurrentBar + 1, Period)); double sum = 0; for (int barsBack = Math.Min(CurrentBar, Period - 1); barsBack >= 0; barsBack--) sum += (Input[barsBack] - avg) * (Input[barsBack] - avg); Value.Set(Math.Sqrt(sum / Math.Min(CurrentBar + 1, Period))); } }```
06-12-2018, 09:39 AM   #4
koganam
Senior Member

Join Date: Feb 2008
Location: omegasigmaindicators.com
Posts: 7,235
Thanks: 56
Thanked 3,012 times in 2,437 posts

Quote:
 Originally Posted by cbadr Thanks for the answer. Here the code i used for the Geometrical Average: But after investigations on google, i realized i couldn't used the same Std calculation once the Mean is calculated. I'll have to update the formula accordingly. Any thoughts on this? Code: ```protected override void OnBarUpdate() { if (CurrentBar < 1) { Value.Set(0); sumSeries.Set(Input[0]); } else { sumSeries.Set(Input[0] * sumSeries[1] * (CurrentBar >= Period ? 1/Input[Period]: 1)); double avg = Math.Pow(sumSeries[0],1/Math.Min(CurrentBar + 1, Period)); double sum = 0; for (int barsBack = Math.Min(CurrentBar, Period - 1); barsBack >= 0; barsBack--) sum += (Input[barsBack] - avg) * (Input[barsBack] - avg); Value.Set(Math.Sqrt(sum / Math.Min(CurrentBar + 1, Period))); } }```
Whereas you are using a rather inefficient algorithm, I nonetheless do not quite understand what you are asking. You appear to be calculating a mean, then looping backwards and summing the squares of the deviations from the mean. That is the basic definition of how to determine the variance and hence StdDev. So what is the issue?

 06-12-2018, 10:02 AM #5 cbadr Junior Member   Join Date: Oct 2014 Posts: 29 Thanks: 9 Thanked 1 time in 1 post The issue is about understanding the result of the StdDev from a Geometric VS Harmonic VS Squared Mean. From wikipedia i found that the formula to get the Geometric Std is mention there: https://en.wikipedia.org/wiki/Geomet...dard_deviation Thats a first thing i have to update on the script. Then really i am wondering if the Std formala is the same for the other Mean?
06-12-2018, 08:47 PM   #6
koganam
Senior Member

Join Date: Feb 2008
Location: omegasigmaindicators.com
Posts: 7,235
Thanks: 56
Thanked 3,012 times in 2,437 posts

Quote:
 Originally Posted by cbadr The issue is about understanding the result of the StdDev from a Geometric VS Harmonic VS Squared Mean. From wikipedia i found that the formula to get the Geometric Std is mention there: https://en.wikipedia.org/wiki/Geomet...dard_deviation Thats a first thing i have to update on the script. Then really i am wondering if the Std formala is the same for the other Mean?
For an unbiased estimator, the standard deviation is always the Root Mean Square Deviation, the square root of the variance, regardless of what mean you are using, albeit the mean used is almost always the arithmetic mean, rather than any other.

Why do you think that a geometric mean may be more appropriate for what you are trying to do, if I may ask.

 Tags average, math, standard deviation

 Thread Tools Display Modes Linear Mode

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is Off Forum Rules
 Forum Jump User Control Panel Private Messages Subscriptions Who's Online Search Forums Forums Home News and Announcements     News and Announcements NinjaTrader 8     Platform Technical Support     Add-On Development     Indicator Development     Strategy Development NinjaTrader 7     Platform Technical Support     General Development     Indicator Development     Strategy Development NinjaScript File Sharing     NinjaScript File Sharing Discussion     Version 8 Add-Ons     Version 8 Indicators     Version 8 Strategies     Version 7 Indicators     Version 7 Strategies     Version 7 Miscellaneous NinjaScript Educational Resources     Reference Samples     Tips Suggestions and Feedback     Suggestions And Feedback Historical Beta Archive     Version 8 Beta

 Similar Threads Thread Thread Starter Forum Replies Last Post behnam Indicator Development 14 08-04-2018 06:31 PM pretender General Development 3 10-06-2015 10:15 AM morphtrade Platform Technical Support 0 10-07-2009 09:18 AM amedhussaini Indicator Development 3 08-30-2009 06:52 PM murfinp Indicator Development 2 08-11-2009 03:39 PM

All times are GMT -6. The time now is 11:05 AM.

Futures, foreign currency and options trading contains substantial risk and is not for every investor. An investor could potentially lose all or more than the initial investment. Risk capital is money that can be lost without jeopardizing ones financial security or lifestyle. Only risk capital should be used for trading and only those with sufficient risk capital should consider trading. Past performance is not necessarily indicative of future results. View Full Risk Disclosure.

CFTC Rules 4.41 - Hypothetical or Simulated performance results have certain limitations, unlike an actual performance record, simulated results do not represent actual trading. Also, since the trades have not been executed, the results may have under-or-over compensated for the impact, if any, of certain market factors, such as lack of liquidity. Simulated trading programs in general are also subject to the fact that they are designed with the benefit of hindsight. No representation is being made that any account will or is likely to achieve profit or losses similar to those shown.