Same EMA but from multiple time frames

[sandman]

Patrick. No more questions on this topic. Thanks for your help and patience.

sandman

[sandman]

.... (2) Tweaking the indicator

I understand that you try to build an EMA for a higher timeframe. For both EMA and SMA there is a simple solution: multiply the period of the EMA with the timeframe factor.

-> a 13-period EMA on a 3 min chart is near-identical to a 39-period EMA on a 1-minute chart
-> a 13-period EMA on a 10 min chart is near identical to a 130-period EMA on a 1 min chart

Of course there are minor differences created by session templates and exponential smoothing, but they can be neglected.

Harry. I had already tried something similar and had a 35-period EMA on a 1Min chart for the 13-period EMA of a 3Min DataSeries. 39 based on the formula you provided is better.

I put that into my various charts and it seems it fulfills its intended purpose VERY nicely.

Thank you for having taken the time to reply in such detail.

sandman

[Harry]

Harry. I had already tried something similar and had a 35-period EMA on a 1Min chart for the 13-period EMA of a 3Min DataSeries. 39 based on the formula you provided is better.

I put that into my various charts and it seems it fulfills its intended purpose VERY nicely.

Thank you for having taken the time to reply in such detail.

sandman

Your thanks should go to Jack Hutson, see his contribution in Stocks & Commodities, Filter Price Data: Moving averages vs. Exponential Moving Averages, published in March 1984.

Welles Wilder used an exponential moving average with a smoothing constant of 1/N, whereas the modification suggested by Jack Hutson lead to an exponential moving average with a smoothing constant of 2/(N+1), where N is the period or comparable length of a linear moving average.

The simple approximation of a higher timeframe EMA is only possible because of the new definition of the smoothing constant.