I'm not sure this is possible??????

[paulm]

Hi

I'd like to ....

1. calculate 4 ema's (tf = timeframe)

o = ema(period1,open(inv(tf));
h = ema(period1,high(inv(tf));
l = ema(period1,low(inv(tf));
c = ema(period1,close(inv(tf));

2. then calculate smoothed heiken ashi values based on values returned from step 1.

ha_close = (o+h+l+c)/4;
ha_open = (ha_open(previous bar) + ha_close(previous bar))/2
ha_high = Math.max(h,ha_open);
ha_low = Math.min(l,ha_open);

3. Smooth again

calculate 4 new smoothed values using the results calculated in step 2. using the wma function.

ha2_close = wma(period2,ha_close values from step 2.)
ha2_open = wma(period2,ha_close values from step 2.)
ha2_high = wma(period2,ha_close values from step 2.)
ha2_low = wma(period2,ha_close values from step 2.)

then if close > open return trend=UP
if close < open return trend=DOWN

And I want to do this on 8 timeframes in the same study ... so i end up with 8 trend variables set to UP/DOWN for 8 different timeframes

Is this possible - I cannot get my head around how to do it!!!!

Thanks again your help is always much appreciated!

Paul

[smeyer55]

Paul,
What you are asking is possible if you convert your variables to series.

Right now your ha_x variables are single values instead of a series. You can convert them to a series using an efsInternal function and putting your ha_x calculations inside an efsInternal function.

Then you could use that series as the source for the wma.

The knowledgebase has articles about efsInternal that explains how it works.

Steve