2 replies to this topic

[spielchekr]

...the output of this formula series spikes away from the zero line. It almost never fails. The formula assigns divergence points to each of 200 moving averages... either a positive 1 point, a negative 1 point or a zero to each of 200 moving averages.

A positive point gets assigned when today's close divided by X moving average, is greater than yesterday's close divided by X moving average, AND the close is less than yesterday's close (in other words, a positive divergence). For a negative point for the same moving average time, the opposite must be true. And if neither is true, zero is assigned. It looks like this:

if( close/mov(close,1) > delay(close/mov(close,1),1) && close<delay(close,1), 1,0)
+
if( close/mov(close,1) < delay(close/mov(close,1),1) && close>delay(close,1), -1,0)
+
if( close/mov(close,2) > delay(close/mov(close,2),1) && close<delay(close,1), 1,0)
+
if( close/mov(close,2) < delay(close/mov(close,2),1) && close>delay(close,1), -1,0)
+
if( close/mov(close,3) > delay(close/mov(close,3),1) && close<delay(close,1), 1,0)
+
if( close/mov(close,3) < delay(close/mov(close,3),1) && close>delay(close,1), -1,0)
+
… (continue series to 200) …
+
if( close/mov(close,200) > delay(close/mov(close,1),1) && close<delay(close,1), 1,0)
+
if( close/mov(close,200) < delay(close/mov(close,1),1) && close>delay(close,1), -1,0)

And today's chart output? It looks like this, of course:



A spike down (virtually 200 negative moving average divergences today), and of course a doji day. Negative divergence climaxes almost always happen during bullish phases (like now). The reverse is true for bearish phases.

Just some funny stuff from my 'puter.

[mss]

Great post, thanks for the formula data. I need something to keep me awake during this grind up till we get some sort of consolidation. Best to you, mss

[jjc]

For those that use this miserable TS easy language: inputs: Price( Close ); variables: x(0), val(0); Array: smaA0[200](1),smaA1[200](1); val = 0; For x= 1 to 200 begin smaA0[x] = Average( Price, x ); if ((close/smaA0[x]) > (close[1]/smaA1[x])) and (close < close[1]) then val = val+1 else if ((close/smaA0[x]) < (close[1]/smaA1[x])) and (close > close[1]) then val = val-1; smaA1[x] = smaA0[x]; end; Plot1( val, "sumofsmadiv" ) ;