need a new macd indicator

[Deleted]  

Hi, i'm looking for a MACD alert (in the pop-up MT4 alert window) that will show the Price and Time when the MACD line crosses the zero line (intrer-candle).

The only MACDzeroLineCrossAlert's i can find give the alert at the close of the candle (4Hr for me) [e.g., MACD ColorHIst Alert] and do not show the time and price precisely when the zero cross occurs.

Thanks for your consideration,
steve

brewmanz
486
brewmanz  

Consider using lower TF for your MACD crossing. This could give you alert within (say) 1 minute of the event. If you're trading on H4, this should be 'close enough'

If using MACD(12,26,9) for H4, maybe use MACD(12*N, 26*N, 9*N) where N is ratio of timeframes (48 for M5, 240 for M1)

[Deleted]  
brewmanz:

Consider using lower TF for your MACD crossing. This could give you alert within (say) 1 minute of the event. If you're trading on H4, this should be 'close enough'

If using MACD(12,26,9) for H4, maybe use MACD(12*N, 26*N, 9*N) where N is ratio of timeframes (48 for M5, 240 for M1)

Thanks for the idea brewmanz. It seems to make much sense. But for some reason, it doesn't work - not even close. I wonder why?
brewmanz
486
brewmanz  
mt2414:
Thanks for the idea brewmanz. It seems to make much sense. But for some reason, it doesn't work - not even close. I wonder why?

I would expect similar results, but not identical. I've just tried it with H4 & M15 (using a ratio of 16). I used that as 16 also works as a zoom factor, so I could use MkI eyeballs to compare them both using an identical base timescale.

The H4 MACD(10,20.10) MACD shape is very similar to M15 MACD(160,320,160), including the signal line. MACD & other 'smoothers' rely on the slowly changing difference of relatively large numbers, an inexact science. Maybe the zero crossing points are different, but it may be arguable that the M15 is more accurate, as it reflects more price points in its calculation!

Also remember - any MACD can produce false zero crossing in the current bar (as in - disappearing after the event triggered) if you use anything but Open price, as the price changes (Close price will range from LOW to HIGH in the course of the bar being drawn).

So, in summary, there is no exact answer. My idea gives an approximation but maybe not close enough for your purposes. Never mind; we'll all continue our search for the perfect indicator.

[Deleted]  
mt2414:
Thanks for the idea brewmanz. It seems to make much sense. But for some reason, it doesn't work - not even close. I wonder why?

Of course it doesn't work because this is not correct

If using MACD(12,26,9) for H4, maybe use MACD(12*N, 26*N, 9*N) where N is ratio of timeframes (48 for M5, 240 for M1)

THE CORRECT METHOD IS THIS: H4 = M240 ; M240 = 8 X M30 also from this M240/8 = M30 THAN

If using MACD(12,26,9) for H4, maybe use MACD(12/N, 26/N, 9/N) where N is ratio of timeframes (12/8, 26/8, 9/8)for M30 and (12/48, 26/48, 9/48) for M5

Zsolt

brewmanz
486
brewmanz  

I think that Zsolt (and maybe mt2414) misunderstood what I was saying.

I'm saying that, given a MA12 curve on an H4 chart, then to get the same (well, similar) MA shape curve on a M5 chart you need to plot a MA(12*48) on the M5 chart, because there are 48 M5 periods in an H4 period.