kozzythedog
7 posts
msg #81796
- Ignore kozzythedog |
10/26/2009 9:24:04 PM
I came across this post on another blog. It looks very interesting, but could be hard to find.
I am not a programmer, but I would like someone to build a scan for this. By adding some additional overbought indicators, could only help to confirmed the move.
So thanks!
zDon
So here it is:
Take the bar that the alleged High was made. The next
bar must take out the low of the bar that made the High and close lower.
Then, the next bar had to take out the High of the bar that took out the low
of the High day's bar and close lower. (Look at Wed (oct 21,2009)., Thurs., Friday on the
SPY. That's it in the flesh.) (This is confirmed on the daily chart)
To be valid, this sequence had to be complete within 5 days
- the sooner the better. Three days, obviously, is the bare minimum for the
set-up to occur.
|
chetron
2,817 posts
msg #81802
- Ignore chetron
modified |
10/26/2009 10:34:44 PM
maybe...
MAYBE....
|
moby
314 posts
msg #81813
- Ignore moby |
10/27/2009 1:07:51 AM
Man Chetron, you Rock!
That is mighty quick work on the code translation and it works perfect. Bravo.
|
chetron
2,817 posts
msg #81814
- Ignore chetron |
10/27/2009 6:57:46 AM
THANX = D
|
kozzythedog
7 posts
msg #81841
- Ignore kozzythedog |
10/27/2009 12:09:07 PM
Great Job Chetron! I'm going to paper trade some of these results and see how they roll out. I've added some overbought indicators as well.
Hope we got a winner!
Thanks again!
|
chetron
2,817 posts
msg #81844
- Ignore chetron |
10/27/2009 12:48:10 PM
kozzythedog, WHAT INDICATORS ARE YOU USING?
|
kozzythedog
7 posts
msg #81862
- Ignore kozzythedog |
10/27/2009 3:57:07 PM
I'm using RSI(3), Stoch(8,4,4), CCI(10), Momentum(10). I'm looking for the RSI to be over 99, Stoch to be above 95, CCI to be moving off the top and the Momentum to be pointing down. I also have a Linear Regression Channel for further confiration of resistance.
Do you or anyone else have any other suggestions on making this more accurate?
|
trendscanner
265 posts
msg #81886
- Ignore trendscanner
modified |
10/27/2009 7:59:03 PM
Just thinking out loud (always a dangerous activity), you might want to add the following line, since you indicate you're looking for a market top, not a top for an individual stock
apply to symlist(spy)
SPY doesn't behave exactly as the SPX but it's a pretty close proxy.
Something else that might help confirm a top is a negative divergence in MACD or similar indicator, i.e., SPX is making a new high but RSI(14) or MACD is not confirming the new high.
I don't know how to search for these divergences particularly well, but one way that might work is something like this (assuming that we're looking for, say, a negative divergence in rsi(14))
set{DRSI, days(rsi(14) reached new 4 month high, 100)}
drsi > 30
rsi(14) below 4 month high for last 30 days
Hope that helps
|