This post was contributed by a guest author, and does not necessarily reflect the views of Richard or MovetheMarkets.com
Here’s the results from the NR7 Scan for tomorrow, sorted by volume:
MAR, CMI, JAH, GGG, CHIC, GDI, LNET, CCMP, HTI
These are not recommendations, but NR7’s that might be of interest to put on your watchlist for tomorrow. I’ll be watching for these stocks to break the high or the low of today by a few ticks, with the stop on the opposite side of the candle, like I did for MCGC and CAH. I’d also look at the daily charts to get a feeling of longer term bias, whether long or short, which could influence me to bypass an entry signal if it was in the “wrong” direction. I’d pick a target based on recent support or resistance, and I wouldn’t enter unless that risk:reward is about 3 or better.
Any comments on how to improve the system are appreciated!
This post was contributed by a guest author, and does not necessarily reflect the views of Richard or MovetheMarkets.com
May 1st, 2007 at 8:31 pm
DNDN April 30th looks like NR7. Guess the volume did not contract enough to meet the scan criteria. Big move on the open but it didn’t last long. I did not trade it.
May 2nd, 2007 at 5:52 am
DOH! I looked at it, and these lines kept DNDN out of the scan yesterday:
/*Find recent huge volume movers from the NR7 results
—————————————————————*/
and average volume(5) is more than 50% above average volume(30)
and volume reached a new 12 week high within the last 3 weeks
The average volume(5) was below average volume(30), not above. Also, the 12 week volume high happened about 5 weeks ago. When I take out these filters, DNDN shows up. I don’t think that these filters are filtering the results in the way I want them to, so I’m killing them. The scan now gives about 27 results yesterday, and 59! today.
Any ideas on better filters to use? I’m thinking about specifying a small candle compared to an Average True Range, or maybe a wide spread between the highest day range and the lowest day range in the last 7 days (WR7 compared to NR7)…
May 2nd, 2007 at 1:28 pm
Just one opinion, but I think you are changing the code too fast. It’s hard to keep track of whether todays changes would break the scan from 5 days ago, you know? It’s ok if it misses some candidates that a human can find. As long as you’re getting something out of it, I’d just collect data for a couple weeks and then decide what needs tweaking, if anything.
May 2nd, 2007 at 2:11 pm
You’ve got a good point, Richard. I want the scan to find the patterns Jamie calls out at least, though.
I guess I should just collect all of the NR7 results and track the performance and analyze it after a while. I think I’ll use Stocktickr for that. :)
May 2nd, 2007 at 2:27 pm
Disregarding whether it’s possible to mechanically find everything Jaime calls out, I do think you’ll be better off tracking it and making changes after you have more data. Good luck, seems like you are off to a good start!
May 2nd, 2007 at 5:19 pm
Prospectus,
If the scan is working for you, don’t change it on my account. I have a makeshift NR7 scan which uses a combination of inside days, volume and ATR. It’s far from perfect but once in a while it gives some good candidates.
Have you traded any new NR7s since your last post?
May 2nd, 2007 at 5:29 pm
Jamie,
Just CHIC today, which is currently underwater but not stopped out yet. The candidates I have been filtering out of the bulk NR7’s don’t seem to move much. I’m still working on better filters, since 50-100 NR7’s to wade through is a tall order. So the scan isn’t working all that well for me yet, but I think that there is promise if I can intelligently filter the results. I’ll keep everybody posted!