Scanning a large stock base

[pkfryer]

I've read in many books (Elder springs to mind amongst others) that it is a bad idea to scan for a large number of stocks for entry signals for your trading system. That it is better to have a handful of stock and study them each day and learn their quirks and behaviour over time.

I do this with a dozen stocks that I have chosen for their good trading patterns, wide channels and volatility. I am also looking for very specific signals which get flagged up EOD. The signals dont come along that often per share, maybe once or twice a month, and I'm out of the market sometimes weeks at a time.

I'm just getting worried that I'm not trading as efficiently as I could and that I might be missing some really good trading opportunities just because I'm concentrating on my nest of eggs. I can't help thinking that if I could scan the entire FTSE100 or Dow etc, I could get a few signals a day that I could check out and see if they look promising and if the stock behaves the way I like... so at least I will have more options as to what and when to trade.

Surely, if a signal works with a fairly robust set of components it doesn't matter that you are intimately familiar with the stock and have paper traded it throughout the last 5 years of historic data etc? If it has the elements that I like, e.g. volitily, in a primary index, good channel width.. it will behave in similar ways as all the other stock I've chosen and open my eyes to other stock that have very easily tradable features and I can add to my nest.

What are your views on this? Is scanning good? Do you really need to be intimately aware of your stock to be able to trade them with a system?

How have you gone about choosing your shares or instruments?

Can you recommend a good (not too expensive) scanning application that allows you to program in reasonably specific (complex) entry signals and patterns

 

[TheBramble]

PK - a subject of many posts in the past.

I did start of down the 'scan the entire market' route myself and tried a number of approaches and packages.

Remarkably, none are really setup for those sorts of volumes.

You might also find (as I understand happened to one trader) that the datafeed supplier pulled the account, or at least limited it in some way, when they attempted to do just that.

To find a product thats is (a) compliant enough to be easily programmed with your selection criteria and (b) broad enough to allow inclusion of as many stocks as you want to monitor was difficult. The best I could come up with is EXCEL. However, it is immensely greedy of PC resource and anything over 100 stocks with DDE links slowed me down noticeably.

I then was advised to be a little more sensible in my approach. To preselect on a few major criteria. Effectively - to end up at precisely the point you're at now. With maybe a couple of dozen stocks that you are getting to know rather well.

I have even considered looking at just one stock. Another t2w member whom I hold in great regard, mentioned they had some success with this method. Getting to understand the 'depth' of one particular instrument.

Indeed, that was a post yesterday on another thread which went into some detail on doing precisely that - get to know just one instrument in depth before expanding your repertoire. Good advice I think.

PK - maybe you should examine your motivations for looking to expand your stock base. Is it impatience with a current flat trend in your normal stable? Happens all the time. And impatience can lead to losses on the 'barely met criteria' trade or the 'looks like it might in a minute so I'll jump in now' trade.

It might be interesting to set up a poll to find out how often people trade. I know some, like Naz trade a few trillion times a day while others make perhaps a couple of trades each month.

Maintaining your comfort level (which is as variable as the seasons and cycles) in your trading activity is essential. You inherently 'know' when you're over-trading or under-trading.

Go with your instincts, but I'd suggest you're just in a temporary flat spell.

Don't forget - you missed all the trades that you didn't take before you started trading. There will always be more.

 

[ithomas21]

pkfryer - You could experiment with Metastock's Explorer. You can specify detailed filter criteria with Metastock's proprietary language and then list particular attributes of the selected stocks.

I have scanned through the S$&P500 in the past and it handles complex filters pretty quickly.

As for the price or whether it fits your trading plan is up to you.

Oh, also try www.bigeasyinvestor.com

Ameritrade's package is quick and simple but not as flexible. (You can try it for free though.) Hope this helps.

 

[Mr Euro]

SSRT has a scanner for uk stocks and I use various scanners for the US market (real time, EOD and pe-market).

 

[pkfryer]

How do you use the scanners Mr Euro? Can you program them with reasonably complex criteria?

 

[Mr Euro]

Pkfryer,

Can you program them with reasonably complex criteria?

Not sure what you call complex but generally yes.

Good luck.

 

[TheBramble]

SSRT has a scanner for uk stocks and I use various scanners for the US market (real time, EOD and pe-market).

What is SSRT?

 

[Rognvald]

Tony
hazarding a guess - Sharescope realtime?

 

[Mr Euro]

Yes I personally don't like th product (as a real time package) but it does have some strengths and the EOD version is not bad. There are many that use it as the core part of their trading set-up just not for me thats all.