3SOF

Quote from: Quasi on December 17, 2006, 10:46:11 PM

sets min 20 day ave liquidity at greater than $500K
and
[SMA(20,close) * SMA(20,volume) > 500,000]

I've had a couple of posters email me as to why I'm multiplying criteria parameters in the above scan section, so thought I would  expand and post the answer here also.

Now this is a feature of the Stockcharts.com advanced scan engine, not all scanning sites allow you to use arithmetic operations within the scan criteria.

Really all I'm looking for is liquidity, for most scans I really don't care that much what the price or volume are independently.  What I need to know is does it trade enough dollars on an average day to make it easy to get in and out.  I usually don't want have a position any larger than 5-10% of the daily average dollars traded.  Note price or volume might be another criteria later in the scan, but in the beginning I want to weed out any stocks without liquidity.

Before the advanced scan engine I would create ranges like between $0.5 - 2 I want min volume of 1M,  and between $2-5 I want min volume 500K, I think you get the picture.

So in the above example a stock at 50 cents would need to have volume of 1M, where as a stock at $10 would only need to have volume of 50K to meet the minimum criteria of trading $500K per day.

Hope that helps
happy trading / investing
Quasi

Hi all, best of the season to you.

I just modified my T3ML scan and posted the results over on Terliso's T3ML board.

Basically lowered the liquidity requirement to $200K and allowed the macd to be slightly more negative.  These mods will pick up more possibilities but will also let in more trash which will have to be sorted out by reviewing the charts.

happy trading / investing
Quasi

[type = stock]
and [Close > 0.5] and [Close < 10]
and
[SMA(20,close) * SMA(20,volume) > 200,000]
and
[daily sma(50,daily close) >= daily sma(200,daily close) * 0.95]
and
[daily sma(50,daily close) <= daily sma(200,daily close) * 1.05]
and
[daily ema(200,daily close) >= daily sma(200,daily close) * 0.95]
and
[daily ema(200,daily close) <= daily sma(200,daily close) * 1.05]
and
[close > daily sma(50,daily close)]
and
[close > daily sma(200,daily close)]
and
[close > daily ema(200,daily close)]
and
[[volume > daily sma(100,daily volume) * 2.5]
or[1 day ago volume >  daily sma(100,daily volume) * 2.5]
or[2 day ago volume >  daily sma(100,daily volume) * 2.5]
or[3 day ago volume >  daily sma(100,daily volume) * 2.5]
or[4 day ago volume >  daily sma(100,daily volume) * 2.5]
or[5 day ago volume >  daily sma(100,daily volume) * 2.5]]
and
[[weekly macd signal(12,26,9) >= -.1]and[weekly macd signal(12,26,9) <=0.5]
or[1 week ago weekly macd signal(12,26,9) >= -0.1]and[1 week ago weekly macd signal(12,26,9) <= 0.5]
or[2 weeks ago weekly macd signal(12,26,9) >= -0.1]and[2 weeks ago weekly macd signal(12,26,9) <= 0.5]]
and
[[1 week ago weekly macd line(12,26,9) < 1 week ago weekly macd signal(12,26,9)]
or[2 weeks ago weekly macd line(12,26,9) < 2 weeks ago weekly macd signal(12,26,9)]
or[3 weeks ago weekly macd line(12,26,9) < 3 weeks ago weekly macd signal(12,26,9)]]
and
[[weekly macd line(12,26,9) > weekly macd signal(12,26,9)]
or[1 weeks ago weekly macd line(12,26,9) >1 weeks ago weekly macd signal(12,26,9)]
or[2 weeks ago weekly macd line(12,26,9) >2 weeks ago weekly macd signal(12,26,9)]]

Wow Quasi, that looks great....too bad i dont understand most of it...lol

I just use the free version of stockcharts so I couldnt use ur great filter anyway, thats too bad for me.

So, is this new scan doing a great job for you? Sure does look like it helps u find those elusive T3ML stocks.

Did it pick up FDMLQ..i posted on Terlisos 3ML site, but got no reply on it.

Well..nice job ...and Happy Holidays

thanks

julia

Quote from: gmarc66 on December 24, 2006, 10:45:17 AM
Did it pick up FDMLQ..i posted on Terlisos 3ML site, but got no reply on it.

Hi Julia, just back from a 3 week vacation and going thru all the messages.  Yes FDMLQ has had a very nice run, good pick.  I'd be wary right now though as it looks ready for a pretty good pullback, probably into the high 0.40's.

My scan wouldn't have picked that one up as I usually don't get involved with BB stocks or anything that trades less than $500K daily on average.  And also I usually go for a stock price between $1 - $10.

However the scan is easy to modify, just change the liquidity setting to say $200K, and adjust the stock price range down to say a lower end of $0.25 and I'm sure you will get many more possibilities to look at.

hope you had a great holiday.

happy trading / investing
Quasi

   Woooah Quasi... Cool Scan.   Much more indepth than mine have been as I've tried to keep them relatively tame but whatever works.  I'll have to run it tonight and see what it pulls up but I'd then likely run the gauntlet of my other signals over the scan results.

As for Aussie Trader and running Automated Scans and using Mechanical Trading (or system trading)....  You may have something there if we were able to write a Scan(s) that was Foolproof,  however if a scan is good but keeps pulling up 80 stocks or so, you need to then keep narrowing down the parameters to scan over and over till you get a short list, kinda like interviewing candidates for a job, eventually getting a top 5 list.  Then add some Fundamental Analysis and we could likely generate some seriously profitable swing trades on the top 3.

Thing is there's a Gazillion parameters to meet everyones trading style so how could we agree on anything?  The other thing is most of here don't day trade, we swing trade so again most anything with a momentum indicator is toast.

I've now got about 6 good scans that I look at periodically but not nearly as much as I'd like.   Perhaps if we put our collective minds together to determine if there are mutual indicators of interest that appear affective and build one scan than encompasses them all...

Would that even be possible?

Quote from: Deadb440 on January 15, 2007, 11:02:01 PM
  Woooah Quasi... Cool Scan.   Much more indepth than mine have been as I've tried to keep them relatively tame but whatever works.  I'll have to run it tonight and see what it pulls up but I'd then likely run the gauntlet of my other signals over the scan results.

As for Aussie Trader and running Automated Scans and using Mechanical Trading (or system trading)....  You may have something there if we were able to write a Scan(s) that was Foolproof,  however if a scan is good but keeps pulling up 80 stocks or so, you need to then keep narrowing down the parameters to scan over and over till you get a short list, kinda like interviewing candidates for a job, eventually getting a top 5 list.  Then add some Fundamental Analysis and we could likely generate some seriously profitable swing trades on the top 3.

Thing is there's a Gazillion parameters to meet everyones trading style so how could we agree on anything?  The other thing is most of here don't day trade, we swing trade so again most anything with a momentum indicator is toast.

I've now got about 6 good scans that I look at periodically but not nearly as much as I'd like.   Perhaps if we put our collective minds together to determine if there are mutual indicators of interest that appear affective and build one scan than encompasses them all...

Would that even be possible?

Hi Dead,

Yes I do get a little in depth on some of my scans, but with the new computers at Stockcharts and the new scan engine it seems to handle them quite well, usually get results in a few seconds, although it might be 15 seconds on a longer scan and I get 999 hits and then have to narrow the parameters.   Also I only run these after hours, you never want to use up the horsepower during the trading hours.

As far as scanning linked to mechanical trading, sounds great, but if you ever get anything that works, you better keep it to yourself.  Otherwise if we all used it, it would stop working.  Same thing all the big guys are doing with automated algorithmic trading, they go to great lengths to keep their secrets, and somebody is always trying to outsmart or outguess your algorithm.

I also run all the scan outputs thru my standard chart parameters for each particular scan setup.  Agree I / we could put them all in the same scan using "or" connectors between the different scan sections, and I sometime do that but only for scans where I would be using the same chart setup to look at them all.  IE I might link ("or" together) several scans that are all looking for setups on the daily chart, then dump to a folder and set all charts to my standard daily chart setup and view full screen in the 10 per page mode.  But the weekly scans I would run them separately because I use a different chart setup for those ones.

For all the picks I find that look interesting but aren't quite ready, I through them into a watch folder.  I then have a separate scan I run every day or two that scans that folder for unusual activity, percent price change or high volume change etc.  that way I get flagged if something is starting to happen.  Still learning and usually putting way too many stocks on the watch list and then miss the initial move by several days because my attention has been sucked to some other stock for a few days., thus trying to automate the flagging process.

Another trick I use (I'm sure most of you also do it) is to build a text file of all your standard scan parameters, like building blocks.  Then it is very easy to build a new scan by cutting and pasting and splicing them together, and you already know that all the individual sections have been previously tested for syntax coding errors.

Always tweaking and looking for new ideas.

happy trading / investing
Quasi

I've never used the ASI but my favorite scans utilize the macd, full stoch, SPX ratio, OBV ma(20), and the linear regression bounces on 3 and 6 months...  http://www.tickertuner.com/lineareg/pricespx.aspx

Quote from: Quasi on January 16, 2007, 12:44:14 AM


As far as scanning linked to mechanical trading, sounds great, but if you ever get anything that works, you better keep it to yourself.  Otherwise if we all used it, it would stop working. 

Interesting point Quasi, although in reality probably not true as 99.99% of people would not follow the system 100% by definition they would emotionally interfere as it were.

Quote from: Quasi on January 16, 2007, 12:44:14 AM
I also run all the scan outputs thru my standard chart parameters for each particular scan setup...... IE I might link ("or" together) several scans that are all looking for setups on the daily chart, then dump to a folder and set all charts to my standard daily chart setup and view full screen in the 10 per page mode.  But the weekly scans I would run them separately because I use a different chart setup for those ones.

For all the picks I find that look interesting but aren't quite ready, I through them into a watch folder.  I then have a separate scan I run every day or two that scans that folder for unusual activity, percent price change or high volume change etc.  that way I get flagged if something is starting to happen. 

Another trick I use (I'm sure most of you also do it) is to build a text file of all your standard scan parameters, like building blocks.  Then it is very easy to build a new scan by cutting and pasting and splicing them together,

I think you are selling yourself short here Quasi, what you describe doing is probably streets ahead of what 90%+ of traders out there are even contemplating. You sound like you have a very methodical, replicatable and systematic approach which should give you an edge in the market overtime. And an edge is basically what you want to have to be succesfull in the market.

Quote from: Deadb440 on January 15, 2007, 11:02:01 PM

As for Aussie Trader and running Automated Scans and using Mechanical Trading (or system trading)....  You may have something there if we were able to write a Scan(s) that was Foolproof,  however if a scan is good but keeps pulling up 80 stocks or so, you need to then keep narrowing down the parameters to scan over and over till you get a short list, kinda like interviewing candidates for a job, eventually getting a top 5 list.  Then add some Fundamental Analysis and we could likely generate some seriously profitable swing trades on the top 3......

What you describe here Deadb is a great systematic approach to filtering from thousands of stocks down to a large candidate list and then further filtering (by fundamentals / favourite chart pattrens etc to get a stock or stocks to enter. That is exactly what David does for the 3SoF picks.

The system trading I am using / working on scans the selected market each night with the system parameters / entry conditions and presents you with a handful of trades to be placed the following day. The only further filtering to be done (if I had too many signals to enter them all) would be to randomly choose from the trade signals (up to my maximum number) which ones to enter. I could also be systematic and rank them by liquidity and work my way from highest liquidity first instead of purely random.

In the next few days I will share 'some' details on where I am at and what tools / resource I have / am accessing for this system trading approach.

I just read the thread and was wondering if you have shared the other tools and resources that would also help us that getting started in this wonderful world.
Thanks.

some of the more valuable tools i'm using is called Trading solutions
you can download it <a href="http://www.jdoqocy.com/click-2459552-10401460" target="_top">here</a> it free
<img src="http://www.tqlkg.com/image-2459552-10401460" width="1" height="1" border="0"/>

I am new to the site and don't know enough about the Stockcharts.com Advanced Scan Interface to make an algorithm.  Is there anyone out there that is willing to write a simple program  that can be used on the Stockcharts.com site? I have a criteria using one indicator with price.
wiley