StockFetcher Forums · View by Author: (62 messages)  [ Display By: Date / Subject ]<< 1 ... 3 4 5 6 7 >> 
Filter Exchange · Explosion Filter??
murknd
msg #27436
7/17/2003 11:49:32 PM

..whoops - ater all that careful editing

... I mean >25000 in my comment about volume.

D


Filter Exchange · Explosion Filter??
murknd
msg #27435
7/17/2003 11:47:25 PM

... well OK maybe not, but this filter (and perhaps a variation or 2) is very close to it. Admittedly it is a variation on BB Squeeze, but I noticed many times right at the squeeze the high is above the UBB and when accompanied by a spike in volume it's not too long thereafter there can be "explosive-type" behavior. I figured the best way to shake it out is to throw it to the "BB squeeze hounds" (I think I'm becoming a convert) and let them chew on it awhile. I've done a fair amount of back testing, but more is definitely prudent. I think one of the keys is the new 30 day BB width low, but you can also get many more matches by deleting this and just modulating the BB width. I think coupled with some rigorous trailing stop discipline (and of course the usual dose of patience) it may be a very powerful filter - but I leave that assessmnet to the judgement of the SF community - to whom I tip my hat for the significant amount of learning over the last few weeks.

Anyway .. have at it ... let me know if you have any questions about varations - I've tried many.

.. and FYI - I think the average volume(90) <25000 is important because many times these "sleeper" stocks don't have much more volume - try even 10K.

D

P.S. I WILL claim authorship on this one.

Fetcher[bollinger width(20) has reached a new 1 month low and bollinger width(20) is below 0.16 and high is above upper bollinger band(20) and low is above lower bollinger band(20) and volume has gained more than 80% over the last 1 day and close is between 0.1 and 10 and average volume(90) is above 25000]





General Discussion · Noah's Gold Mine
murknd
msg #27396
7/15/2003 11:28:00 PM

... and I would be very careful with your $25K if you're new to trading - it could evaporate before your eyes - start slow and be VERY patient ...

D


General Discussion · performance
murknd
msg #27351
7/14/2003 8:28:32 AM

performance is from the date of the match until the most current close ... if you click on performance (column heading) it will open a new window where you can check 1,2,3,4 day, 1 week , etc. "performance" for any date offset ...

Hope that helps - good luck,

D


Filter Exchange · Bollinger Band Squeeze
murknd
msg #27346
7/13/2003 9:36:09 PM

I've been fiddling with these BB squeeze filters as well and simply wanted to post where I'm at - I think they are largely variations on "themes" already posted so I apologize if they're redundant and I don't claim "authorship" - I wouldn't mind a comment or two if anyone see's anything new. I think the first one (variation on ig0r) and the fourth one (variation on petrolpeter) look pretty good - the key being to "detect" the BB expansion immediately the day it happens and to make sure bullish signals (volume up, green at open, etc.) are present.

I noticed BSQR meets this criteria and I think met Muddy's filter a couple of days ago. Surprisingly (perhaps not so surprisingly) BSQR also passed my MACD filter Friday.

I saw Todd's post on yahoo about "too many" filters and I agree it can get very confusing - I spent a couple of hours today just trying to organize "My Filters" in SF - but I like his idea about filter "owners" and a little more method to this madness - I'm also working on the "explosion" filter (I think Karen's idea) and getting close - I'll post separately here in a few days.

One other comment - the global filter is the same one I'm using now for my MACD filters - however it is slightly different than the global filter in my previous post "Building MACD Filters" - notice I got wise to other sectors (besides technology) - I guess there's money to be made elsewhere - LOL ...

Global Filter

Fetcher[close is between 0.1 and 10 and volume is above volume 1 day ago and average volume(90) is above 50000 and low is increasing and close is increasing and open is above close 1 day ago]



BB Squeeze 1

Fetcher[bollinger width(20) 1 day ago dropped between 0 and 50 percent over the last 4 days and bollinger width(20) is between 0 and 0.5 and bollinger width(20) gained between 5 and 100 percent and price touched upper bollinger band(20)]



BB Squeeze 2

Fetcher[bollinger width(20) reached a new 1 month low 3 weeks ago and bollinger oscillator(20) has been increasing the past 3 days and price touched upper bollinger band(20) the past 3 days and FI(15) is above 0 and OBV has been increasing the past 2 days]



BB Squeeze 3

Fetcher[bollinger width(20) reached a new 1 month low 3 weeks ago and bollinger width(20) has been increasing the past 3 days and price touched upper bollinger band(20) the past 3 days and FI(15) above 0 and OBV has been increasing the past 2 days]



BB Squeeze 4

Fetcher[bollinger width(20) reached a new 30 day low within the last 2 days and bollinger width(20) is increasing and high is above upper bollinger band(20)]




Filter Exchange · Short Term Filter
murknd
msg #27269
7/8/2003 8:17:09 PM

you've missed a great opportunity once the MACD fast line has crossed above the slow line

Try this -

Fetcher[Low is above Low 1 day ago
and Volume is above Volume 1 day ago
and open is below close
and open 1 day ago is below close 1 day ago
and close is between 1 and 2
and MACD histogram is below 0
and Average Volume(90) is above 50000
]



run it 1,2,3,4 days ago and you should see a number of stocks that had a MACD xover in the last day or so

see my "building MACD filters" post in filter exchange - I'm working on some enhancements.

D


General Discussion · MA turning up
murknd
msg #27266
7/8/2003 6:07:05 PM

Fetcher[show stocks where MA(10) has been decreasing for 4 days lag 4 days and MA(10) has been increasing for 4 days]



Just adjust the number of decrease/increase days and the number of lag days until you get the desired "saucer."

D


Filter Exchange · Building MACD filters
murknd
msg #27229
7/6/2003 12:31:35 AM

... just remember they have many "personalized" (I like technology, I like price between 0.7 and 12, etc., etc.) components ... and I don't use them solely by themselves - I use them in combination with others in my previous post ... but here's the complete filters (actually 2 - they are similar, but give slightly different results and if you look at the histogram it's hard to argue it's not trending up and "looks as if it will xover in the next few days") ...

Fetcher[MACD Histogram has been decreasing for 4 days lag 4 days and MACD Histogram is between -0.10 and 0 and MACD Histogram has been increasing for 2 days and MACD(12,26) has been decreasing for 4 days lag 3 days and MACD(12,26) has been increasing for 1 days and fast stochastic(15,3) fast %K is above 30 and fast stochastic(15,3) fast %K is increasing for 2 days and close is between 0.7 and 12 and volume is above volume 1 day ago and volume is above 100000 and sector is technology]



Fetcher[MACD(12,26) is decreasing for 4 days lag 1 days and MACD(12,26) is increasing for 1 days and MACD Histogram is between -0.12 and 0 and MACD histogram is increasing for 2 days and fast stochastic(15,3) fast %K is above 15 and fast stochastic(15,3) fast %K is increasing for 2 days and close is between 0.7 and 12 and volume is above volume 1 day ago and volume is above 100000 and sector is technology]



D




Filter Exchange · Building MACD filters
murknd
msg #27226
7/6/2003 12:14:02 AM

I've used the following filters with pretty good success (for swing trading - see my "level-set" post in general discussion) recently - again, the filters don't "magically" reveal winners every time, so you have to couple them with chart-reading. I also try to watch pre-market and pretty reliably the stocks that are trading and bidding up in pre-market usually do well 1-3 days - especially through the 1-2 days following a MACD xover.

Here's my logic ...

MACD xover is a very reliable signal in virtually any market - those that have a steeper slope of the MACD histogram are even more so (reliable) - so the basic filter is ...

Fetcher[MACD fast line(12,26) crossed above MACD slow line(12,26)]



... the bullish signal is even stronger when the fast line is increasing (rather than when the slow line is decreasing - although the slow line may very well still be decreasing at the xover point). So I add this to the basic xover and get ...

Fetcher[MACD fast line(12,26) crossed above MACD slow line(12,26) and MACD(12,26) has been increasing for 2 days]



... the next obvious thing is that the histogram has to be increasing in order for there to be a xover. I like to see the histogram increasing (that is getting less and less negative) for 3 days and the fast line increasing for 2 days. This helps filter out stocks that are acting more volatile above / below MACD xover. So now I get ...

Fetcher[MACD fast line(12,26) crossed above MACD slow line(12,26) and MACD histogram has been increasing for 3 days and MACD(12,26) has been increasing for 2 days]



Now imagine if you could filter for xover's 1 or 2 or 3 days prior to the xover. Not only do you potentially increase your probability for gains, but you have a "rolling" watch list.

I've experimented with a number of variations on this theme and you can (should) do the same, but typically the MACD histogram will reverse (i.e. a bottom) 1-2 days before the fast line reverses (forms a bottom). If you see a histogram "bottom" it will be followed by a fast line bottom (if the stock is really forming a bottom). Here's what these filters look like .. first the MACD Histogram bottom ...

Fetcher[MACD Histogram is decreasing for 4 days lag 1 day and MACD Histogram is increasing for 1 day]



... I like to add a histogram value qualifier (to filter out stocks that are many days to weeks away from a xover) and I get ...

Fetcher[MACD Histogram is decreasing for 4 days lag 1 day and MACD Histogram is between -0.14 and 0 and MACD Histogram is increasing for 1 day]



... now adding the MACD fast line increasing (recall the histogram is already increasing) so I get ...

Fetcher[MACD(12,26) is decreasing for 4 days lag 1 days and MACD(12,26) is increasing for 1 days and MACD Histogram is between -0.12 and 0 and MACD histogram is increasing for 2 days]



... the logic here again is that the histogram will bottom first, followed by a fast line bottom and then the histogram will approach zero from below.

I've tried to put these 2 concepts together and got this ...

Fetcher[MACD Histogram has been decreasing for 4 days lag 4 days and MACD Histogram is between -0.10 and 0 and MACD Histogram has been increasing for 2 days and MACD (12,26) has been decreasing for 4 days lag 3 days and MACD(12,26) has been increasing for 1 day]



MACD is a trend indicator so I also like to couple it with an oscillator like Fast Stochastic so I add (or chain) the following to all the filters above because I like to sort for those that have definitely formed a bottom and are now cycling back up ...

Fetcher[fast stochastic(15,3) fast %K is above 30 and fast stochastic(15,3) fast %K is increasing for 2 days]



... using a level of 30 once both the histogram and fast line have bottomed, and a level of 20 if I'm just looking for a bottom on the histogram.

I also use the following (mostly from posts on this forum - thank-you) as validation to help "narrow the field" ...

Fetcher[Williams %R (25) crossed above -40.00 within the last 1 day]



Fetcher[linear regression slope(10) crossed above 0]



Fetcher[RSI(2) crossed below 10% within the last 5 days and RSI(2) crossed above 10%]



Fetcher[MA(3) crossed above MA(8)]



I also use the following global filter ...

Fetcher[close is between 0.7 and 12 and volume is above volume 1 day ago and volume is above 100000 and sector is technology]



I usually end up developing a daily watch list of 10-20 stocks (once I've studied the charts) that pretty reliably include some nice (~5%) winners every day - do I pick those winners everyday? - of course not or I would be trading for a living - but I think this is a pretty logical system for those that swing trade and can't watch the market 24/7.

I'd appreciate any in-sights (constructive criticism welcome) you more experienced traders / analysts could give to tweak these concepts for various market conditions.

D



General Discussion · Levelset
murknd
msg #27208
7/5/2003 7:42:55 PM

I've tried to keep an open mind looking at all the filters / discussion everyone posts and it's pretty obvious there's a varied experience level and trading "personality" represented. I don't know about everyone else, but I thought it might be helpful to keep the following filters (pun intended) in mind as we post and read about this screen and that screen ...

1) There's no such thing as a "magic" filter that returns 100% (or perhaps 70% or 80%) winners day-in and day-out in all market conditions.
2) I've back-tested many of the filters that have been posted in filter exchange and some are "better" than others, but they all have "varied" performance depending on market conditions and countless other factors. Screening should be about choosing which stocks (of 1000's) to study further. Anyone can build a filter that returns 200 or 300 stocks - what good does that do?
3) I don't hear too much talk about entry and exit strategies. Inevitably any "good" filter will return numerous stocks that meet the filter criteria - which one or ones do you enter? There's no substitute for studying the individual charts at this point - that's one of the reasons I joined SF - because I could quickly scan through the charts associated with my specific citeria.
4) I think there are some of you that trade for a living (many perhaps that are day-traders) and your filters have been posted as "the best" this and "the best" that. Well I, for one, don't trade for a living and can't watch the market constantly throughout the day - so any filter that's posted with the caveat -- I then watch the "such and such" at "stocksrus.com" to determine entry and exit throughout the day -- is a filter that doesn't really help me (I know, I know, it does help some of you). It would be nice if people at least indicated their filters work best in "xyz" conditions or with "abc" exit and entry strategies.

That's probably enough rambling for now - just trying to keep everything in perspective ... thanks for listening.

D


StockFetcher Forums ·  · << 1 ... 3 4 5 6 7 >>

*** Disclaimer *** StockFetcher.com does not endorse or suggest any of the securities which are returned in any of the searches or filters. They are provided purely for informational and research purposes. StockFetcher.com does not recommend particular securities. StockFetcher.com, Vestyl Software, L.L.C. and involved content providers shall not be liable for any errors or delays in the content, or for any actions taken based on the content.


Copyright 2018 - Vestyl Software L.L.C.Terms of Service | License | Questions or comments? Contact Us
EOD Data sources: DDFPlus & CSI Data Quotes delayed during active market hours. Delay times are at least 15 mins for NASDAQ, 20 mins for NYSE and Amex. Delayed intraday data provided by DDFPlus


This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.