| StockFetcher Forums · General Discussion · OBV crosses.... what is the correct syntax? | << 1 2 3 >>Post Follow-up |
| ImperialWhazoo 26 posts msg #38970 - Ignore ImperialWhazoo |
11/19/2005 4:58:42 PM I seldom ask anybody for help because I'm not trying some of the realy sophisticated things many of you folks are into. That having been said, I have a question of the community regarding the cross of the red & blue lines when the OBV is used. I've hammered away on this for literally hundreds of hours and I'm still not able to figure it out. So please lend me a hand, folks. First: Heres a filter I run: Show stocks where MA(40) has been increasing over the last 3 days and average volume(40) is above 300000 and Slow Stochastic(12) Fast %K is below 50 and close is between .05 and 5 and draw OBV(12) Now, here are the settings I am using: Draw MA(7) Draw MA(12) Draw MA(40) Draw Bollinger Bands(20,2.0) Here's what I'm aiming at: I am a day trader. I trade stocks between 5 cents and 5 dollars. I trade for a living. I have observed that, when these two sets of crosses happen within the same timeframe, a very profitable day-trade and/or swing trade follows just like clockwork. The problem is not one of screening for crosses of the %K & %D. The problem is screening for the crosses of the OBV lines. In my opinion, the event I'm trying to screen for will work regardless of the price range of the stocks you trade. Also, it can be used in reverse if you want to short. So, do the following steps Start by running the above filter with the above settings as I list them. Today (Saturday, Nov 19, 2005) I got 40 hits from this screen. Now, start pulling up the graphs in the list and notice a blue OBV line and a red OBV EMA(12) line. There is also a Slow Stochastic(12,3) graph with %K displayed in blue and %D in red. The screen I have not succeeded in constructing would spot the following events: a.) the %K below the %D but near to a cross of the %D from below. (Easy to do in Stockfetcher... not the problem I'm stumped by). b.) the blue OBV line about to cross the red OBV EMA(12) line from below. (I'm unable to figure out the correct StockFetcher syntax for testing for this event... this IS the crux of my question). c.) these two crosses setting up in the same timeframe. To be 100% precise, I'm not looking for these two crosses happening on exactly the same day. Neither am I screening to find these crosses on the day they have happened. The setup I'm after is where one (or both) of these crosses appear to be ready to happen tomorrow or the next day after that. The idea is to assemble a list of stocks where the event appears to be about to happen in the next day or so. It is simply my observation that in a high percentage of instances, when these two crosses coincide, there is normally a highly profitable trade that follows immediately afterwards (sometimes a really profitable day trade and sometimes a very profitable swing trade). I have tried and tried but I can not figure out how to test for the cross of the OBV (Blue) of the OBV EMA(12). As I said before, the %K %D cross is a no-brainer... not the problem. StockFetcher makes this an easy thing to test for. I'm at a dead-end trying to figure out how to test for the crosses of the two OBV lines. I can see them on the graph but I'm at my wits end trying to screen for them. So, thats my question: What is the correct StockFetcher syntax for screening for crosses of the red and the blue OBV lines? To see exactly what I'm talking about, run my screen (see above). Look at the following stocks from today's results: AVII - look at the Sept 29 & Nov 11 areas BKHM - look at the Sept 2 to Sept 12 area... & the Oct 20 to Oct 28 timeframe You can click thru the whole list and see many more examples of what I'm talking about, but these two will do. I trust that I'm not the only guy out here frustrated in his attempts to figure out how to use the crosses of the red & blue OBV lines in correct StockFetcher syntax. So, please help all of us out here who may have struggled with this. Best Regards ImperialWhazoo |
| dangreene 229 posts msg #38972 - Ignore dangreene |
11/19/2005 9:21:55 PM ImperialWhazoo, Here is a start for you using your price range and OBV(12) only. Tweak this by adding other indicators to your liking. I will be unable to respond further until 11/28/2005 as I will be out of town. Looking forward to your final version and comments from others. Good luck. Dan Greene |
| ImperialWhazoo 26 posts msg #38986 - Ignore ImperialWhazoo |
11/20/2005 9:07:22 PM Thanx dangreene - Here is your contribution incorporated into my screen. Show stocks where MA(40) has been increasing over the last 3 days and average volume(40) is above 300000 and Slow Stochastic(12) Fast %K is below 50 and OBV(12) crossed above CEMA(OBV(12),12) and close is between .05 and 5 and draw OBV(12) Please run it and you will get 3 stocks today (11/20/05): FEEC, ISO, and USEY So, click thru them and I think you will see my frustration. Your contribution shows up as a green line APPROXMATING the red line I am looking at. Now take a good hard look at the other displays: you will notice that the green line always is a very close approximation of the red line but it is always only that.... an approximation. If I have to live with an approximation of the redline then so be it but in my opnion, it seems to me that the following should be the applicable rule: a line such as the red OBV line, being obviously displayable, ought to be addressable via the syntax so that an event such as the cross I'm studying can be exactly studied (as opposed to approximated). So, when you get back in the saddle after the 28th, if you have any ideas on this (or if anyone out there in the community has anything to contribute... explanations, improvements, rotten eggs tossed at me.... whatever) I'd sure be appreciative. Best regards ImperialWhazoo |
| ImperialWhazoo 26 posts msg #38987 - Ignore ImperialWhazoo |
11/20/2005 9:10:41 PM dangreene- I ommitted this detail ---- look first at ISO to see my frustration in the OBV graph area.... The other two (FEEC & USEY) are less glaring examples of this approximation matter. ImperialWhazoo |
| ImperialWhazoo 26 posts msg #38993 - Ignore ImperialWhazoo |
11/21/2005 12:09:38 AM dangreene - I have wrestled all night with your suggested screen syntax and here's the issue: Your syntax throws a new line on screen and tests for when the original blue line crosses it. This is different from my need, which is to flag instances where the original blue line crosses the red line. Yours tests for a cross that is different. In the case of stock symbol ISO, the cross I am looking for happened Thursday of last week (Blue crossing Red from below) but Blue did not cross green (your new line) until Friday. A day late and the trade is not profitable. Thus, using your syntax "OBV(12) crossed above CEMA(OBV(12),12)" generates a result list but the fact that the green line and the red line are different matters in terms of timing the trade. Thanks greatly but I'm still looking for a working syntax for flagging instances where the red OBV line is crossed from below by the blue OBV line. Thanks for your help, dangreene. I thought you were onto something but I couldn't make it work. I wish somebody could tell me why it is impossible to screen for when the blue OBV line crosses the red OBV line. ImperialWhazoo |
| jclaffee 81 posts msg #39022 - Ignore jclaffee |
11/21/2005 9:40:41 PM ImperialWhazoo, I haven't pursued this thought all the way through but I offer it to you for your consideration. . . . .the SF User Guide does not appear to support any parameter variation for OBV. Could the use of "OBV(12)" be throwing it a curve that it is unprepared to handle? As an aside, when I ran your filter, the sytax checker told me there was a syntax error in the line "and Slow Stochastic(12) Fast %K is below 50". Regards (and happy filtering)! Jim |
| ImperialWhazoo 26 posts msg #39154 - Ignore ImperialWhazoo |
11/28/2005 9:12:48 AM jclaffee - Thanx 4 the thought. My testing of this grew out of a lengthy search thru past forums where I found the syntax in question. Since it is a known fact that the manual inadequately describes the present state of functionlity, it is my practice to do this manual scrutiny of the past forums in order to pick up ideas on things that are inapparent otherwise. So, in testing this thought, I've been pretty thorough. To my eye, (and again, I've dedicated a considerable amount of time testing), it appears to me that the results are uniformly as described previously regardless of whether I use 7, 10, 12, 22 or whatever. So, I don't think using 12 is the problem. This I can tell you.... and please take a few moments to set this experiment up because you will not be disappointed.... there is an astonishingly regular pattens that could be used to signal buys when both the Stochastic and the OBV are set up as I note. When both have crosses occuring withig days of each other, there appears to me to be a highly useful signal of an immenent upswing. I simply want one of the many genuinely brilliant forum participants to focus on this phenomena and either tell me that it is possible to construct a syntax that generates a daily list of tradable stocks to whom this "doiuble-cross" event has occurred or to tell me that there is a syntactical deficiency in Stochfetcher that prevents using the cross of the OBV EMA as a signal. Since my lengthy tests (albeit manual due to this problem with the syntax) have convinced me that this "double-cross" is a very useful, regularly occurring, and potentially profitable signal, I would be devestated if the outcome were to be that the OBV EMA could not be used as a signal. Help... anyone... I'm puzzled. Oh, and jclaffee, thanx again. ImperialWhazoo |
| dangreene 229 posts msg #39191 - Ignore dangreene |
11/29/2005 1:23:02 PM ImperialWhazoo, You said, ".... there is an astonishingly regular pattens that could be used to signal buys when both the Stochastic and the OBV are set up as I note. When both have crosses occuring withig days of each other, there appears to me to be a highly useful signal of an immenent upswing." I certainly see your point and am surprised that no one other than jclaffee offered any comment. Perhaps it is time to contact Stockfetcher support via "Comments, Bugs and Requests" to see if what you are trying to achieve is attainable. Good luck. |
| ImperialWhazoo 26 posts msg #39207 - Ignore ImperialWhazoo |
11/29/2005 9:34:49 PM dangreene - Boy do I ever agree with you. There is a pattern to things on the stockfetcher blog and I must tell you (and anyone else kind enough to chime in) that I've tried to contribute in the past but have never felt particularly welcome. Candidly, there are some pretty noisy voices out there in this blog and I've learned not to bother trying to contribute. So mostly, I just plod along on my own. Over the months, I've worked out almost 100 filters. I kind of think that, just maybe, some of them might be of interest to someone other than myself, but I don't think it is reasonable to have to tolerate being verbally slapped for the horroble crime of having a novel approach or for daring to crayon outside of the lines. This current filter, which I've dubbed "The DoubleUp-Cross", looks to be heavily loaded with potential. As you were so kind as to say, this thing looks pretty promising, which makes the lack of responses frustrating.... to say the least. HEY FOLKS!!! Doesn't anyone beside dangreene and myself have any interest in unearthing a filter that frequently signals dual digit returns in brief timeframes? Hello?!? Knock knock?!? Anybody home?!? There's something askew in this odd problem because the answer you (dangreene) offered (ie... CEMA OBV) was the one I'd tried at some point prior to asking for help.... and logiaclly, it would seem to be the right answer, but it is not. As I pointed out in my reply to you, the two lines are only approximately the same. In any event, I remain of the firm opinion that it shouldn't even be necessary to have to resort to approximating the OBV EMA(12) with the CEMA OVB(12,12).... it should be that the OBV EMA(12) is itself addressable within proper stockfetcher syntax, thereby making resorting to a custom EMA unnecessary. So, can anybody shed some light on this? Regards etcetera (and more thanx again to dangreene), ImperialWhazoo |
| ImperialWhazoo 26 posts msg #39208 - Ignore ImperialWhazoo |
11/29/2005 9:45:40 PM Oh and jclaffee - I apologize for not making mention of your comments too. To be 100% accurate, let me restate that it puzzes me that only you (jclaffee) and dangrene have respondes... and thanx to you both. Oh also, if folks want to see 4 prime examples of tickers I've found via my screen that clearly show what I'm talking about, set your settings as I described in my first blog entry and run the filter. Then go look at the confluence of the Stochastic signal and the OBV signal on these tickers for the dates noted. You will be amazed at how they signal the correct timing for a sizable gain immediately following the signals: XYBRQ (11/17), UPDA (9/30), PWTC (9/29), & LTHU (9/29). Regards etcetera... ImperialWhazoo |
| StockFetcher Forums · General Discussion · OBV crosses.... what is the correct syntax? | << 1 2 3 >>Post Follow-up |
Copyright 2022 - 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