- Katılım
- 23 Eki 2020
- Mesajlar
- 1,826
Before I even attempt to do this, first can someone tell me if would I be able to create an Expert for the following that would color each bar according to the following criteria which includes whether it's an up/down day, whether it's a wider or narrower spread than the previous bar, whether volume is higher or lower than the previous bar, and whether the close is in the upper/lower half of the bar's range: Up Day - Positive Close Up day, higher vol, narrow spread, close higher half. Up day, lower vol, narrow spread, close higher half. Up day, higher vol, wider spread, close higher half. Up day, lower vol, wider spread, close higher half. Up Day - Negative Close Up day, higher vol, narrow spread, close lower half. Up day, lower vol, narrow spread, close lower half. Up day, higher vol, wider spread, close lower half. Up day, lower vol, wider spread, close lower half. Down Day - Positive Close Down day, higher vol, narrow spread, close higher half. Down day, lower vol, narrow spread, close higher half. Down day, higher vol, wider spread, close higher half. Down day, lower vol, wider spread, close higher half. Down Day - Negative Close Down day, higher vol, narrow spread, close lower half. Down day, lower vol, narrow spread, close lower half. Down day, higher vol, wider spread, close lower half. Down day, lower vol, wider spread, close lower half. Thanks, Harold Harold Yes I'm positive this can be done. I've just posted a chart on the MetastockUsers Yahoo group which clearly shows how my own charts have bars with one of four colours. I achieve this by having a simple custom expert imbedded in my default template. The colours I use are white, yellow, blue and red. The biggest problem you might face is getting the logical priority right. If the logical selection of colours is clear-cut then you're home free. If there are "overlaps" in the colour selection then you might need to do a bit more thinking before setting up the expert or you'll end up with a fair amount of confusion. I haven't run down through your selection criteria so you may already have the logic sorted. Roy Roy, Great. I looked at your chart on the Yahoo group. Exactly the kind of thing I want to do. These are based on Wyckoff criteria for supply/demand. Some days represent "meeting supply", some "lessening of demand", some "strong demand" and so on. The reason this is interesting to a "Wyckoffian" is because, say in an uptrend previous to a pullback, you have several "lessening of demand" days as opposed to lots "meeting supply" days. This would certainly be more positive in respect to the uptrend resuming after the pullback. So even though there are 16 different days, there aren't that many interpretations. I'm still trying to work them out. I could also use symbols for some days. Normally there aren't that many Strong Supply or Strong Demand days, so I could just have a symbol and a neutral color for both of those, since you'd want those to stand out anyway. Thanks, Harold OK, I need some help writing these formulas. I'd appreciate if someone could check these and see if they're right. I think they're pretty close. C > ((H-L)*.5) - Today's close in the upper half of bar range (H-L) > ((H, -1) - (L, -1)) - Today's bar range greater than yesterday's bar range C > (C,-1) - Close higher than yesterday V > (V,-1) - Volume higher than yesterday Thanks very much, Harold C > ((H-L)*.5) - Today's close in the upper half of bar range (H-L) > ((H, -1) - (L, -1)) - Today's bar range greater than yesterday's bar range C > (C,-1) - Close higher than yesterday V > (V,-1) - Volume higher than yesterday Should read (C>((H-L)*.5)+L) and ((H-L)>Ref((H-L),-1)) and C>Ref(C,-1) and V>Ref(V,-1) Up Day - Positive Close Up day, higher vol, narrow spread, close higher half. Up day, lower vol, narrow spread, close higher half. Up day, higher vol, wider spread, close higher half. Up day, lower vol, wider spread, close higher half. Up Day - Negative Close Up day, higher vol, narrow spread, close lower half. Up day, lower vol, narrow spread, close lower half. Up day, higher vol, wider spread, close lower half. Up day, lower vol, wider spread, close lower half. Down Day - Positive Close Down day, higher vol, narrow spread, close higher half. Down day, lower vol, narrow spread, close higher half. Down day, higher vol, wider spread, close higher half. Down day, lower vol, wider spread, close higher half. Down Day - Negative Close Down day, higher vol, narrow spread, close lower half. Down day, lower vol, narrow spread, close lower half. Down day, higher vol, wider spread, close lower half. Down day, lower vol, wider spread, close lower half. Your Spread and volume are irrelevent to your explorations, what is relevent is if the close was greater or less than yesterday and it's position to the days range so your formula would be UDPC:=(C>((H-L)*.5)+L) and C>Ref(C,-1); UDNC:=(C<((H-L)*.5)+L) and C>Ref(C,-1); DDPC:=(C>((H-L)*.5)+L) and C<Ref(C,-1); DDNC:=(C<((H-L)*.5)+L) and C<Ref(C,-1); UDPC:=(C>((H-L)*.5)+L) and C>Ref(C,-1); UDNC:=(C<((H-L)*.5)+L) and C>Ref(C,-1); DDPC:=(C>((H-L)*.5)+L) and C<Ref(C,-1); DDNC:=(C<((H-L)*.5)+L) and C<Ref(C,-1); HVWS:=((H-L)>Ref((H-L),-1)) and V>Ref(V,-1); HVNS:=((H-L)<Ref((H-L),-1)) and V>Ref(V,-1); LVWS:=((H-L)>Ref((H-L),-1)) and V<Ref(V,-1); LVNS:=((H-L)<Ref((H-L),-1)) and V<Ref(V,-1); UDPC:=(C>((H-L)*.5)+L) and C>Ref(C,-1); UDNC:=(C<((H-L)*.5)+L) and C>Ref(C,-1); DDPC:=(C>((H-L)*.5)+L) and C<Ref(C,-1); DDNC:=(C<((H-L)*.5)+L) and C<Ref(C,-1); HVWS:=((H-L)>Ref((H-L),-1)) and V>Ref(V,-1); HVNS:=((H-L)<Ref((H-L),-1)) and V>Ref(V,-1); LVWS:=((H-L)>Ref((H-L),-1)) and V<Ref(V,-1); LVNS:=((H-L)<Ref((H-L),-1)) and V<Ref(V,-1); B:=If(UDPC and HVWS,8,If(UDPC and HVNS,7,If(UDPC and LVWS,6,If(UDPC and LVNS,5, If(UDNC and HVWS,4,If(UDNC and HVNS,3,If(UDNC and LVWS,2,If(UDNC and LVNS,1, If(DDPC and HVWS,-8,If(DDPC and HVNS,-7,If(DDPC and LVWS,-6,If(DDPC and LVNS,-5, If(DDNC and HVWS,-4,If(DDNC and HVNS,-3,If(DDNC and LVWS,-2,If(DDNC and LVNS,-1,0)))))))))))))))); B I forgot to mention,plot this as a histogram in it's own window Crimey Henry! Outstanding! You're a genius. Thanks very much for all your trouble. Here's a link to a Yahoo chart of ADBE: So, is this anything? Can it help me w/short-term plays? I have no idea yet. Just looking at this one chart: During the first circled period in Sept/Oct the indicator plummets sharply, showing much effort to the downside but very little result as price goes sideways. Immediatedly following price takes off. In the second circled period in Dec the same thing happens. Following this price continues sideways while the indicator is in a slow steady uptrend. Does this give a clue as to accumulation? Note however the last several weeks the indicator shows great effort while price shows little result. Distribution? I don't know yet. But this is interesting, I gotta say. Thanks again, Harold Henry, or anyone, Forgive me 'cause I'm not exactly sure what I'm asking here. This histogram, and therefore the ma's, can only go up or down so far, 8 is the limit. So theoretically if you had a whole bunch of -8 days in a row, the ma's would flat line. Is there way this could be written so that limit doesn't exist? So that a bunch of -8 days could go to a lower level, -20 or whatever. This would be better for reading lower lows, higher lows, ect, correct? Sorry if this is a dumb question. Harold Harold, I think that you would have to rank the type of bars from most bullish to most bearish Harold, Once you rank the bars from most bullish to most bearish,then plot the difference from yesterday to today B-Ref(B,-1) My Final version |
|
Wyckoff PV Histogram I UDPC:=(C>=((H-L)*.5)+L) AND C>Ref(C,-1); UDNC:=(C<((H-L)*.5)+L) AND C>Ref(C,-1); DDPC:=(C>=((H-L)*.5)+L) AND C<Ref(C,-1); DDNC:=(C<((H-L)*.5)+L) AND C<Ref(C,-1); HVWS:=((H-L)>=Ref((H-L),-1)) AND V>Ref(V,-1); HVNS:=((H-L)<Ref((H-L),-1)) AND V>Ref(V,-1); LVWS:=((H-L)>=Ref((H-L),-1)) AND V<Ref(V,-1); LVNS:=((H-L)<Ref((H-L),-1)) AND V<Ref(V,-1); B:=If(UDPC AND LVWS,8, If(UDPC AND HVWS,7, If(UDPC AND HVNS,6, If(UDPC AND LVNS,5, If(UDNC AND LVWS,4, If(UDNC AND HVWS,3, If(UDNC AND HVNS,2, If(UDNC AND LVNS,1, If(DDPC AND LVNS,-1, If(DDPC AND HVNS,-2, If(DDPC AND HVWS,-3, If(DDPC AND LVWS,-4, If(DDNC AND LVNS,-5, If(DDNC AND HVNS,-6, If(DDNC AND HVWS,-7, If(DDNC AND LVWS,-8,0)))))))))))))))); Mov(B,5,S);Mov(B,15,S);B-Ref(B,-1); |
Henry, Thanks again for your latest. I've experiment w/both and I think I like the first one better. However, I have a problem, I'd be interested if my solution is viable. As I said, I'm using the first formula. I then overlay a 21d sma of the histogram. The problem is this: If I choose scaling for the sma to "Overlay w/o scale", then when I zoom in and out on the chart the sma moves around. If I choose "Merge w/scale on right" then the sma does not flucuate enough to be useful. So I chose "Display new scale on left", then set the min/max levels of that Y-axis to fixed levels, so the sma does not fluctuate. It will sometimes go "off the chart" a bit but not often. Does that sound like a proper solution? Thanks again, Harold Harold, Mov(B,21,S); should fix what you want. just paste it under the formula Harold: if the indefatigable Henry devotes that much time & effort to your formula, it's got to have something going for it. I have done the histogram for the 6/4/12:43 version, having made only one change from the formula. where you specify V>ref(V,-1); I took the liberty to make that (V>Ref(V,-1) OR V>Ref(V,-2)); a very minor change. came to the following conclusion: Attention when 3 consec increasing bars or 3rd bar larger than middle bar (has to be consecutive)! examples using the chart for ANZ 30/5/02, 8/8, 22/8, 17/10, 24/10, 22/11 all conform with that requirement but not 6/1/03 because the 3 bars are equally high. it will not work to the exclusion of all else. example the recent strong rise in ANZ would not have been signalled at all, not by my superficial interpretation anyway. an interesting study & I shall be reading your & Henry's text more carefully, maybe I have overlooked something relevant. amory Hi Tony, Sorry it took so long to get back to you. I thought this thread was dead. Here's the current formula I'm using, I forget which version of Henry's it is. You'll have to look thru the messages to see how I tweaked it to get the histogram and such. Harold |
|
Wyckoff PV Histogram II UDPC:=(C>((H-L)*.5)+L) AND C>Ref(C,-1); UDNC:=(C<((H-L)*.5)+L) AND C>Ref(C,-1); DDPC:=(C>((H-L)*.5)+L) AND C<Ref(C,-1); DDNC:=(C<((H-L)*.5)+L) AND C<Ref(C,-1); HVWS:=((H-L)>Ref((H-L),-1)) AND V>Ref(V,-1); HVNS:=((H-L)<Ref((H-L),-1)) AND V>Ref(V,-1); LVWS:=((H-L)>Ref((H-L),-1)) AND V<Ref(V,-1); LVNS:=((H-L)<Ref((H-L),-1)) AND V<Ref(V,-1); B:= If(UDPC AND HVWS,8, If(UDPC AND HVNS,7, If(UDPC AND LVWS,6, If(UDPC AND LVNS,5, If(UDNC AND HVWS,4, If(UDNC AND HVNS,3, If(UDNC AND LVWS,2, If(UDNC AND LVNS,1, If(DDNC AND HVWS,-8, If(DDNC AND HVNS,-7, If(DDNC AND LVWS,-6, If(DDNC AND LVNS,-5, If(DDPC AND HVWS,-4, If(DDPC AND HVNS,-3, If(DDPC AND LVWS,-2, If(DDPC AND LVNS,-1,0))))))))))))))));B |
�
