 Gold Customer
Joined: 11/11/2006
Posts: 359
|
Tom: Have you had any success with a PCF, et al, for the VPCI indicator?
My gadget is not Block-able and I haven't a clue for a PCF.
Seems like a usefull tool for the toolbox, however. If you have any success in this area, I would be appreciative of any assistance.
Mammon
|
|
Registered User
Joined: 3/21/2006
Posts: 4,308
|
Mammon, I am trying to construct the indicator from scratch. I prefer indicators over writing a PCF.
The Volume Price confirmation indicator (VPCI) exposes the relationship between the prevailing price trend and Volume. Some experts state, (and I am starting to agree with them) that Volume and price are all that there is to it. Of course trend is important as well.
Lord knows that I have back-tested many a technical indicator. And although some are accurate (to a degree) the truth is that no indicator can predict future price movment. Stocks are like a pitbull, you think that you can predict their behavior and the next thing you know they turn on you. I am sure that I will catch some flack for saying this (in light of the major thread going on currently involving MACD-H). Hohandy and I are in disagreement on the predictive value of this indicator.
Since all of my current holdings are wraped up in long term, I have a lot more time to devote to research. I have studied for the last 5-6 months the relationship between price and their moving averages. Now I will study the relationship between Price-trend and Volume. The secret to a greater understanding I believe can be found there..
If you are interested then I will keep you updated...
|
|
Registered User
Joined: 12/21/2004
Posts: 902
|
QUOTE (tommy21) I am sure that I will catch some flack for saying this (in light of the major thread going on currently involving MACD-H). Hohandy and I are in disagreement on the predictive value of this indicator.
LOL - nothing's guaranteed Tommy - all it is is just a means to hopefully increase the odds.
And I don't know if you missed my post last week or not, but check out CNH on a weekly or monthly going back to before 2006 - the thing that impresses me most about that stock is volume
|
|
Gold Customer
Joined: 11/11/2006
Posts: 359
|
Tom: I am interested. Almost everything , indicator wise, derives from price and/or volume. I still think of Dan Zanger, who used ONLY price and volume with chart patterns to amass a fortune. And thats in recent history. He was profiled twice in Stocks and Commodities.
I think the trouble arrives when one predicts movement rather than anticipating it.
I have followed your comments in this vein closely, in addition to my own floundering around in the research areas.
I don't think there are "right and "wrong" approaches. It appears that each method has it's followers who will defend them against all others. As long as it works for them, its working. But perhaps not for everyone...
Thank you,
Mammon
|
|
Registered User
Joined: 3/21/2006
Posts: 4,308
|
First off Hohandy I meant no harm with my remarks. I enjoy the friendly debates between you and I; it helps us to look at investing from different points of view and keeps our skills sharp. One last thing to mention, I am going through changes within the context of stock investing. My thoughts and theories are changing, so if I appear to be contrary to my prier trading techniques you will hopefully understand why.
Lets look at CNH and the volume pattern. On a weekly chart plot an 18-week linear regression line on price and then one on volume. Notice how the visual slope for the price is pretty steep and yet the slope on volume is polar opposite. Granted since the beginning of 2006 the Volume for this stock has been a big change compared to 2005, and price has confirmed that by moving from $17 to $50, but to continue up price is going to need more volume not less. If I were analyzing this security I would say that the ride is almost over (IMO). There might be another long consolidation or even a correction in store. Maybe even another huge burst in volume to push price higher. All I am saying is that there is no indicator that will give us that answer. Finally it is not all about just the volume of shares traded but the relationship of that compared to the range of the price (High, low, open close) during that trading period, and then compares simlarly to the folowing trading periods.
Mammon, There are traders that profit highly from understanding the real reasons that stocks move. And they do not rely on the run of the mill indicators that we see here. If you have access to Stocks and Commodities magazine (July issue) look at page 20-30 there is a great article that discusses VPCI. Let me know what you think…
|
|
Registered User
Joined: 12/21/2004
Posts: 902
|
QUOTE (tommy21) Lets look at CNH and the volume pattern. On a weekly chart plot an 18-week linear regression line on price and then one on volume. Notice how the visual slope for the price is pretty steep and yet the slope on volume is polar opposite. Granted since the beginning of 2006 the Volume for this stock has been a big change compared to 2005, and price has confirmed that by moving from $17 to $50, but to continue up price is going to need more volume not less. If I were analyzing this security I would say that the ride is almost over (IMO). There might be another long consolidation or even a correction in store. Maybe even another huge burst in volume to push price higher. All I am saying is that there is no indicator that will give us that answer. Finally it is not all about just the volume of shares traded but the relationship of that compared to the range of the price (High, low, open close) during that trading period, and then compares simlarly to the folowing trading periods.
I haven't given much rigorous thought yet to the price/volume relationship, except in broad general terms, so I expect that I will have a lot to learn from you in the next few months are you proceed with your study. I'm looking forward to it.
But in broad general terms, what comes to mind, is of course supply and demand - the more demand on the buying side, the higher the price will go, etc. But the economist part of me wonders if there necessarily has to be a direct correlation between rate of increase in volume and rate of increase in price for the lr comparisons to work. If the float is only X amount, the average daily volume can only go so high- at some point there can't be a 1:1 correlation of stock price and volume increase because there's not enough stock available to increase volume, but that doesn't mean that stock price can't keep going - in fact I would think the closer you get to the float cap, the more affect additional amounts of higher volume would have on price rise as buyers bid up those additional shares (or looking at it another way, buyers would have to bid up in order to coax those additional shares back onto the market).
My gut tells me that a simple slope comparison of price and volume growth isn't necessarily determinative. Does that make sense to you or do you that I'm just missing it?
Maybe the answer lies in volume/price slope comparisons of the beginning of the volume increase period and now (or something along that line) to see how the relationship of volume increase to price increase has changed over time or something like that. It's very plausible to me that as volume increases, unless the float increases with it, that the slope of the demand curve will get steeper meaning less additional volume is needed to generate a price increase. Not even mentioning exogenous factors such as the gangbuster strength of the ag equipment sector right now. I don't really know - just thinking in general terms here.
Always enjoy kicking things around with you Tommy.
|
|
Registered User
Joined: 1/28/2005
Posts: 6,049
|
I hope this is correct.
(I only glanced at the article)
ST= short -term
LT= long term
VWMA= volume weighted moving average.
Since some of these indicators are interesting by themselves.
I broke it up so it can be seen how they are calculated.
I choose ST=5 and LT=20
-------------------------------------------------------------------------------
ST VWMA (5):
(C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4) / (AVGV5 * 5)
--------------------------------------------------------------------------------
LT VWMA (20):
(C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9 + C10 * V10 + C11 * V11 + C12 * V12 + C13 * V13 + C14 * V14 + C15 * V15 + C16 * V16 + C17 * V17 + C18 * V18 + C19 * V19) / (AVGV20 * 20)
---------------------------------------------------------------------------------------
VPC = ( ST VWMA - LT VWMA )
((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4) / (AVGV5 * 5) - ((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9 + C10 * V10 + C11 * V11 + C12 * V12 + C13 * V13 + C14 * V14 + C15 * V15 + C16 * V16 + C17 * V17 + C18 * V18 + C19 * V19) / (AVGV20 * 20)))
-----------------------------------------------------------------------------------
VPR = (ST VWMA / ST SMA )
((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4) / (AVGV5 * 5) / AVGC5)
------------------------------------------------------------------------------------
VM = (AVGV5 / AVGV20)
--------------------------------------------------------------------------------------
VPCI = VPC * VPR * VM
VPCI
((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4) / (AVGV5 * 5) - ((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9 + C10 * V10 + C11 * V11 + C12 * V12 + C13 * V13 + C14 * V14 + C15 * V15 + C16 * V16 + C17 * V17 + C18 * V18 + C19 * V19) / (AVGV20 * 20))) * ((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4) / (AVGV5 * 5) / AVGC5) * (AVGV5 / AVGV20)
This should be plotted with center zero line checked.
You can click on the indicator and add a 20 period mav.
Thanks
diceman
|
|
Gold Customer
Joined: 11/11/2006
Posts: 359
|
Diceman:
I thank you for your efforts on my behalf.
Mammon
|
|
Registered User
Joined: 1/28/2005
Posts: 6,049
|
It appears there was a mistake.
VPC should be:
VPC = ( LT VWMA - LT SMA )
VPC=
((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9 + C10 * V10 + C11 * V11 + C12 * V12 + C13 * V13 + C14 * V14 + C15 * V15 + C16 * V16 + C17 * V17 + C18 * V18 + C19 * V19) / (AVGV20 * 20) - AVGC20)
Which makes VPCI:
((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9 + C10 * V10 + C11 * V11 + C12 * V12 + C13 * V13 + C14 * V14 + C15 * V15 + C16 * V16 + C17 * V17 + C18 * V18 + C19 * V19) / (AVGV20 * 20) - AVGC20) * ((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4) / (AVGV5 * 5) / AVGC5) * (AVGV5 / AVGV20)
Thanks
diceman
|
|
Registered User
Joined: 1/28/2005
Posts: 6,049
|
5 / 20 appears better on weekly charts.
This is the 10 50 version
VPCI ST=10 LT=50
(((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9 + C10 * V10 + C11 * V11 + C12 * V12 + C13 * V13 + C14 * V14 + C15 * V15 + C16 * V16 + C17 * V17 + C18 * V18 + C19 * V19 + C20 * V20 + C21 * V21 + C22 * V22 + C23 * V23 + C24 * V24 + C25 * V25 + C26 * V26 + C27 * V27 + C28 * V28 + C29 * V29 + C30 * V30 + C31 * V31 + C32 * V32 + C33 * V33 + C34 * V34 + C35 * V35 + C36 * V36 + C37 * V37 + C38 * V38 + C39 * V39 + C40 * V40 + C41 * V41 + C42 * V42 + C43 * V43 + C44 * V44 + C45 * V45 + C46 * V46 + C47 * V47 + C48 * V48 + C49 * V49) / (AVGV50 * 50)) - AVGC50) * (((C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9) / (AVGV10 * 10)) / AVGC10) * (AVGV10 / AVGV50)
Thanks
diceman
|
|
Registered User
Joined: 3/21/2006
Posts: 4,308
|
Diceman thanks for constructing the indicator so quickly (you beat the wordens to it) and me.
That is my weakness, I am great at coming up with good ideas but not as swift about constructing
indicators to make my job easier. I hope you stick around until I retire. I am trying to speed that
process up by following your advice, always look for new and better ways, never stop learning.
I believe that finding chart patterns with obvious signs of accumulation (as stated in my Bonanza bottom thread)
and waiting for the right volume pattern to emerge for the break-out and initial buy-in is the way to enter a stock.
Then you need to monitor your progress by means of the multiple moving averages (as stated in my FAN thread). The ones
that build there fortunes and have found success, (IMO) have done so using similar techniques that I am trying to build here.
Short term trading is fun (believe me I know) but it is not the most efficient way to build wealth. (IMO) you have to find
stocks that demonstrate proven patterns of success and buy and hold them for at least an intermediat time frame.
One question about the indicator VPCI take a look at MODT notice that from February 20 to March 06 the indicator is shooting up
when price and volume are going down. I know that you cannot speak for the ones that created the idea for VPCI but I would like
to hear your comments about this supposed contradiction.
Thanks Diceman for all your help..
Tommy21..
|
|
Registered User
Joined: 1/28/2005
Posts: 6,049
|
I would need to do a more detailed study.
My first impression is the "core" of the indicator:
(LT VWMA - LT SMA)
Is "adjusted" by a short-term ratio:
(ST VWMA / ST SMA)
and by a volume ratio:
(ST AVGV / LT AVG V )
Lets say the indicator is -10
When you multiply it by a number larger than 1
it becomes more negative.
(a number less than 1 more positive)
-10 * 1 = -10
-10 * 1.2 = -12
-10 * .8 = -8
-------------------------------------------------------------
So it is saying. there is a decrease in selling strength.
(even though the trends are still negative. The ratios
are smaller)
It is in effect saying selling pressure is drying up.
-----------------------------------------------------------------
(good eye for spotting that)
Thanks
diceman
|
|
Registered User
Joined: 3/21/2006
Posts: 4,308
|
I like the VPC portion of this Indicator for a 10 day period. Try this indicator
in the middle window, Indicator - indicator use center zero line -
(AVGC10 - (C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9) / (AVGV10 * 10))
On price in the top window add a 10 day sma, also add this indicator in the top
window, indicator, indicator check plot using price scale -
(C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9) / (AVGV10 * 10))
In the top window you now have a 10 day moving average and a 10 day volume weighted
moving average.The VPC indicator in the middle window measures the difference between
the two moving averages in the top window. When the VPC indicator crosses down below
the zero line from above zero, it is a buy signal. This indicator is a little more
difficult to interpret then the usual indicator, so do a lot of back - testing to get
used to its nuances. I know that I will...
|
|
Registered User
Joined: 3/21/2006
Posts: 4,308
|
I am sorry. The VPC indicator will be easier to read if you use it this way instead -
(C * V + C1 * V1 + C2 * V2 + C3 * V3 + C4 * V4 + C5 * V5 + C6 * V6 + C7 * V7 + C8 * V8 + C9 * V9) / (AVGV10 * 10) - AVGC10
Still use the center zero line and the buy signal will now come when the indicator
crosses up over the zero line. Still do some back testing thought, to make sure that
you are convinced of the indicators reliability...
|
|
Gold Customer
Joined: 11/11/2006
Posts: 359
|
Tom: When I paste the formula in the top window (C*V+C1...)plotted with price scale, I get no
indicator line,as I do with the 10 day moving average. Plotting the other indicator in the middle window, I
get an single indicator line that crosses "O"(AVGC10...).This indicator seems to be inverted.
Where have I fouled up...
Thank you.
Mammon
|
|
Gold Customer
Joined: 11/11/2006
Posts: 359
|
Tom: Ignore my tastless comment about the middle window.
Mammon
|
|
Registered User
Joined: 3/21/2006
Posts: 4,308
|
Mammon, have you figured it out yet? I am still working on the nuances of the indicator.
You have to watch the volume as the indicator crosses zero. for instance when the indicator spikes high
for one day, then it is usually a false start (like a one day large white candle that corrects itself
the next day or two). if the indicator crosses zero gradually then price performs better and will continue
climbing. This is part of my learning curve concerning volume. I am currently looking at stocks that have
been running up for over two years (I have traced them back to the bottom formations) and I am studing the
Volume patterns, like I did a while ago with moving averages.
Let me know if you still need help seting this up..
|
|
Registered User
Joined: 1/19/2015
Posts: 1
|
Hi Diceman, its been a few years since your post and your explanation of VPCI has really helped me.
I've read Buff's 2011 "Investing w/ Volume Analysis" book at least 3 times, looking for hints in other parts of the book, but the calculation of his Anti-Volume Stop Loss still eludes me. On page 254 Buff says
AVSL = Lower Bollinger Band - (Price, Length, Standard Deviation)
where
Length = Round (3+VPCI)
Price = Average (Lows x 1/ VPC x 1 / VPR, Length)
Standard Deviation = 2 x (VPCI X VM)
It looks like you use a rounded up whole number "Length" to calculate the average. The part I dont understand is once you have the "Standard Deviation" and the "Price", how is it subtracted from the Lower BB? I mean -(Price, Length, STD) isnt really a correct mathematical expression.
Is there something obvious that you can see?
Thanks,
Mike
|
|
Registered User
Joined: 1/28/2005
Posts: 6,049
|
QUOTE (MHN)
Hi Diceman, its been a few years since your post and your explanation of VPCI has really helped me.
I've read Buff's 2011 "Investing w/ Volume Analysis" book at least 3 times, looking for hints in other parts of the book, but the calculation of his Anti-Volume Stop Loss still eludes me. On page 254 Buff says
AVSL = Lower Bollinger Band - (Price, Length, Standard Deviation)
where
Length = Round (3+VPCI)
Price = Average (Lows x 1/ VPC x 1 / VPR, Length)
Standard Deviation = 2 x (VPCI X VM)
It looks like you use a rounded up whole number "Length" to calculate the average. The part I dont understand is once you have the "Standard Deviation" and the "Price", how is it subtracted from the Lower BB? I mean -(Price, Length, STD) isnt really a correct mathematical expression.
Is there something obvious that you can see?
Thanks,
Mike
Its difficult say from that.
It almost looks like a software language.
Standard Deviation looks like the only clear equation.
He doesnt show any examples?
Obviously assuming you could calculate the parts
you can guess at things like lower BB-(Price+Length+STD) but I couldnt say if its correct.
You could also look at the logic.
Does (price length STD) increase when he wants a widening stop?
Without the book his intent and the calculations are unclear.
Thanks
|
|
|
Guest-1 |