BlowFish

If date = date[1] And StartPrice > 0 Then Begin Value2 = AvgPrice - StartPrice; V2VolLevel = 5000+(Value2*(1/(minmove/pricescale))); PVPVolArray[V2VolLevel] = PVPVolArray[V2VolLevel] + MyVolume; [i] {Plot new volume level here volume is PVPVolArray[V2VolLevel] the level to plot at can be got from V2VolLevel}[/i] If PVPVolArray[V2VolLevel] > PVPVolume Then Begin PVPVolume = PVPVolArray[V2VolLevel]; PriceDiff = 5000-V2VolLevel; PVPPrice = StartPrice - (PriceDiff*(minmove/pricescale)); [i] {If we are here then we have a new peak volume, you might need to rescale the chart unless you used a 'fixed scaling'} [/i] End; End;

As Tams points out the array index itself is used to derive the price. The days starting price starts in the middle of the array. V2VolLevel = 5000+(Value2*(1/(minmove/pricescale))); converts a tick to be an integer. So whether it is the ES (.25 tick) or DAX (.5 tick) this will convert it do an array index. (Don't ask me how I struggled with that!) Edit: Another way of looking at this is V2VolLevel is an integer representing Where the tick is in the array compared to the opening tick at 5000. It's not a very well chosen name to be honest. Im not sure I have explained it well enough?

Input: SD3(false), PlotOldVWAP (false), NumberofDays (1); if date > date[1] then begin n = n + 1; if n = NumberofDays then begin s = 0; n = 0; VWAP1 = VWAP; SD1 = SD; VWAP = 0; SumWeights = 0; SumWeightsOld = 0; VWAPOld = 0; end; I use something like this. If it's a new day increment a counter. If the counter is up to n days fire your reset code. You can tack that on to the front of the VWAP code in the indicator section here. Should probably change it to decrement a counter and compare to zero, more efficient like that, not sure what I was thinking that day .

You have picked a good place to start with that code. Mind you I would say that I re-wrote it and sent it to Dbtina! The way to do things is construct an array of volume and use a tick's price to index it (after suitable scaling so each change in price is '1'). That is exactly what I wrote and runs efficiently enough to process every tick. All the previous code I have seen uses messy, slow and inefficient loops. My intention was always to expand it to plot a volume profile though never got round to it. One important idea. Again do not use loops to plot just add volume at the appropriate level when a tick arrives. If that happens to be at the mode (PVP, peak, wahtever you call it) You may need to loop through the whole array just for that tick to rescale things to the new peak. This depends on how you choose to scale in the first place and is avoidable if you use a fixed scale of n volume per x axis unit. I have been away from TL for quite a while and have a lot of catching up to do, Once I have I'll take a proper look at what you are up to

Well there is a whole other issue with forex and that is whether using bid and ask chanes as a proxy is good enough for PVP, I don't think so. It can vary tremendously depending on how you allocate the volume to a level. Looking at every single print and allocating it to the apropriate level gives quite different results to averaging across a 2 min bar. Actually it's debatable (that's being generous) whether FX data is 'OK' for VWAP and SD bands. If the lines it gives you are a useful framework who cares.

HI Aki, It al started here Algorithms for calculating variance - Wikipedia, the free encyclopedia seacthing round the internet revealed a couple of online methods. I'd really like to do the skew in a non iterative fashion but my maths is nowhere near strong enough! I have a couple of 'issues' with PVP as a proxy to determine skew that I have mentioned in the threads.

Not only that there are brokers that will run the popular ones for you so all you have to do is read your satements each month. The thing is with 100% mechanical systems is there is still discretion to be applied. e.g. When to retire them, when to change parameters, how much to risk, which markets to run them on etc. etc.

Metatrader 4 speak for "Expert Advisor" I believe.

Absolutely. Interesting post. I was in two minds whether to say anything but I figured it might have some value to people that are following your work (and that is quite a few if the PM's I have had about code are anything to go by). Certainly it wasn't just to be argumentative and I apologise if it seemed overly critical. Some people perhaps don't have a good idea of 'scale' and perhaps it might be helpful to them (particularly if trying to come up with their own indicators, that level of precision is not necessary and probably unattainable in windows on PC hardware). For example when an app requests data from disk it can take 10 or even 20 milliseconds before the first byte is delivered. DSL local loop delay can be 30 ms. Light will get about 300km in a vaccum in a bit over a millisecond.

I don't. It's looking good though!

I think it was your very post here that I quizzed you about your infrastructure. I thought you where using Tradestation data? Or are you getting a data feed that is time stamped at the exchange? I am going to go out on a limb here and call hyperbole. That's not to say that this is not a pretty interesting post but as I've said a million times (hehe) milisecond accuracy isn't plausible. Transactional latency on the CME systems is likely to be up to an order of magnitude greater than milliseconds. I hope I am not sounding pedantic but it might shape other peoples expectations.

Ahh the good ole 'scientific method' many and varied applications

Thanks for your detailed reply....seemed pretty clear to me

Indeed, the question Is how much of a turn do you need to see to confirm a change in the force .....err I mean flow (old starwars habits die hard).

Could be a variety of things. I am having similar issues my self and its driving me potty. Changing the router fixed one particular problem (intermittent packet loss) however the line still looses sync. Routers do fail especially if they are somewhere without good airflow. I have had the ISP run all sorts of checks and the telco engineers have been to my house twice to run tests on the line (there test gear is quite sophisticated now and should be capable of spotting intermittent problems). As I have a rough idea what the figures for signal to noise ratio and line attenuation should be (and the router shows these) I can keep telling them that things are un acceptable Despite this the problem remains and the fault can not be found. I am just about to call them to get them to put me on a completely different port on the DSLAM at the exchange. If that still has problems I will ask the telco to use a different cable pair to the house. If I where you I would get the ISP to run checks first and be prepared to swap the router (simply to eliminate it as the issue).

HI FT, What constitutes an 'order flow transition' ? In the second example for instance I notice 3 green bars in the last down leg that did not result in a change in price direction in price or order flow. Put another way if watching those 3 events live it would look as if the order flow had changed (indeed it did) however the initial flow resumed. How much paper do you need to see to quantify it as a transition? Cheers.

HI TT, I seem to recall that you mentioned early in this thread (or maybe another) that while your approach for stocks and the ES (and FX for that matter) is essentially the same across the board, there where some subtle differences. Looking at your recent ES trades I am struggling to see much difference at all The most obvious is I see you using some 15 min charts on ES. I wonder if you might perhaps highlight the more important differences. Do you find the ES to suffer from more false break outs? It seems Index futures are particularly prone to those tests pokes and prods that go a tick or two beyond a range and then shoot off he other way.

The short answer is you don't As there is no centralised exchange different participants can be conducting trades at different prices. Of course if things get out of line arbitrage opportunitiues would exist so things would be brought into line fairly quickly. I see other people have already said this but as I have typed it already I may as well post.

Short answer is no Silly.

Isn't that just a MACD with XMA's instead of SMA's?

All implementations are pretty highly iterative though I managed (with some help from a mathematician) to come up with a continuous algorithm. as an aside as MT4 'brokers' don;t usually have complete volume info it is not an ideal platform anyway.

This is misleading lingo UB. I presume you are actually tracking the imbalance in aggresive/passive (or patient/impatient) buyers/sellers? It is a complete fallacy to assume that just because someone sells aggressively.that they are opening a short position. I am sure you know this, however you have a duty of care to your more naive readers! (actually you don't in the slightest but I am sure as a responsible poster you would not want them to go away with any misguided notions). For newer traders this sort of wrong thinking can cause no end of problems down the road. Talking of naivety i think it is naive to assume that when the commercials are accumulating or distributing a large position they do so impatiently. Now whether cumulative contracts aggressively bought or sold is a good indicator of short term price movement is not what I am questioning. what I am questioning is whether you can assume that this represents short interest. Of course if you have a way of getting real time open interest information from the exchange hats off to you!

It is not completely accurate none of the TS implementations are due to the data handling. TS handles best bid and best ask asynchronously. It may be possible by using 3 data Plots one assigned to the instruments bid one to the ask and one to the last. You would need to ask TS whether price change 'events' of bid and ask are handles sycronously they look like they might be if you simply watch them on a chart. On the merit of Fulcrums approach does it not assume that trader acumuluate by buying aggresively? Isn't this the opposite to what occurs? When accumulating a large position you would buy passively at the bottom of the range and sell aggressively at the top (to keep price down).

I think the core concepts are pretty good. However he uses a series of terms that despite being well defined I did not find that intuitive. It is also a somewhat 'difficult' look at price behaviour. If you want a better understanding of price behaviour it is likely to provide it. If you want a simple method of extracting a couple of points a day in your favourite market there might be better sources. With those caveats I would say go for it.

All V@B or V@A really shows is who is prepared to buy/sell aggressively/passively or put another way who is patient (limit orders) and who is anxious (market order). People seem to assume that anxious participants are the ones who know where price is going.