Trading with Market Statistics XI. HUP

[jperl 11]

The price around vwap is tough for me I want to buy at -2sd no matter where price is relative to VWAP and sell at +2sd. I am working on this. I covered my SPY at 90.50.

 

Two points:

Don't trade at the VWAP if VWAP is close to the PVP. Skew is usually small at that point, so market doesn't know which way it wants to go.

 

Fading the second SD is ok, provided price has already passed through it and is coming back toward the VWAP. For example, if price is rising toward +SD2, wait for it to pass through it and then comes back down through it a second time before fading it. You may miss a few good trades this way, but at least you won't be caught in a continued range expansion to the upside.

[ockrazor 10]

I am pretty sure price was under vwap when I faded -1sd so no was price was moving toward VWAP. vwap under POV and Price under VWAP so OK to fade. Is that right?

[jperl 11]

I am pretty sure price was under vwap when I faded -1sd so no was price was moving toward VWAP. vwap under POV and Price under VWAP so OK to fade. Is that right?

 

No. I would never fade the 1st SD (that is trade toward the VWAP). You are trading against the trend. Always dangerous at the 1st SD.

[darthtrader2.0 10]

Well hopefully I haven't bit the hand that...ect..

But, Jerry have you ever messed around with trying to compute the cash VWAP for the underlie of the derivative your trading?

From my research, I just have fundamental problems with the VWAP on futures. I really question with the depth of something like ES along with what the ES functions as in the market that the WVAP matters that much on a futures contracts. However, on a single security that makes up an index, big institutions have some serious liquidity problems so that then puts a natural regression towards the mean on any single security, a natural regression towards the VWAP that is probly not there on the equity futures or is at least highly distorted from hedging/arbitrage.

Basically, a tick precise, cash weighted agragate VWAP I suspect would be pretty baddass..

[jperl 11]

But, Jerry have you ever messed around with trying to compute the cash VWAP for the underlie of the derivative your trading?

The short answer is no.

From my research, I just have fundamental problems with the VWAP on futures.

Yes that's been obvious from the last bunch of posts you have made on these threads. I certainly will not dispute your contention that there may be some hidden variable that we should be examining(sort of like hidden variables in quantum mechanics) but until that something comes along and can be examined in detail, this is all we have.

When you find something better, let us all know about it.

[budfox99 10]

Is there a link to a summation of all of the rules in using VWAP/PVP/SD for JPERL's Market Statistics?

[BlowFish 10]

It is not presented as a fixed system but having said that there is more than enough information published to trade it as such.

 

If the 11 threads are daunting just read the first post in each and you will have all you need. The rest is questions and ongoing discussions. It is worth doing this as a simple newbie trade is first presented and then the ideas built on. You would be doing yourself a dis-service by not spending the 10 or 20 minutes to read that information at least.

 

I do have a summary (with some key observations) in a reporters notepad but I guess that's not much use to you

[budfox99 10]

Thanks Blowfish, I've read and watched all 11 parts of the thread several times. I've also made some notes after reading the replies to several questions posted. I was just curious if someone had posted all of this info in one place.

[BlowFish 10]

I have to say I quite like bullet points, dosen,t take long to go through and make a short list.

[budfox99 10]

I've been using BigCharts.com to get the 10 day PVP (Volume by Price). It proved to be a nice HUP for the SPY's today. The market bounced off the 10 day PVP $85 region twice.

 

Does anyone else use longer term HUP's?

[Anton 10]

Having only just found this series, I have as yet to read and digest the whole set. What I have done is try to prepare the material by collecting the various downloads, films and images first. In so doing, I realise that I have no way of knowing if I have all of those that I should have.

 

Would it be too much of an imposition for you, if I asked if you could simply do a check list of all of the films and any other item that I should have downloaded? Perhaps with a reference to which training lesson it relates. A list at least would be really helpful.

That way I can ascertain that I have everything that I should have at the outset.

 

Thank you in advance for what looks to me like a masterly training course.

[BlowFish 10]

Anton, As far as I know everything is where it should be if you follow it in sequence. I think at one time some videos required premium access if you have issues it may be to do with that. Just start at the beginning and work your way through.

[darthtrader3.0beta 10]

The short answer is no.

Yes that's been obvious from the last bunch of posts you have made on these threads. I certainly will not dispute your contention that there may be some hidden variable that we should be examining(sort of like hidden variables in quantum mechanics) but until that something comes along and can be examined in detail, this is all we have.

When you find something better, let us all know about it.

 

Well this might be interesting.

I've put trading on hold until ninja 7 comes out, cancelled my CME data fees and was just going to hack around on the historical data I have. I didn't realize until last week though that I get 10 minute delayed YM feeds still with real time NYSE TICK...I was just going to paper trade this morning off the tape on YM and mess around with order entries/stops but felt naked without the TICk. Then I realized I literally found the holly grail. I can say without question that if you could look at the NYSE TICK 10 minutes into the future then you have the holly grail....

Going off the TICK 10 minutes into the future using 10 contracts and a 15 point stop I pulled 135 wins and 2 losers out today...

The only thing to me this shows is its completely nuts to not look at the cash when trading futures(ie the underlie of the derivative your trading)

The problem I see with index futures analysis is that we are not taking into account the nature of the instrument...its ultimately an arbitrage and hedging vehicle, us 2 bit pirates are just in the way of what the elephants are trying to do. Reading the tape on YM works great right up the point the programs come in and sweep the market..then your either lucky or a victim if you only reading the tape.

If there is one thing we know though its that the elephants have to be aware of the VWAP on the stocks they are trading..thats their stomping ground as far as liquidity goes.

The hidden variable isn't really hidden, its just the cash vs futures.

An interesting question to me is if there is some way to build a composite measure of how the elephants are stomping around the wvaps on the underlie...that probly could be alot more interesting than TICK.

[darthtrader3.0beta 10]

For computation of mode, mean or median of the volume distribution function you need to keep an array with volume at each price.

Such an array needs to have exactly that number of elements as how many different prices are there in evaluated period (in this case in one day). So if tick size is 0.25 and current daily range is 10 you need to maintain an array of 41 elements. You dont need to care about concrete trades.

To calculate median of such an array you need to order its elements by size and find the middle one in this sequence. If there is an even number of them you can take any number between and including the middle two. Search Wikipedia for median.

 

This is exactly where the C# based trading platforms shine. A sorteddictionary in C# is a lovely way to do a hash table which is the data structure you want, not an array if your talking basically keeping track of counting..

As I've posted before, there is alot of "meat" to the ideas here but they are bogged down in the wrong use of very defined terms.

"volume distribution function"....none of these threads really talk about a probabillity distribution of volume, "volume distribution function" in the context of these threads means assuming a normal distribution then counting volume..there is a HUGE difference between that and a ""volume distribution function"..

Comeon, if I'm talking to the wall here I seriously give up and will just keep my thoughts private.

[BlowFish 10]

You are talking to the wall to be honest Darth Actually maybe you are the wall There is absolutely no assumption about a normal distribution. Actually the assumption is quite the reverse until later on in the presentation. It is actually assumed that the distribution will be skewed. I just don't see how you can walk away with the normal distro idea. Really I truly and genuinely don't understand how you don't 'get it' if you have read the early threads. You need to read it without the blinders mate 'cause frankly you are plain wrong. It's a real shame as Jerrys work would seem to be a pretty close fit with the sort of stuff that interests you.

[Head2k 10]

Darth, there is no assumption about the distribution. It is not assumed that it is normal, it is not assumed that it is skewed. No assumptions are needed.

 

Actually there is one point where you need to assume some distribution, so maybe that is what you are talking about. When you calculate volume distribution say for a day, you calculate it from bars. Since you dont see inside the bars you have to assume distribution of volume inside them. Usually the assumption is that volume is spread equally along every bar. That is of course a simplification and it introduces some error into calculation. To eleminate this error one would need to run the calculation on a 1 tick chart, where every bar (or point, in this case) means one transaction. But if you play with it for a while you will find that these errors are insignificat (say under 1 tick) when certain detail in data sampling is reached.

 

So as I said, this is the only point when you assume something. The rest is simply mathematics.

[VolumeJedi 10]

My version.

 

I do not want to distract from this wonderful thread, but I really wanted to post a pic. I understand this is not a VSA thread, but I must make one reference to it. Please bare with me.

 

Below is a 4 min Euro chart from today/last night (12/31/2008). As is the last day of the year, no trade was taken but it was "called" as it happened.

 

Note there are various HUPs on this chart. There is a major difference between my HUPs and Jerry's. All the solid lines represent dynamic HUPs used in a static fashion. That is, at 1700 hrs (the close as per my data feed) the values are taken and projected forward to the next trading day. Therefore, the VWAPd is the daily VWAP as of the close the previous day.

 

The two dotted lines are dynamic. The Aqua line is the VWAPc (C for current) and the Yellow is the PVPc. Note that on the chart, the PVPc is equal to the PVP from the prior day, so it may be a bit hard to see.

 

As for the trade. We were in a down trend. As defined by price trading below both the VWAPd and the VWAPc. There was a trade at the VWAPd, but that is in the VSA thread. Here we want to look at a "Market Statistics" trade at the fist standard deviation of VWAPd. Up until about 0612, the VWAPc was above the PVPc. Newbie was looking to make shorts due to the trend but did not yet have skew in his favor. Things changed at 0612. The VWAPc finally went below the PVPc. Price was near the SD-1. Time to look to get short.

 

According to the Shapiro Effect, we would want to look for an up bar that pass through the SD-1 and then a down bar that violates the low of that bar. In VSA terms, we get a two bar reversal, which also qualifies as the Shapiro Effect @ 0640. Short trade set up with a target of SD-2. Two bars and roughly 40 pips later, SD-2 is hit.

 

Jerry hasn't mentioned this yet, but we pass thru SD3d-1 (first standard deviation for a 3 day VWAP) and it may be possible to then target the second standard deviation of this WVAP rather than just SD-2. Am I correct on this Jerry? As it happens, price does move down to SD3d-2 and along the way moves thru SDw-1 (weekly) and ultimately stops at SDw-2.

 

Thank you for indulging me. Thanks Jerry for your considerable time and effort. They say when you teach you learn twice. I hope you have gotten at least half of what you have given me (us).

[snowbird 10]

VolumeJedi,

 

I am also interested in jperls reply, I imagine you can treat all these HUPs identically, and rather than targeting the additional 3 day SD2, just re-analyze price action at each HUP and make a decision as to hold or exit depending on how many contracts you have on. I also imagine if multiple HUP's were to align at the same spot, you may choose just to exit.

 

On a seperate note: It appears you may be one of the first to apply jperls strategies to forex.

 

I'm interested if anyone else has also experimented with pvp and vwap using market statistics with any of the currency pairs, and if so, what their experience has been.

 

Specifically, are there any tradestation users on the thread who have figured out the right code changes/array parameters neccesary to alter dbtinas original PVP & VWAP code to work on the currency pairs?

 

Happy New Year All... wishing you a great 2009!!!!

 

snowbird

[abboman 10]

Hi Jerry,

I've been through your threads a few times and I want to thank you for taking the time and effort to lay out your perspectives on trading.

I have a few questions that I'm hoping you might respond too.

 

1.) When you scale in do you use the Shapiro effect or do you pull the trigger when then price hits the scale in point.

 

2.) Could you give some thoughts on when you reverse a trade. When do you do it?

 

3.) I'm a recent ensign subscriber and I don't see where you can extend the HUP lines on the charts. Did you do that through a DYO?

 

4.) In an earlier thread you wrote about using hard stops in case you lose power or your equipment fails. How far out do you place these stops and do you enter them and cancel them every trade? Or are they sitting out there all day? Or something else?

 

As you can tell by the questions I am a newbie. Any help would be greatly appreciated.

 

Thank you for your generosity.

 

Mark

[federicoo 10]

hi, could you inform me how to setup the studies and template in ew to use all of these indicators that you are showing us?

 

Thanks in advance,

Fede

[jperl 11]

Hi Jerry,

1.) When you scale in do you use the Shapiro effect or do you pull the trigger when then price hits the scale in point.

 

Sorry for the delay in answering your query. I just noticed it.

As a newbie if you are using the Shapiro effect, you should use it on every entry. That includes scale ins as well. The whole idea behind the Shapiro effect is to give you confidence that the trade will move in your direction.

 

2.) Could you give some thoughts on when you reverse a trade. When do you do it?

I was going to discuss trade reversals in the Market Statistics threads, but never got around to it. At some point perhaps I will add a thread about this. Basically, if you are in a trade and the trade begins to fail, you have 3 choices: a) exit at some predefined stop loss b) wait for a scale in point or c)reverse the trade. I've already discussed a) and b). Here is a simple procedure to do c): Wait for a scale in point and apply the Shapiro effect. If the Shapiro effect does not materialize, reverse the trade and double up on the number of contracts you are trading. The whole idea here is you want the trade to be profitable and you want it to be so, quickly.

3.) I'm a recent ensign subscriber and I don't see where you can extend the HUP lines on the charts. Did you do that through a DYO?

You can extend HUP lines by assigning global variables to them and then use the price line graphic to extend them.

4.) In an earlier thread you wrote about using hard stops in case you lose power or your equipment fails. How far out do you place these stops and do you enter them and cancel them every trade? Or are they sitting out there all day? Or something else?

I use ZeroLine Trader for entries. ZLT automatically enters a stop loss with every trade entry and deletes them on every exit. The stop losses are placed far enough away so they don't get hit.

 

 

 

Ok Mark hope this helps

[BlowFish 10]

Glad to see you are still around Jerry. An idea for c) above. If whilst waiting for the Shapiro effect the other end of the trigger bar is broken, reverse.

[n123 10]

Thank you Jerry for introducing this interesting topic! I have read all your threads and watched all the videos. I have a fundamental question about HUP.

 

I am very curious to know why do HUPs work. Why are they support and resistance? For example, if SD1 is a support/resistance area, why not SD1.5 or SD 2.3? What is so special about them?

 

To me, support and resistance work because people around the world use them, like fibs... why do 50% or 61.8% retracement often rejects price? because many people watch them and use them. So the reason for them to work is self-fulfill prophecy.

 

I am sure they work well for you, but I am just curious why they work.

 

Thanks again!

[jperl 11]

 

I am very curious to know why do HUPs work. Why are they support and resistance? For example, if SD1 is a support/resistance area, why not SD1.5 or SD 2.3? What is so special about them?

 

 

It depends on what you mean by working, n123. The standard deviation points above and below the VWAP represent the volatility of the price data over the period for which you computed the SD, nothing more. It thus represents the amount you should expect the price to fluctuate. It's a quantitative measure of volatility.

How you use this information is of course up to you. I do not think of this as support/resistance. In fact I don't like that term since it suggests a point of reversal which in fact it is not. Rather I prefer to describe it as points where the market slows it's motion for a time before deciding what to do next. I thus call them HUP or hold up prices rather than support/resistance.

[n123 10]

It depends on what you mean by working, n123. The standard deviation points above and below the VWAP represent the volatility of the price data over the period for which you computed the SD, nothing more. It thus represents the amount you should expect the price to fluctuate. It's a quantitative measure of volatility.

How you use this information is of course up to you. I do not think of this as support/resistance. In fact I don't like that term since it suggests a point of reversal which in fact it is not. Rather I prefer to describe it as points where the market slows it's motion for a time before deciding what to do next. I thus call them HUP or hold up prices rather than support/resistance.

 

Thank you for your reply Jerry. Oh I should not confuse hold up price with support and resistance, sorry about that.

 

I understand that SD quantitatively measures the volatility. But I still wish to know why SD1 and SD2 are hold up prices but not SD1.5 or SD 1.3 or anything in between? Is there some discontinuity in the distribution in the SD1 and SD2 level so that when price goes there it tends to hold up or slow down? Basically, why those particular points?