Trading with Market Statistics VI. Scaling In and Risk Tolerance

[Dogpile 10]

which std dev are you talking about walter. I use dbtina's code as proxy for Jerrys std dev... I use my own code to watch the expansion/contraction of volatility in intraday basis.

 

btw, I have to thank Jerry. I have not had a losing day since I started filtering trades with VWAP (nearly 3 weeks now). part of this is of course these juicy markets. very forgiving market. bails you out of a bad trade with a monster swing at some other point.

[jperl 11]

 

Is plotting on a (2 min) intraday chart enough or does the ELD for pvp needs more functionality for the upcomming lessons ? (plotting weekly ? etc.)

 

thanks!

 

If you want to do more advanced trades (which I only recommend once you have passed the basic test, lol), you will want the volume histogram and its VWAP,SD's for 1 day,2 day,1 week, 2 week, 1 month, 2 month and 1 year. You will also want these lines to update dyanamically for the day you trade.

We will get to what all this means when we discuss Hold Up Prices ( HUP )

[walterw 10]

this one Dog... is it yours ?... thanks Walter

[Dogpile 10]

walter,

 

that is dbntinas code but I played with the # of Std Devs on there.. that one I think had 0.5 1.0 and 2.0 std devs... but I am playing around with those settings and was using something different today.

[walterw 10]

Ok Dog I see.... now this other, would that be diferent ? sorry for bothering you... thanks Walter.

[Dogpile 10]

<<Ok Dog I see.... now this other, would that be diferent ? sorry for bothering you... thanks Walter>>

 

yes that is different. that is the code I wrote to watch how the bands compress and widen -- not necessarily for levels. and you aren't ever bothering me, I am as addicted to chatting about this as I am about trading.

 

here it is:

 

{I use this on 1 and 2-minute charts with a 'custom session' start time to collect 45 2-minute bars -- thus it starts plotting this band using 60 minutes of pre-market trading and first 30-mins of regular trading}

 

value1=

square(c-vwap_h)+

square(c[1]-vwap_h)+

square(c[2]-vwap_h)+

square(c[3]-vwap_h)+

square(c[4]-vwap_h)+

square(c[5]-vwap_h)+

square(c[6]-vwap_h)+

square(c[7]-vwap_h)+

square(c[8]-vwap_h)+

square(c[9]-vwap_h)+

square(c[10]-vwap_h)+

square(c[11]-vwap_h)+

square(c[12]-vwap_h)+

square(c[13]-vwap_h)+

square(c[14]-vwap_h)+

square(c[15]-vwap_h)+

square(c[16]-vwap_h)+

square(c[17]-vwap_h)+

square(c[18]-vwap_h)+

square(c[19]-vwap_h)+

square(c[20]-vwap_h)+

square(c[21]-vwap_h)+

square(c[22]-vwap_h)+

square(c[23]-vwap_h)+

square(c[24]-vwap_h)+

square(c[25]-vwap_h)+

square(c[26]-vwap_h)+

square(c[27]-vwap_h)+

square(c[28]-vwap_h)+

square(c[29]-vwap_h)+

square(c[30]-vwap_h)+

square(c[31]-vwap_h)+

square(c[32]-vwap_h)+

square(c[33]-vwap_h)+

square(c[34]-vwap_h)+

square(c[35]-vwap_h)+

square(c[36]-vwap_h)+

square(c[37]-vwap_h)+

square(c[38]-vwap_h)+

square(c[39]-vwap_h)+

square(c[40]-vwap_h)+

square(c[41]-vwap_h)+

square(c[42]-vwap_h)+

square(c[43]-vwap_h)+

square(c[44]-vwap_h);

 

value2=squareroot(value1/(45-1));

value3=vwap_h+value2;

value4=vwap_h-value2;

value5=vwap_h+(value2*2);

value6=vwap_h-(value2*2);

 

if time>659 then begin

{note, I am in California timezone >659 = >959 in New York Time}

 

Plot1(vwap_h, "UpperBand" ) ;

{Plot2(value3, "UpperBand" ) ;

Plot3(value4, "UpperBand" ) ;}

Plot4(value5, "UpperBand" ) ;

Plot5(value6, "UpperBand" ) ;

 

end;

[walterw 10]

Allright Dogpile ¡¡ nice to have the same adictions jejeje... but that is a nice formula too ¡¡ I like to see this compression/expansion being more amplified visually with tis one.... thanks again ¡¡ cheers Walter.

[Dogpile 10]

btw, as rumor-filled as the market was today -- look how 'normal' the volume distribution ended up...

[BlowFish 10]

Hi there I can't log in to TS forums (no account). The one I have has is by ghkramer. It has numerous parameters the most important is probably delta price which should be the tick size or larger if you try larger you may get more success because of how the array is indexed.

 

It takes a little while to first display as it has quite a lot of stuff to do. The rest can be left on defaults. Every now and then it gives an array subscript error but oddly re-applying the indicator fixes it. I havent looked in to why yet though have taken the time to familiarise myself with the code.

 

I'd very much like to shoot ghkramer an email a)thanking him for his great work and b) asking for the latest version. Anyone know how I canget in touch (just to say again I don't have a TS forum logon).

 

Cheers.

[jperl 11]

btw, as rumor-filled as the market was today -- look how 'normal' the volume distribution ended up...

 

Good observation Dogpile. Also note in the attachment here that the VWAP = POC = 1497.50 at the end of the day, a sign of a symmetric distribution. We'll discuss trading a symmetric distribution in a later thread.

 

Cooter if you are reading this, here is an example of price action touching the 3rd SD at 15:26 EST.

[Nvesta81 10]

This might be better served for a later thread, but I'm wondering if the previous day's PVP has any bearing on where things might go in the present day's open? For example, if yesterday's PVP is at 13000 and the market gaps up to say 13080, wouldn't the skew be to the upside from the start if you account for yesterday's volume distribution? Or is the present day data the only thing we should be concerned about to gauge the skew of the present day's market?

[BlowFish 10]

Here's an interesting thing. This mornings DAX. The trend is down but as it just started and went with no base today to work from the PvP is jumping ahead of the VWAP which eventually catches up and overtakes. This causes a flip however it is clearly (to me) quite a different type of flip to that caused by a new base being built. Essentially we are going down on ever increasing volume with no base in the background (not a base today at least) for reference.

 

A couple of pictures that illustrate. I cant help feel this is exploitable, but I am trying hard not to second guess you Jerry! Actually I have a bit but would rather wait before going off in the wrong direction.

 

Any comments? I guess things like this will be covered in 'flips' and perhaps the section looking at longer time profiles.

 

Comments welcome, appreciation growing.

 

Cheers.

[O66 10]

blowfish is the pvp eld you are using available?

im using also multicharts but cant find a working eld

 

thanks

[jperl 11]

This might be better served for a later thread, but I'm wondering if the previous day's PVP has any bearing on where things might go in the present day's open? For example, if yesterday's PVP is at 13000 and the market gaps up to say 13080, wouldn't the skew be to the upside from the start if you account for yesterday's volume distribution? Or is the present day data the only thing we should be concerned about to gauge the skew of the present day's market?

 

Good question Nvesta. Previous days statistics do play a role in the price action for today, but not in the way you might think. They act as Hold Up Prices which I call HUP. We will discuss these in a future thread.

[BlowFish 10]

As to whether the indicator is available ...well its a bit quirky I took ghkramer's volume profile and severely hacked it about. Its a bit 'fragile' to let out in the wild. There seem to be a couple of problems that need sorting.

 

Having said that I learnt enough about he he did his volume profile to be able to attempt a simple PvP using his concepts. Its just finding the time. To be honest I was just messing while hoping Dbtina or someone else came up with something a bit more elegant, and more to the point with PvP in mind from the start.

 

Cheers,

[Guest cooter]

 

Cooter if you are reading this, here is an example of price action touching the 3rd SD at 15:26 EST.

 

Yep. That's what I was looking for - thxs Jerry.

 

Note the pullback at that point.

 

[O66 10]

As to whether the indicator is available ...well its a bit quirky I took ghkramer's volume profile and severely hacked it about. Its a bit 'fragile' to let out in the wild. There seem to be a couple of problems that need sorting.

 

Ok no problem.

 

i did find an eld which im testing now. (cost $100,-)

see screenshot

red lines are yesterdays high and low area and in the middle yesterdays closing pvp.

The dark blue lines are todays pvp.

is this chart correct ?

 

also this one is working (cost 150,-)

http://www.theswingmachine.com/Price_Distrib_Anal/PriceDist_Main.htm

 

but he if i can find it for free.....

 

also thinking about switching to Investor/RT

need good charts...any ideas about this?

[Z_trade 10]

O66

you might want to take a look on eSignal's Volume Profile and Vwap, though its not as customized as an Ensign software.

[O66 10]

3 newbie trades in a row today on the ER, each good for 4 points

nice vwap shuffle!

[dbntina 10]

Alright TS users (blowfish do you have TS and you were just banned from forums??)

 

Anyway...Here is where we are:

 

I do have code that will calculate PVP on 1 tick chart. Btw I have 4 megs of ram and I can't run the code that will calculate VWAP and SD's on a 1 tick chart...way too memory intensive..computer keeps locking...so I will have to settle for being a tick or two off during the day with current eld I posted here.

 

I was able to get the ADE working and I now can see the PVP updating realtime with the VWAP and SD bands.

 

To get my setup at a minimum you will need to:

 

1) Keep a 1 tick chart running with the first PVP.eld indicator on it.

2) Then you can have your higher timeframe chart (2min) open with the updated VWAPSD indicator on it and it will plot the PVP from the 1T chart and the VWAP and SD bands.....

 

However...to those who are familiar with programming, posting indicators and working with ADE and eld's I need some guidance.

 

For people who don't have the ADE library, ELCollections and ZeroSynch library are they able to still use the indicators? Or do they have to install that first....My gut feeling tells me that the indicators that they need will get installed with the eld (I hope).

 

Also, I would like to send the eld's to someone first who knows more about this stuff to troubleshoot and figure out what is needed to post for everyone's use and to help me with it. Test and see basically if you only need the 2 eld's to work.

 

One more thing (sorry to hijack the thread jperl), I don't know if the ticks go back historically because throughout the day the PVP updated well...looking at the matrix and comparing...however towards the end of the day 1973 and 1971.50 or so lower were different in volume by 200 or so and it didn't move lower...however I lost the data feed twice during the day and I don't know if that is the reason why it was a little off....so caution with that.

 

Let me know if anyone can help me out.

 

I tell you what, the code is currently keeping track of all the volume at every price tick...to plot the volume histogram would be very easy using trendlines...I just need to add that into the code...should not be hard at all...then we got everything we need! If anyone feels like running with it let me know.

 

Thanks,

 

dbntina

[jperl 11]

3 newbie trades in a row today on the ER, each good for 4 points

nice vwap shuffle!

 

Good trading O66.

[dbntina 10]

Ok here a pic of today's action on 2min of NQ...I think at the end of day it should have moved lower like I was talking about and losing data twice.

 

dbntina

[Blu-Ray 10]

Ok here a pic of today's action on 2min of NQ...I think at the end of day it should have moved lower like I was talking about and losing data twice.

 

dbntina

 

Impressive work there dbntina, if you don't mind I would like a try.

 

Thanks

 

Blu-Ray

[dbntina 10]

No problem Blu-Ray...will send it tomorrow morning...would like to get some people to test it out...and I can make changes if we need to and get it documented and once we get the bugs out I will post for everyone here...next version I will get the histogram put on and then we are in business...then the following version will have an option to combine multiple days of data as an option...then I am through with this....

 

dbntina

[walterw 10]

No problem Blu-Ray...will send it tomorrow morning...would like to get some people to test it out...and I can make changes if we need to and get it documented and once we get the bugs out I will post for everyone here...next version I will get the histogram put on and then we are in business...then the following version will have an option to combine multiple days of data as an option...then I am through with this....

 

dbntina

 

 

dbntina... nice actitude ¡¡ thanks for great collaboration and welcome aboard TL... that indicator looks very nice... cheers Walter.