Iterative Refinement

Quote from tobbe:

Mostly for fun - here's another way of viewing the pace data for ES posted previously. Imagine you need at least 3 point bars to trade profitably - check the table, you need at least 25700 PRV to trade. The point numbers are up to the following number, i.e. 1 is for 1-1.75 points, 2 is for 2-2.75 etc.

<img src=http://www.elitetrader.com/vb/attachment.php?s=&postid=1825411>

Please, don't forget there are ... lies ... lies ... and statistics ... :p The above table is based on far too little data (february only). If I post it again I'll do it for more bars and with relevant confidence intervals etc.

cheers,
tobbe
More...

Hi Tobbe, thanks for the info. Not to get too far OT (especially since I'm not a maths person), but I once did a PRV study using Vorzo's version of Mak's excel spreadsheet thinking that one month of 5 min. bars would be enough data. How much data do you think is necessary? And, if you believe someone like me can understand, why? It's interesting that you included the first three bars of the day in your study.

For anyone wondering, here's where Vorzo pasted the PRV study spreadsheet.
http://www.elitetrader.com/vb/showthread.php?s=&threadid=83604&perpage=6&pagenumber=1504

Right now my focus has shifted to nailing down other parts of the equities/futures methodology (thanks Spyder), but since you brought up pace lines, I thought to ask now and incorporate later.
 
I am also using Vorzo's spredsheet but get different volume levels than Tobbe.

Tobbe:
8000, 12500, 17000, 24500 and 30000

Vorzo's spreadsheet:
Data is last 30 days.

EXTREME 40000 and higher
FAST 26500-40000
MEDIUM 19000-26500
SLOW 14000-19000
DU 9000-14000

Anyone else using this sheet?

regards,
Ivo


Quote from Padawan:

Hi Tobbe, thanks for the info. Not to get too far OT (especially since I'm not a maths person), but I once did a PRV study using Vorzo's version of Mak's excel spreadsheet thinking that one month of 5 min. bars would be enough data. How much data do you think is necessary? And, if you believe someone like me can understand, why? It's interesting that you included the first three bars of the day in your study.

For anyone wondering, here's where Vorzo pasted the PRV study spreadsheet.
http://www.elitetrader.com/vb/showthread.php?s=&threadid=83604&perpage=6&pagenumber=1504

Right now my focus has shifted to nailing down other parts of the equities/futures methodology (thanks Spyder), but since you brought up pace lines, I thought to ask now and incorporate later.
More...
 
Quote from Padawan:
How much data do you think is necessary? And, if you believe someone like me can understand, why? It's interesting that you included the first three bars of the day in your study.
More...
I think 30 days of data or so is enough for the pace table as posted by MAK and Vorzo. It's not enough for a table like the second one I posted. For example there were only 2 bars that made up the volume for 10 points. Statistics is mostly about large samples and only 2 would be considered far too little.
That said, I personally believe market data is non-stationary and not normally distributed, so "normal" statistics isn't of much use anyways. Sorry if I caused any confusion.
 
Quote from ivob:
I am also using Vorzo's spredsheet but get different volume levels than Tobbe.
More...
I had forgotten about Vorzo's spreadsheet (and do I feel kind of stupid :p ), thanks Padawan and Ivo for the reminder. The pace table/chart I posted was for february. If you used the last 30 days of data, that would explain the difference.

Even if you used data from the same period but from another provider there would probably be some difference (but hopefully smaller). I don't think your average feed vendor (esignal, qcharts, iqfeed, ib etc) spend very much time cleaning their tick data so there's bound to be differences.
 
Well, I've been away for a while so this is my first full-day for a while. The Guassians sure seemed to want to change shape alot. I did find glancing at a 15min chart helpful today.
 

Attachments

Sorry about the messy chart, I found it difficult to match the gaussians today, very hard day. I traded 1 ES contract but 2 accounts. :D
1ES.png
 
You must log in or register to reply here.
Back
Top