Anyone Trading FTSE/ASE 20 Athens Index at this time?

General discussions about futures.
Post Reply
leslie
Roundtable Knight
Roundtable Knight
Posts: 135
Joined: Tue Apr 12, 2011 11:46 am

Anyone Trading FTSE/ASE 20 Athens Index at this time?

Post by leslie »

Has anyone made a trade FTSE/ASE 20 Athens Index with following FuturesInfo settings?

CSI #659, Symbol "ATF"
_Big Point Value=5
_Decimal=2
_Tick Unit=0.01
_Minimum Tick=25

Are these the correct settings?
Does expected round turn profit (loss) match BROKER statement?

The exchange fact sheet is here:

http://www.adex.ase.gr/AdexDownload/GBs ... e20_if.pdf

so I'd expect to see price increments of 0.25, 0.50, 0.75 (equal to EUR 1.25, EUR 2.5, etc).

However, I am seeing price increments of 0.01 in CSI data!
Feeling suspicious on this one.

Thank you,
L
p.s. late in the day... staring at screen too long




Is this correct?
Can someone please

Not the issue is that when I look into the
sluggo
Roundtable Knight
Roundtable Knight
Posts: 2987
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo »

viewtopic.php?p=20189&highlight=burmese#20189

The follow up post by Nickmar is especially relevant / chilling .
leslie
Roundtable Knight
Roundtable Knight
Posts: 135
Joined: Tue Apr 12, 2011 11:46 am

Post by leslie »

Sluggo,

Thank you.
Yes, messing up BPV is one of the costliest mistakes one can make. Thank you for the suggestion. I'll follow up & make a test trade with the specific broker-datafeed combination.

At moment, I am configuring, back testing, & coding, so just need to have setting verified.

Someone on this forum *does* trade Athens.
It was their suggestion which prompted me to check liquidity. (I just cannot find the posting.) Could they please have a look at the numbers and say 'ye or ney'.

Thanks,
L
sluggo
Roundtable Knight
Roundtable Knight
Posts: 2987
Joined: Fri Jun 11, 2004 2:50 pm

Post by sluggo »

For purposes of simulation, an estimated BPV will probably be good enough. It affects the simulated position size:
  • #Contracts = (betsize * Equity) / (distance_to_stop * BPV)
If your estimated BPV is a factor of 3 too high, then your simulations will trade a factor of 3 too few contracts. BUT, each of those contracts will have a factor of 3 greater profit and loss. Voila: the same dollar profits
  • (Correct#Contracts / 3) * (CorrectBPV * 3) = CorrectDollarProfit
The same is true if your estimated BPV is a factor of 3 too low. Now your simulations will trade a factor of 3 too many contracts, but each of those contracts will have a factor of 3 smaller profit and loss. Voila part 2: the same dollar profits.

The only discrepancies will be (1) in rounding fractional-contract position sizes to integer-contract position sizes; (2) Slippage and Commissions. You may decide that these are acceptably small errors, and that you're willing to live with them while you wait for an anonymous benefactor to provide the gifts you seek.

One way you can estimate the BPV is to assume that this contract has about the same (BPV * ATR) as its peers. Perhaps
  • Make a list of contracts that you consider to be peers of this one
  • Compute BPV * ATR for each of the peers (I'd compute the average value of BPV*ATR20 for the past two years but, hey, that's only me)
  • Compute the mean value of (BPV * ATR) for the peer group
  • Choose BPV for this contract so that its (BPV * ATR) is approximately equal to the mean value of its peers' (BPV * ATR)
  • Done
Once you are ready to actually trade it with actual money, you'll want to replace the estimate with a correct (experimentally verified) value!
leslie
Roundtable Knight
Roundtable Knight
Posts: 135
Joined: Tue Apr 12, 2011 11:46 am

Post by leslie »

Sluggo

Thank you indeed.
You are quite correct.

In fact, I was going to delay the test trade 'till after I finish going through every desired market in the FuturesInfo file. Given you suggesting, a very useful one, I shall do just that.

Data administration is such a dull task.
It can tire one quickly, so reading your advice cheered my up. You offered good reason why going on (as opposed to fussing) is the correct solution. Thank you.

L
Chris67
Roundtable Knight
Roundtable Knight
Posts: 1052
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 »

USING the following will get you the correct back tested results and a P/L that matches what your broker tells you you have made

BPV = 5
Disp Dig = 2
0.01
Min Tick = 1
Euro 5* Index

Whether or not the unadjusted close prices are displayed in a way slighty different maybe the case - but the above works for me
C
leslie
Roundtable Knight
Roundtable Knight
Posts: 135
Joined: Tue Apr 12, 2011 11:46 am

Post by leslie »

Hi Chris,

Thank you for posting.
You confirmed my suspicion that
Minimum Tick should be 1 and not "25".

With so many markets, confirming every detail is grunt work.

L
Post Reply