[ This is 2nd supporting document, that was attached with mid-year rebuttal , with EMMI team explaining how EMMI is different from EMTS ]
There are differences expected at every level. Per EMMI team EMTS feed can not be considered as a benchmark.
Mark Yadid's comments regarding differences:
The source of the differences are:
1) switch of emphasis from sensitivity to price based
2) change in methodology for generation of sensitivities
3) poor quality of current sensitivity data sent to VaRs which EMMI is expected to solve
Alik Barg’s email about differences:
Diane,
I discussed the "EMTS feed vs Murex feed reconciliation" issue with BrownMan, Siddharth, and with other project participants individually and at our joint meetings:
I am strongly convinced that there are too many differences between the Murex and the EMTS that make the EMTS vs. Murex feed comparison questionable.
As you know EMTS feed can not be considered as a benchmark.
Mark Yadid in his email sent today remarks that there many differences on the individual position level.
The source of the differences are
1) switch of emphasis from sensitivity to price based
2) change in methodology for generation of sensitivities
3) poor quality of current sensitivity data sent to VaRs which EMMI is expected to solve
I can add at least several other reasons that make me liken the EMTS vs Murex comparison to the apples-oranges comparison:
EMTS feed carries duplicated positions;
EMTS does not handle correctly dual currencies;
Murex introduces forward bond positions;
Murex aggregates trades by positions and the EMTS sends individual trades;
Any measure - VAR, or LMV, or quantities, or instruments, etc. - one would like to use for the feeds reconciliation is to be distorted by these factors.
As I expressed prior, in my opinion dealing with such reality it would be more prudent to approach to the Murex feed as to a new one.
The relationship between the EMTS and the Murex is only that one should assure:
the completeness of the EMTS to the Murex position migration;
the correctness of the position;
that required positions are being included in the VARS feed fully and correctly;
that feed has a proper format for the further processing;
I believe that the first two conditions were fulfilled during EMTS->Murex migration, which have been accompanied by multiple reconciliations and by business sign-offs. I don't think there is a reason to go through this process again. Items ##3,4 we were working on the last several months and did our best to satisfy them too.
I hope that we'll find a middle-ground in our approach and we'll accomplish this our mission soon.
Thank you.
Regards,
-Alik
P.S. Quantities in the feeds won't match because of duplicating issue in the EMTS. I did reconcile quantities sent in the feed vs Murex DB data.
Christian's comments:
Multiphase bonds were incorrectly calculated in EMTS
EQTY positions are not migrating from EMTS to EMMI feed
