Subject: uk submission of positions
just to follow up on james ' s note of yesterday - my apologies for being out
the elimination of the requirement to ' grab ' the fx and ir market environment
from houston ( and consequently use european data ) can improve our ability
to kick the overnight batch processing off earlier in the evening . james and
i will be working with brian hudson to determine the timetable for effecting
this change
the revised batch start time would improve the opportunity to detect any
system failure and complete a rerun of the valuation process so as to deliver
completed results by the time the risk management team arrive at their desks
in the morning . system failures are the most significant problem we face in
delivering timely information to houston . consequently the probability of
meeting current reporting deadlines would be greatly improved given that we
will have full it overnight support covering for any it failure and curve
input validation processes on trade date .
indeed given the successful completion of all overnight runs we are able to
deliver
officialisation of all valuation systems and most spreadsheets ( david hardy
is in final testing of the eastern spreadsheet feed ) by the 10 am deadline (
houston 4 am )
flash p & l by the lpm deadline ( houston 7 am )
completion of the final dpr to be submitted to houston by the 5 pm deadline (
houston 11 am )
james and i are currently finalising a document reviewing all possible
changes to business processes that could improve these times . this is in
addition to working with commercial and it to assess the possibility of
delivering significantly faster revaluation systems that could assist in
delivering a trade date control process .
i can confirm that there is significant work to do in this area , however , we
are dedicated to meeting the objectives of improved / more real time control .
please feel free to call if the above requires any additional commentary .
regards
mike
- - - - - - - - - - - - - - - - - - - - - - forwarded by mike jordan / lon / ect on 19 / 12 / 2000 15 : 06
- - - - - - - - - - - - - - - - - - - - - - - - - - -
james new
18 / 12 / 2000 09 : 10
to : rick buy / hou / ect @ ect
cc : john sherriff / lon / ect @ ect , michael r brown / lon / ect @ ect , fernley
dyson / lon / ect @ ect , mike jordan / lon / ect @ ect , ted murphy / hou / ect @ ect , gary
hickerson / hou / ect @ ect
subject : uk submission of positions
rick ,
thanks you for your note below . we are today implementing a flash p & l
process . we aim to report daily numbers at 7 am houston time and will start
today with the p & l . we will build on this and hope to have draft positions
and var for most books within a week or so . there will be a reconciliation of
flash to final numbers which will be included in our return to houston .
i will put together a note on the london dpr production process which goes
into the process we currently have , the process we actually need to have , and
the obstacles that are in the way . i would expect this to be finalised
tomorrow . obtaining the usd interest rate curve on a more timely basis is
just one of our problems but it is unfortunately by no means our only or most
serious problem .
james
from : rick buy
15 / 12 / 2000 21 : 59
to : john sherriff / lon / ect @ ect , michael r brown / lon / ect @ ect , mike
jordan / lon / ect @ ect
cc : gary hickerson / hou / ect @ ect , ted murphy / hou / ect @ ect
subject : uk submission of positions
need your help on the following : each day we are delayed in finalising var ,
p & l and positions because the uk must wait for a usd interest rate curves
before submitting their data to houston . i am also told that it is really not
necessary to wait for this curve and the data could be submitted close of
business london . even if there was some minor inaccuracy from this method it
would be better than what we have now . this would greatly improve the
efficiency ( by 4 to 6 hours ) in reporting to senior management . can you guys
initiate this change or get me to the right person there . thanks , rick