Should we set OrdStatus to Filled in the replace confirm if an order becomes fully filled in the middle of a replace request in 4.1 ?

, ,
1

Imported from previous forum

2

[ original email was from David Hu - david.hu@csfb.com ]

Appendix D of FIX 4.2 spec explains how we should set OrdStatus according to precedence really well. Just wondering, for FIX 4.1, does it clearly state a Filled status takes precedence over Replaced so if an order got fulled filled while it is in a pending replace state, then we should set OrdStatus=2 and ExecType=5 in a replace confirmation if the replace is successful ?

Many thanks

3

[ original email was from John Agbai - john.agbai@drkw.com ]
>

Appendix D of FIX 4.2 spec explains how we should set OrdStatus
according to precedence really well. Just wondering, for FIX 4.1, does
it clearly state a Filled status takes precedence over Replaced so if an
order got fulled filled while it is in a pending replace state, then we
should set OrdStatus=2 and ExecType=5 in a replace confirmation if the
replace is successful ?

Many thanks

Strictly speaking if the order became filled while in a pending replace state then you should never be confirming the replace as it should fail.
e.g.
→ New Order Single (A)
← Exec Report:(A, Pending New)
← Exec Report:(A, New)
→ Order Modification Request (B,A)
← Exec Report:(A, Filled)
← CancelReject: B,A (“Order Filled”)

4

[ original email was from David Hu - david.hu@csfb.com ]
Hi John,

So there are no circumstances that the cancel replace can be accepted ? Not even say, in the case the cancel replace is to change from mkt to limit and involves no quantity changes ?

Appendix D of FIX 4.2 spec explains how we should set OrdStatus
according to precedence really well. Just wondering, for FIX 4.1, does
it clearly state a Filled status takes precedence over Replaced so if
an order got fulled filled while it is in a pending replace state,
then we should set OrdStatus=2 and ExecType=5 in a replace
confirmation if the replace is successful ?

Many thanks

Strictly speaking if the order became filled while in a pending replace
state then you should never be confirming the replace as it should fail.
e.g. → New Order Single (A) ← Exec Report:(A, Pending New) ← Exec
Report:(A, New) → Order Modification Request (B,A) ← Exec
Report:(A, Filled) ← CancelReject: B,A (“Order Filled”)

5

[ original email was from John Agbai - john.agbai@drkw.com ]
> Hi John,

So there are no circumstances that the cancel replace can be accepted ?
Not even say, in the case the cancel replace is to change from mkt to
limit and involves no quantity changes ?

Appendix D of FIX 4.2 spec explains how we should set OrdStatus
according to precedence really well. Just wondering, for FIX 4.1,
does it clearly state a Filled status takes precedence over Replaced
so if an order got fulled filled while it is in a pending replace
state, then we should set OrdStatus=2 and ExecType=5 in a replace
confirmation if the replace is successful ?

Many thanks

Strictly speaking if the order became filled while in a pending
replace state then you should never be confirming the replace as it
should fail.
e.g. → New Order Single (A) ← Exec Report:(A, Pending New) ←
Exec Report:(A, New) → Order Modification Request (B,A) ←
Exec Report:(A, Filled) ← CancelReject: B,A (“Order Filled”)

Some systems do allow an order to be re-opened after it has been filled when the volume was being increased (e.g. Euronext), however in most systems after the sell side has said the order is filled then any subsequene request should be rejected. (Except OrderStatus Request)

John Agbai