You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have come across an unusual bug:
When backtesting, if a situation is created for a historical event where:
Trade created, back Order created with trade
back Order reaches EXECUTION_COMPLETE
New Trade created, lay Order created with new trade
lay Order reaches EXECUTION_COMPLETE
However, if this is exact same execution is modified so that in step 3. the lay Order is created with the same Trade used in step 1. for the back Order, then the lay Order never reaches EXECUTION_COMPLETE?
The text was updated successfully, but these errors were encountered:
I think I have found the bug: in FlumineBacktest._process_backtest_orders(), it has
if order.trade.status.value == "Complete":
blotter.complete_order(order)
However, if a Trade has one complete Order then its status will go to Complete, and only after a new Order has been placed will the Trade status go back to LIVE, at which point the blotter has already deemed the Order as Complete
I have come across an unusual bug:
When backtesting, if a situation is created for a historical event where:
Trade
created, backOrder
created with tradeOrder
reachesEXECUTION_COMPLETE
Trade
created, layOrder
created with new tradeOrder
reachesEXECUTION_COMPLETE
However, if this is exact same execution is modified so that in step 3. the lay
Order
is created with the sameTrade
used in step 1. for the backOrder
, then the layOrder
never reachesEXECUTION_COMPLETE
?The text was updated successfully, but these errors were encountered: