Backtesting (slippage, historical data)  [SOLVED]

Questions about MultiCharts .NET and user contributed studies.
chosenek
Posts: 26
Joined: 01 Mar 2015
Has thanked: 2 times

Backtesting (slippage, historical data)

Postby chosenek » 13 Mar 2015

Hi,
I noticed that during backtesting slippage value is added twice in case of short order and only once in case of long order.
Is there a specific reason for that?

Also I am not able to use more ~65 days of historical data during backtesting (although I set 1 year). The tested instrument was e-mini Russel 2000 using custom future.

I am still unsure with this multichart .net edition and CQG combination. Lots of things seem to be weird. I hope its just me being a newcomer to MC.

Here is link to my other thread with questions:
http://www.multicharts.com/discussion/v ... 19&t=48309

User avatar
JoshM
Posts: 2195
Joined: 20 May 2011
Location: The Netherlands
Has thanked: 1544 times
Been thanked: 1565 times
Contact:

Re: Backtesting (slippage, historical data)

Postby JoshM » 15 Mar 2015

I noticed that during backtesting slippage value is added twice in case of short order and only once in case of long order.
Is there a specific reason for that?
Stop and market orders in MultiCharts .NET have slippage added to them (see here). Were your long orders stop/market orders or did you use limits?

chosenek
Posts: 26
Joined: 01 Mar 2015
Has thanked: 2 times

Re: Backtesting (slippage, historical data)

Postby chosenek » 16 Mar 2015

Hi Josh,

My PT is set to 230$/SL is set to 110$./slippage to 2$.
As you can see in attached screenshort slippage differs depending on whether it was PT or SL. In case of PT only one slippage is applied in case of SL, slippage is reduced twice from the trade . These are just a backtesting results.

I am using the basic methods to generate PT/SL (GenerateProfitTarget, GenerateStopLoss)

Maybe its just me having a gaps in slippage/order type logic. But an explanation would help me sleep better.

Thank you in advance.


EDIT: I can see that the screenshot contains only short orders. But the same result is in case of long orders.
Attachments
slippage.jpg
(95.05 KiB) Downloaded 675 times

User avatar
JoshM
Posts: 2195
Joined: 20 May 2011
Location: The Netherlands
Has thanked: 1544 times
Been thanked: 1565 times
Contact:

Re: Backtesting (slippage, historical data)  [SOLVED]

Postby JoshM » 20 Mar 2015

(...) slippage differs depending on whether it was PT or SL. In case of PT only one slippage is applied in case of SL, slippage is reduced twice from the trade .
If you enter with a market order and exit with a limit order (PT) --> one market order, so slippage applied once.

If you enter with a market order and exit with a market order (SL) --> two market orders, so slippage applied twice.

That is at least my understanding of it.

chosenek
Posts: 26
Joined: 01 Mar 2015
Has thanked: 2 times

Re: Backtesting (slippage, historical data)

Postby chosenek » 20 Mar 2015

Make sense.

Thank you.
Petr


Return to “MultiCharts .NET”