search

zcakhaa / DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books

This jupyter notebook is used to demonstrate our recent work, "DeepLOB: Deep Convolutional Neural Networks for Limit Order Books", published in IEEE Transactions on Singal Processing. We use FI-2010 dataset and present how model architecture is constructed here. The FI-2010 is publicly avilable and interested readers can check out their paper.
374 stars 205 forks source link

A question regarding threshold for the percentage change #9

Open shadow111 opened 3 years ago

shadow111 commented 3 years ago

Hello, Can you please explain how you decide to fix the threshold to 0.002 for labelling? is there any logic behind Best Regards Lafi RAED

zcakhaa commented 3 years ago

Hi,

Are you referring to the threshold in FI-2010? We didn't use 0.002. In our case, we choose a threshold for each instrument to have a balanced label in the training set.

From: Lafi Raed @.> Sent: Friday, March 26, 2021 3:19 PM To: zcakhaa/DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books @.> Cc: Subscribed @.***> Subject: [zcakhaa/DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books] A question regarding threshold for the percentage change (#9)

Hello, Can you please explain how you decide to fix the threshold to 0.002 for labelling? is there any logic behind Best Regards Lafi RAED

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/zcakhaa/DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books/issues/9, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AFN66TDQBTF3Z4OA44XADWDTFSX2BANCNFSM4Z3SVUCQ.

shadow111 commented 3 years ago

i'm referring to the threshold in general for the labeling ,i'm aware that for each dataset this threshold will change , is there a logic on how to choose it? can you explain how to fix this threshold

zcakhaa commented 3 years ago

It depends on your application. You can relate it to transaction cost or choose to have a balanced label set.

From: Lafi Raed @.> Sent: Friday, March 26, 2021 5:06 PM To: zcakhaa/DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books @.> Cc: Zihao Zhang @.>; Comment @.> Subject: Re: [zcakhaa/DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books] A question regarding threshold for the percentage change (#9)

i'm referring to the threshold in general for the labeling ,i'm aware that for each dataset this threshold will change , is there a logic on how to choose it? can you explain how to fix this threshold

— You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://github.com/zcakhaa/DeepLOB-Deep-Convolutional-Neural-Networks-for-Limit-Order-Books/issues/9#issuecomment-808376718, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AFN66TH45YP3F76LMCK4A6TTFTEJDANCNFSM4Z3SVUCQ.

shadow111 commented 3 years ago

yes I wanna to have a balanced labeled dataset in my case