Skip to main content

Table 9 Training accuracy rates of own and VAE-reduced stock features

From: An efficient stock market prediction model using hybrid feature reduction method based on variational autoencoders and recursive feature elimination

Stocks

LSTM (basic)

LSTM (with attention)

SVM

LightGBM

Own

VAE

Own

VAE

Own

VAE

Own

VAE

ALBRK

0.729

0.691

0.773

0.756

0.863

0.823

0.783

0.761

AKBNK

0.626

0.589

0.685

0.686

0.820

0.784

0.720

0.691

HALKB

0.653

0.637

0.665

0.652

0.841

0.780

0.721

0.696

ISCTR

0.619

0.638

0.696

0.679

0.825

0.773

0.726

0.708

SKBNK

0.723

0.723

0.780

0.742

0.862

0.804

0.771

0.741

TSKB

0.640

0.681

0.721

0.723

0.835

0.799

0.767

0.733

VAKBN

0.637

0.618

0.682

0.659

0.839

0.795

0.717

0.685

YKBNK

0.669

0.623

0.677

0.662

0.839

0.781

0.709

0.671

Average

0.662

0.651

0,709

0.694

0.840

0.792

0.739

0.710