FactorVQVAE: Discrete latent factor model via Vector Quantized Variational Autoencoder

This study introduces FactorVQVAE, the first integration of the Vector Quantized Variational Autoencoder (VQVAE) into factor modeling, providing a novel framework for predicting cross-sectional stock returns and constructing systematic investment portfolios. The model employs a two-stage architectur...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Knowledge-based systems Ročník 318; s. 113460
Hlavní autori: Kim, Namhyoung, Ock, Seung Eun, Song, Jae Wook
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Elsevier B.V 07.06.2025
Predmet:
ISSN:0950-7051
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract This study introduces FactorVQVAE, the first integration of the Vector Quantized Variational Autoencoder (VQVAE) into factor modeling, providing a novel framework for predicting cross-sectional stock returns and constructing systematic investment portfolios. The model employs a two-stage architecture to improve the extraction and utilization of latent financial factors. In the first stage, an encoder–decoder-quantizer compresses high-dimensional input data into discrete latent factors through vector quantization, addressing posterior collapse and ensuring distinct representations. In the second stage, an autoregressive Transformer captures sequential dependencies among these latent factors, enabling precise return predictions. Empirical results in the CSI300 and S&P500 markets demonstrate FactorVQVAE’s superior performance. The model achieves the best Rank IC and Rank ICIR scores, surpassing the state-of-the-art latent factor models in varying market conditions. In portfolio evaluations, FactorVQVAE consistently excels in both Top-k Drop-n and Long–Short strategies, translating predictive accuracy into robust investment performance. In particular, it delivers the highest risk-adjusted returns, highlighting its ability to balance returns and risks effectively. These findings position FactorVQVAE as a significant advancement in integrating modern deep learning methodologies with financial factor modeling. Its adaptability, robustness, and exceptional performance in portfolio investment establish it as a promising tool for systematic investing and financial analytics. [Display omitted] •This study introduces FactorVQVAE, integrating VQVAE into dynamic factor modeling.•A two-stage design extracts latent factors and models sequential dependencies.•FactorVQVAE outperforms benchmarks in return prediction for CSI300 and S&P500.•FactorVQVAE demonstrates robustness in portfolio investment across different market conditions.
AbstractList This study introduces FactorVQVAE, the first integration of the Vector Quantized Variational Autoencoder (VQVAE) into factor modeling, providing a novel framework for predicting cross-sectional stock returns and constructing systematic investment portfolios. The model employs a two-stage architecture to improve the extraction and utilization of latent financial factors. In the first stage, an encoder–decoder-quantizer compresses high-dimensional input data into discrete latent factors through vector quantization, addressing posterior collapse and ensuring distinct representations. In the second stage, an autoregressive Transformer captures sequential dependencies among these latent factors, enabling precise return predictions. Empirical results in the CSI300 and S&P500 markets demonstrate FactorVQVAE’s superior performance. The model achieves the best Rank IC and Rank ICIR scores, surpassing the state-of-the-art latent factor models in varying market conditions. In portfolio evaluations, FactorVQVAE consistently excels in both Top-k Drop-n and Long–Short strategies, translating predictive accuracy into robust investment performance. In particular, it delivers the highest risk-adjusted returns, highlighting its ability to balance returns and risks effectively. These findings position FactorVQVAE as a significant advancement in integrating modern deep learning methodologies with financial factor modeling. Its adaptability, robustness, and exceptional performance in portfolio investment establish it as a promising tool for systematic investing and financial analytics. [Display omitted] •This study introduces FactorVQVAE, integrating VQVAE into dynamic factor modeling.•A two-stage design extracts latent factors and models sequential dependencies.•FactorVQVAE outperforms benchmarks in return prediction for CSI300 and S&P500.•FactorVQVAE demonstrates robustness in portfolio investment across different market conditions.
ArticleNumber 113460
Author Song, Jae Wook
Ock, Seung Eun
Kim, Namhyoung
Author_xml – sequence: 1
  givenname: Namhyoung
  orcidid: 0000-0003-4191-8830
  surname: Kim
  fullname: Kim, Namhyoung
  email: x7jeon8gi@hanyang.ac.kr
– sequence: 2
  givenname: Seung Eun
  surname: Ock
  fullname: Ock, Seung Eun
  email: ockdaniel@hanyang.ac.kr
– sequence: 3
  givenname: Jae Wook
  orcidid: 0000-0001-6455-6524
  surname: Song
  fullname: Song, Jae Wook
  email: jwsong@hanyang.ac.kr
BookMark eNp9kFFLwzAUhfMwwW36D3zIH2hN0qZZfRDG3FQYyED7GtLkBjK7RpJsMH-9dfXZpwv33nM455uhSe97QOiOkpwSWt3v88_ex3PMGWE8p7QoKzJBU1JzkgnC6TWaxbgnhDBGF1PUbJROPjS7Zrl-wE8u6gAJcKcS9AnbyxEfvIEOn5zCDVwWu6Pqk_sGgxsVnErO96rDy2Py0OvhOdygK6u6CLd_c44-Nuv31Uu2fXt-XS23maYLnrIWBLVcc1pTUpJqAQUTtW5tZUUNNTFtYQFaUWjGQFGoDW9LIRjwVvDW1LqYo3L01cHHGMDKr-AOKpwlJfKXh9zLkYf85SFHHoPscZTBkO3kIMio3RAdjAtDQ2m8-9_gB-72cIo
Cites_doi 10.1109/CVPR52688.2022.01103
10.1111/j.1540-6261.1997.tb03808.x
10.1109/CVPR46437.2021.01268
10.1016/j.jfineco.2014.10.010
10.1109/CVPR.2017.113
10.1111/j.1540-6261.2011.01671.x
10.2307/2325486
10.1145/2939672.2939785
10.1016/0304-4076(92)90072-Y
10.1016/j.jeconom.2020.07.009
10.1609/aaai.v38i1.27767
ContentType Journal Article
Copyright 2025 Elsevier B.V.
Copyright_xml – notice: 2025 Elsevier B.V.
DBID AAYXX
CITATION
DOI 10.1016/j.knosys.2025.113460
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Computer Science
ExternalDocumentID 10_1016_j_knosys_2025_113460
S0950705125005076
GroupedDBID --K
--M
.DC
.~1
0R~
1B1
1~.
1~5
4.4
457
4G.
5VS
7-5
71M
77K
8P~
9JN
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AATTM
AAXKI
AAXUO
AAYFN
AAYWO
ABAOU
ABBOA
ABIVO
ABJNI
ABMAC
ACDAQ
ACGFS
ACRLP
ACZNC
ADBBV
ADEZE
ADGUI
ADTZH
AEBSH
AECPX
AEIPS
AEKER
AENEX
AFJKZ
AFTJW
AFXIZ
AGCQF
AGHFR
AGRNS
AGUBO
AGYEJ
AHHHB
AHJVU
AHZHX
AIALX
AIEXJ
AIIUN
AIKHN
AITUG
AKRWK
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
AOUOD
APXCP
ARUGR
AXJTR
BJAXD
BKOJK
BLXMC
BNPGV
CS3
DU5
EBS
EFJIC
EO8
EO9
EP2
EP3
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
GBOLZ
IHE
J1W
JJJVA
KOM
LG9
LY7
M41
MHUIS
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
PQQKQ
Q38
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SSH
SST
SSV
SSW
SSZ
T5K
WH7
XPP
ZMT
~02
~G-
29L
77I
9DU
AAQXK
AAYXX
ABDPE
ABWVN
ABXDB
ACLOT
ACNNM
ACRPL
ACVFH
ADCNI
ADJOM
ADMUD
ADNMO
AEUPX
AFPUW
AGQPQ
AIGII
AKBMS
AKYEP
ASPBG
AVWKF
AZFZN
CITATION
EFKBS
EFLBG
EJD
FEDTE
FGOYB
G-2
HLZ
HVGLF
HZ~
R2-
SBC
SET
UHS
WUQ
~HD
ID FETCH-LOGICAL-c185t-be71f5c519104068e3279cbf6f79e90db3feeb73c22ea1e9d5b4772e5b75bd9c3
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001473508700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0950-7051
IngestDate Sat Nov 29 07:54:42 EST 2025
Sat Jun 07 17:01:27 EDT 2025
IsPeerReviewed true
IsScholarly true
Keywords Portfolio investment
Transformer
Vector quantization
Dynamic latent factor model
Autoencoder
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c185t-be71f5c519104068e3279cbf6f79e90db3feeb73c22ea1e9d5b4772e5b75bd9c3
ORCID 0000-0001-6455-6524
0000-0003-4191-8830
ParticipantIDs crossref_primary_10_1016_j_knosys_2025_113460
elsevier_sciencedirect_doi_10_1016_j_knosys_2025_113460
PublicationCentury 2000
PublicationDate 2025-06-07
PublicationDateYYYYMMDD 2025-06-07
PublicationDate_xml – month: 06
  year: 2025
  text: 2025-06-07
  day: 07
PublicationDecade 2020
PublicationTitle Knowledge-based systems
PublicationYear 2025
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Ramesh, Pavlov, Goh, Gray, Voss, Radford, Chen, Sutskever (b22) 2021
P. Esser, R. Rombach, B. Ommer, Taming transformers for high-resolution image synthesis, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 12873–12883.
Carhart (b13) 1997; 52
Hendrycks, Gimpel (b30) 2016
Yang, Liu, Zhou, Bian, Liu (b31) 2020
Duan, Wang, Zhang, Li (b6) 2022
Veličković, Cucurull, Casanova, Romero, Lio, Bengio (b16) 2017
Liu, Hu, Zhang, Wu, Wang, Ma, Long (b19) 2023
Lundberg (b32) 2017
C. Lea, M.D. Flynn, R. Vidal, A. Reiter, G.D. Hager, Temporal convolutional networks for action segmentation and detection, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 156–165.
Chung, Gulcehre, Cho, Bengio (b15) 2014
Gu, Kelly, Xiu (b7) 2021; 222
T. Chen, C. Guestrin, Xgboost: A scalable tree boosting system, in: Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794.
Van den Oord, Kalchbrenner, Espeholt, Vinyals, Graves (b21) 2016; 29
T. Li, Z. Liu, Y. Shen, X. Wang, H. Chen, S. Huang, MASTER: Market-Guided Stock Transformer for Stock Price Forecasting, in: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 1, 2024, pp. 162–170.
Fama, French (b3) 2015; 116
Jensen (b29) 1968; 23
Van Den Oord, Vinyals (b8) 2017; 30
Sharpe (b11) 1964; 19
Fama (b27) 1970; 25
Kingma, Welling (b4) 2013
Ng, Engle, Rothschild (b28) 1992; 52
Cochrane (b1) 2011; 66
Vaswani, Shazeer, Parmar, Uszkoreit, Jones, Gomez, Kaiser, Polosukhin (b18) 2017; 30
Rezende, Mohamed (b9) 2015
Huh, Cheung, Agrawal, Isola (b26) 2023
H. Chang, H. Zhang, L. Jiang, C. Liu, W.T. Freeman, Maskgit: Masked generative image transformer, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 11315–11325.
Fama, French (b2) 1992; 47
Ross (b12) 1973
Wei, Rao, Dai, Lin (b5) 2023
Lee, Malacarne, Aune (b25) 2023
Nelson, Pereira, De Oliveira (b10) 2017
Veličković (10.1016/j.knosys.2025.113460_b16) 2017
Vaswani (10.1016/j.knosys.2025.113460_b18) 2017; 30
Fama (10.1016/j.knosys.2025.113460_b27) 1970; 25
Ng (10.1016/j.knosys.2025.113460_b28) 1992; 52
Jensen (10.1016/j.knosys.2025.113460_b29) 1968; 23
Yang (10.1016/j.knosys.2025.113460_b31) 2020
Hendrycks (10.1016/j.knosys.2025.113460_b30) 2016
Fama (10.1016/j.knosys.2025.113460_b2) 1992; 47
Carhart (10.1016/j.knosys.2025.113460_b13) 1997; 52
Chung (10.1016/j.knosys.2025.113460_b15) 2014
Van Den Oord (10.1016/j.knosys.2025.113460_b8) 2017; 30
Cochrane (10.1016/j.knosys.2025.113460_b1) 2011; 66
10.1016/j.knosys.2025.113460_b24
10.1016/j.knosys.2025.113460_b23
10.1016/j.knosys.2025.113460_b20
Nelson (10.1016/j.knosys.2025.113460_b10) 2017
Lundberg (10.1016/j.knosys.2025.113460_b32) 2017
Ross (10.1016/j.knosys.2025.113460_b12) 1973
Duan (10.1016/j.knosys.2025.113460_b6) 2022
Fama (10.1016/j.knosys.2025.113460_b3) 2015; 116
Sharpe (10.1016/j.knosys.2025.113460_b11) 1964; 19
Huh (10.1016/j.knosys.2025.113460_b26) 2023
Gu (10.1016/j.knosys.2025.113460_b7) 2021; 222
Van den Oord (10.1016/j.knosys.2025.113460_b21) 2016; 29
Ramesh (10.1016/j.knosys.2025.113460_b22) 2021
Kingma (10.1016/j.knosys.2025.113460_b4) 2013
Rezende (10.1016/j.knosys.2025.113460_b9) 2015
Liu (10.1016/j.knosys.2025.113460_b19) 2023
10.1016/j.knosys.2025.113460_b17
Wei (10.1016/j.knosys.2025.113460_b5) 2023
Lee (10.1016/j.knosys.2025.113460_b25) 2023
10.1016/j.knosys.2025.113460_b14
References_xml – year: 2023
  ident: b25
  article-title: Vector quantized time series generation with a bidirectional prior model
– volume: 47
  start-page: 427
  year: 1992
  end-page: 465
  ident: b2
  article-title: The cross-section of expected stock returns
  publication-title: J. Financ.
– start-page: 1530
  year: 2015
  end-page: 1538
  ident: b9
  article-title: Variational inference with normalizing flows
  publication-title: International Conference on Machine Learning
– volume: 23
  start-page: 389
  year: 1968
  end-page: 416
  ident: b29
  article-title: The performance of mutual funds in the period 1945–1964
  publication-title: J. Financ.
– volume: 66
  start-page: 1047
  year: 2011
  end-page: 1108
  ident: b1
  article-title: Presidential address: Discount rates
  publication-title: J. Financ.
– volume: 30
  year: 2017
  ident: b18
  article-title: Attention is all you need
  publication-title: Adv. Neural Inf. Process. Syst.
– reference: P. Esser, R. Rombach, B. Ommer, Taming transformers for high-resolution image synthesis, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021, pp. 12873–12883.
– year: 2014
  ident: b15
  article-title: Empirical evaluation of gated recurrent neural networks on sequence modeling
– volume: 25
  start-page: 383
  year: 1970
  end-page: 417
  ident: b27
  article-title: Efficient capital markets
  publication-title: J. Financ.
– volume: 52
  start-page: 245
  year: 1992
  end-page: 266
  ident: b28
  article-title: A multi-dynamic-factor model for stock returns
  publication-title: J. Econometrics
– start-page: 1419
  year: 2017
  end-page: 1426
  ident: b10
  article-title: Stock market’s price movement prediction with LSTM neural networks
  publication-title: 2017 International Joint Conference on Neural Networks
– volume: 116
  start-page: 1
  year: 2015
  end-page: 22
  ident: b3
  article-title: A five-factor asset pricing model
  publication-title: J. Financ. Econ.
– year: 1973
  ident: b12
  article-title: Return, Risk and Arbitrage
– start-page: 8821
  year: 2021
  end-page: 8831
  ident: b22
  article-title: Zero-shot text-to-image generation
  publication-title: International Conference on Machine Learning
– year: 2023
  ident: b26
  article-title: Straightening out the straight-through estimator: Overcoming optimization challenges in vector quantized networks
– year: 2017
  ident: b32
  article-title: A unified approach to interpreting model predictions
– reference: H. Chang, H. Zhang, L. Jiang, C. Liu, W.T. Freeman, Maskgit: Masked generative image transformer, in: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2022, pp. 11315–11325.
– year: 2020
  ident: b31
  article-title: Qlib: An ai-oriented quantitative investment platform
– start-page: 4468
  year: 2022
  end-page: 4476
  ident: b6
  article-title: Factorvae: A probabilistic dynamic factor model based on variational autoencoder for predicting cross-sectional stock returns
  publication-title: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36
– year: 2013
  ident: b4
  article-title: Auto-encoding variational bayes
– volume: 29
  year: 2016
  ident: b21
  article-title: Conditional image generation with pixelcnn decoders
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 222
  start-page: 429
  year: 2021
  end-page: 450
  ident: b7
  article-title: Autoencoder asset pricing models
  publication-title: J. Econometrics
– year: 2016
  ident: b30
  article-title: Gaussian error linear units (gelus)
– volume: 19
  start-page: 425
  year: 1964
  end-page: 442
  ident: b11
  article-title: Capital asset prices: A theory of market equilibrium under conditions of risk
  publication-title: J. Financ.
– year: 2023
  ident: b5
  article-title: HireVAE: An online and adaptive factor model based on hierarchical and regime-switch VAE
– reference: T. Li, Z. Liu, Y. Shen, X. Wang, H. Chen, S. Huang, MASTER: Market-Guided Stock Transformer for Stock Price Forecasting, in: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 38, no. 1, 2024, pp. 162–170.
– reference: T. Chen, C. Guestrin, Xgboost: A scalable tree boosting system, in: Proceedings of the 22nd Acm Sigkdd International Conference on Knowledge Discovery and Data Mining, 2016, pp. 785–794.
– reference: C. Lea, M.D. Flynn, R. Vidal, A. Reiter, G.D. Hager, Temporal convolutional networks for action segmentation and detection, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 156–165.
– year: 2017
  ident: b16
  article-title: Graph attention networks
– volume: 52
  start-page: 57
  year: 1997
  end-page: 82
  ident: b13
  article-title: On persistence in mutual fund performance
  publication-title: J. Financ.
– volume: 30
  year: 2017
  ident: b8
  article-title: Neural discrete representation learning
  publication-title: Adv. Neural Inf. Process. Syst.
– year: 2023
  ident: b19
  article-title: Itransformer: Inverted transformers are effective for time series forecasting
– volume: 19
  start-page: 425
  issue: 3
  year: 1964
  ident: 10.1016/j.knosys.2025.113460_b11
  article-title: Capital asset prices: A theory of market equilibrium under conditions of risk
  publication-title: J. Financ.
– volume: 47
  start-page: 427
  issue: 2
  year: 1992
  ident: 10.1016/j.knosys.2025.113460_b2
  article-title: The cross-section of expected stock returns
  publication-title: J. Financ.
– year: 2017
  ident: 10.1016/j.knosys.2025.113460_b16
– year: 2016
  ident: 10.1016/j.knosys.2025.113460_b30
– volume: 30
  year: 2017
  ident: 10.1016/j.knosys.2025.113460_b18
  article-title: Attention is all you need
  publication-title: Adv. Neural Inf. Process. Syst.
– start-page: 8821
  year: 2021
  ident: 10.1016/j.knosys.2025.113460_b22
  article-title: Zero-shot text-to-image generation
– year: 2014
  ident: 10.1016/j.knosys.2025.113460_b15
– ident: 10.1016/j.knosys.2025.113460_b24
  doi: 10.1109/CVPR52688.2022.01103
– year: 2017
  ident: 10.1016/j.knosys.2025.113460_b32
– start-page: 1530
  year: 2015
  ident: 10.1016/j.knosys.2025.113460_b9
  article-title: Variational inference with normalizing flows
– year: 2023
  ident: 10.1016/j.knosys.2025.113460_b19
– volume: 52
  start-page: 57
  issue: 1
  year: 1997
  ident: 10.1016/j.knosys.2025.113460_b13
  article-title: On persistence in mutual fund performance
  publication-title: J. Financ.
  doi: 10.1111/j.1540-6261.1997.tb03808.x
– ident: 10.1016/j.knosys.2025.113460_b23
  doi: 10.1109/CVPR46437.2021.01268
– year: 2023
  ident: 10.1016/j.knosys.2025.113460_b26
– volume: 23
  start-page: 389
  issue: 2
  year: 1968
  ident: 10.1016/j.knosys.2025.113460_b29
  article-title: The performance of mutual funds in the period 1945–1964
  publication-title: J. Financ.
– year: 2020
  ident: 10.1016/j.knosys.2025.113460_b31
– year: 2013
  ident: 10.1016/j.knosys.2025.113460_b4
– volume: 116
  start-page: 1
  issue: 1
  year: 2015
  ident: 10.1016/j.knosys.2025.113460_b3
  article-title: A five-factor asset pricing model
  publication-title: J. Financ. Econ.
  doi: 10.1016/j.jfineco.2014.10.010
– volume: 29
  year: 2016
  ident: 10.1016/j.knosys.2025.113460_b21
  article-title: Conditional image generation with pixelcnn decoders
  publication-title: Adv. Neural Inf. Process. Syst.
– ident: 10.1016/j.knosys.2025.113460_b14
  doi: 10.1109/CVPR.2017.113
– start-page: 4468
  year: 2022
  ident: 10.1016/j.knosys.2025.113460_b6
  article-title: Factorvae: A probabilistic dynamic factor model based on variational autoencoder for predicting cross-sectional stock returns
– volume: 66
  start-page: 1047
  issue: 4
  year: 2011
  ident: 10.1016/j.knosys.2025.113460_b1
  article-title: Presidential address: Discount rates
  publication-title: J. Financ.
  doi: 10.1111/j.1540-6261.2011.01671.x
– volume: 25
  start-page: 383
  issue: 2
  year: 1970
  ident: 10.1016/j.knosys.2025.113460_b27
  article-title: Efficient capital markets
  publication-title: J. Financ.
  doi: 10.2307/2325486
– start-page: 1419
  year: 2017
  ident: 10.1016/j.knosys.2025.113460_b10
  article-title: Stock market’s price movement prediction with LSTM neural networks
– ident: 10.1016/j.knosys.2025.113460_b17
  doi: 10.1145/2939672.2939785
– volume: 52
  start-page: 245
  issue: 1–2
  year: 1992
  ident: 10.1016/j.knosys.2025.113460_b28
  article-title: A multi-dynamic-factor model for stock returns
  publication-title: J. Econometrics
  doi: 10.1016/0304-4076(92)90072-Y
– volume: 222
  start-page: 429
  issue: 1
  year: 2021
  ident: 10.1016/j.knosys.2025.113460_b7
  article-title: Autoencoder asset pricing models
  publication-title: J. Econometrics
  doi: 10.1016/j.jeconom.2020.07.009
– year: 1973
  ident: 10.1016/j.knosys.2025.113460_b12
– year: 2023
  ident: 10.1016/j.knosys.2025.113460_b5
– ident: 10.1016/j.knosys.2025.113460_b20
  doi: 10.1609/aaai.v38i1.27767
– volume: 30
  year: 2017
  ident: 10.1016/j.knosys.2025.113460_b8
  article-title: Neural discrete representation learning
  publication-title: Adv. Neural Inf. Process. Syst.
– year: 2023
  ident: 10.1016/j.knosys.2025.113460_b25
SSID ssj0002218
Score 2.4260125
Snippet This study introduces FactorVQVAE, the first integration of the Vector Quantized Variational Autoencoder (VQVAE) into factor modeling, providing a novel...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 113460
SubjectTerms Autoencoder
Dynamic latent factor model
Portfolio investment
Transformer
Vector quantization
Title FactorVQVAE: Discrete latent factor model via Vector Quantized Variational Autoencoder
URI https://dx.doi.org/10.1016/j.knosys.2025.113460
Volume 318
WOSCitedRecordID wos001473508700001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: ScienceDirect Freedom Collection - Elsevier
  issn: 0950-7051
  databaseCode: AIEXJ
  dateStart: 19950201
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0002218
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1LT9wwELZa6KEXWkoroA_50NvKq42drNe9rdpFLVSrVkC0t8h2HLHQZhFkEe2v7_iVDQWhcuglipzEtjKfxt-MxzMIva8ARhkzjAy1HJKUKUWE1JJUUo8AQoILt6Obf-XT6Wg2E9_CRvulKyfA63p0fS3O_6uooQ2EbY_OPkDcbafQAPcgdLiC2OH6T4LfcxV08u_5eGLN_U9zUAzAjHs_gFXaREzusa-A07uay17u_PY2urNu5r-Bf-ZgPkcX4XjZLGyqyzIE8QYaexA9ccSugmXIB73aF_Ilmqfy58kvq01aT67XvYcG2nqTZYvLwxgXLG10T2D9wRVBMxcyxW_4FAcEZJ901SsL-tUryCRhqS8gcEt3ezfCaf-sXsCs-3aA_ur1m6my_1rC2sDCGLN2WvheCttL4Xt5jNYpzwRo7_Xxl8lsv12wKXVu4Hb28YSlCwO8PZu7GUyHlRw9RxvBnMBjD4NN9MjUL9CzWKoDB829hfIOKj7giAnsMYE9JrDDBAZMYI8J3GICdzCBO5h4iY73JkcfP5NQUoNoIGYNUYYnVaaBtoMZPhiODKNcaFUNKy6MGJSKVcYozjSlRiZGlJlKwf4ymeKZKoVmr9BavajNNsJJaioqlT20J1JllFICyGKmKVjUqkrUDiLxNxXnPnNKcZ94dhCP_7II7M-zugIAcu-Xuw8c6TV6ukLvG7TWXCzNW_REXzXzy4t3AR1_AMjifvc
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=FactorVQVAE%3A+Discrete+latent+factor+model+via+Vector+Quantized+Variational+Autoencoder&rft.jtitle=Knowledge-based+systems&rft.au=Kim%2C+Namhyoung&rft.au=Ock%2C+Seung+Eun&rft.au=Song%2C+Jae+Wook&rft.date=2025-06-07&rft.issn=0950-7051&rft.volume=318&rft.spage=113460&rft_id=info:doi/10.1016%2Fj.knosys.2025.113460&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_knosys_2025_113460
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0950-7051&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0950-7051&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0950-7051&client=summon