Shape-based Representation and Abstraction of Time Series Data along with a Dynamic Time Shape Wrapping as a Dissimilarity Measure

Gmati, Fatma Ezzahra; Chakhar, Salem; Chaari, Wided Lejouad; Xu, Mark

doi:10.23919/icac50006.2021.9594127

Cited by 1 publication

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Dynamic Time Warping (DTW) technique, which is a non-parametric method [19] based on dynamic programming techniques [20], allows for the calculation of the distance between two time series and the ability to handle complex time series [21]. We adopted this technique to filter out similar patterns between specific and shared representations.…”

Section: Introductionmentioning

confidence: 99%

Cross-Domain Disentanglement: A Novel Approach to Financial Market Prediction

Abdulsahib,

Ghaderi

2024

IEEE Access

View full text Add to dashboard Cite

Profit maximization and risk mitigation require good financial market predictions. Financial markets have a correlated nature, which means that there are some shared patterns between them. Therefore, learning about one market might help understand the behavior of others. End-to-end training techniques have proven successful in financial markets, but they have flaws, such as picking up noise and failing to account for complicated relationships across markets. We present a promising model for predicting financial markets using the correlation between the two markets, which draws inspiration from the recent progress in disentanglement learning. This model learns to disentangle representations of features shared between markets from specific representations, and removes features that cause interference. We utilized a dilated convolutional neural network as an encoder to extract features while using self-attention and cross-attention to capture specifics and shared patterns. Our model uses Dynamic Time Warping (DTW) to minimize the similarity between specific and shared patterns, which has been used along with the Mean Square Error (MSE) to determine the optimal balance between minimizing the similarity and prediction accuracy. We conducted our experiments using datasets that included the closing prices of Apple, Samsung, Bitcoin, Ethereum, Meta platforms, and the X platform. Spearman's rank correlation coefficient was used to evaluate the disentanglement by describing the relationship between the extracted representations. The findings confirm that our model surpasses state-of-the-art approaches in prediction error, financial risk assessment, correlation evolution, and prediction net curves, thereby giving market participants more trust in their decisions.INDEX TERMS Attention mechanisms, cross domain, disentangled representation, financial market, time series prediction.

show abstract

Section: Introductionmentioning

confidence: 99%