Paper: arXiv 2610.09654

Authors: Aashish Bohra, Lokendra Vishwakarm

Abstract

Wavelet-based financial forecasters typically use the transform only to denoise, or reduce it to a single spectral snapshot at the forecast origin, and the convolution that produces the coefficients is usually bilateral, so it can read past the forecast origin. DSTNet instead retains the recent evolution of filter-bank magnitudes as a causal Dynamic Spectral Trajectory, built from seven trailing technical indicators over a twenty-day lookback with a one-sided Morlet-derived filter bank and an explicit burn-in for the left-boundary transient. A factorized Scale-Temporal Spectral Transformer attends along the time and filter-bank axes separately, a learned gate fuses the spectral branch with a CNN-BiLSTM, and horizon-specific gates emit one, three, five, and ten day forecasts in a single pass. We evaluate seven equity indices and gold under a common expanding-window protocol and an untouched one-year hold-out, against nine learned baselines and a random-walk persistence benchmark. Under MAE and MAPE, persistence is the strongest of the ten fixed competitors in 29 of the 32 series-horizon cells and DSTNet is the only model below it in every cell, by 0.7 to 0.9 percent at one day and 3.4 to 4.5 percent at ten days. At one day, paired testing favours DSTNet against the weaker learned baselines but is inconclusive against persistence and the strongest learned forecasters. A downstream allocation diagnostic does not support an equity-timing advantage on any of the seven indices.

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper presents a highly novel approach to financial forecasting using dynamic spectral trajectories and a sophisticated deep learning architecture. It demonstrates strong empirical rigor through extensive evaluation against multiple baselines and a robust hold-out period. The mathematical underpinnings, including causality proofs and complexity analysis, are significant.

Research Flowchart

  flowchart TD
    A[Research Goal: Causal Multi-Horizon Financial Forecasting] --> B{Key Methodology: DSTNet}

    B --> C[Data/Inputs: 7 Equity Indices, Gold, 7 Trailing Technical Indicators (20-day lookback)]
    C --> D[Computational Processes: One-sided Morlet Filter Bank, Dynamic Spectral Trajectory, Factorized Scale-Temporal Spectral Transformer, CNN-BiLSTM, Horizon-specific Gates]
    D --> E[Outcomes: 1, 3, 5, 10 Day Forecasts]

    E --> F{Key Findings: DSTNet consistently outperforms baselines and persistence by 0.7-4.5% MAE/MAPE. No equity-timing advantage found.}

Cited References (206)

  1. Time Series Analysis: Forecasting and Control (1976)
  2. Forecasting seasonals and trends by exponentially weighted moving averages (2004)
  3. Long short-term memory (1997)
  4. A theory for multiresolution signal decomposition: the wavelet representation (1989)
  5. Ten Lectures on Wavelets (1992)
  6. Ideal spatial adaptation by wavelet shrinkage (1994)
  7. De-noising by soft-thresholding (1995)
  8. A practical guide to wavelet analysis (1998)
and 198 more
  1. Wavelet neural networks for nonlinear time series analysis (2010)
  2. Forecasting across time series databases using recurrent neural networks on groups of similar series: A clustering approach (2020)
  3. Financial time series forecasting with deep learning: A systematic literature review: 2005–2019 (2020)
  4. Elements of Information Theory (2006)
  5. Approximation by superpositions of a sigmoidal function (1989)
  6. Multilayer feedforward networks are universal approximators (1989)
  7. The jackknife and the bootstrap for general stationary observations (1989)
  8. XGBoost : A scalable tree boosting system (2016)
  9. Attention is all you need (2017)
  10. Informer : Beyond efficient transformer for long sequence time-series forecasting (2021)
  11. Autoformer : Decomposition transformers with auto-correlation for long-term series forecasting (2021)
  12. Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting (2022)
  13. A time series is worth 64 words: Long-term forecasting with transformers (2023)
  14. iTransformer : Inverted transformers are effective for time series forecasting (2024)
  15. TimeMixer : Decomposable multiscale mixing for time series forecasting (2024)
  16. MASTER : Market-guided stock transformer for stock price forecasting (2024)
  17. N-BEATS : Neural basis expansion analysis for interpretable time series forecasting (2020)
  18. Temporal fusion transformers for interpretable multi-horizon time series forecasting (2021)
  19. Are transformers effective for time series forecasting? (2023)
  20. Linformer: Self-attention with linear complexity (2020) — link
  21. Rethinking attention with Performers (2021)
  22. Bridging short- and long-term dependencies: A CNN -Transformer hybrid for financial time series forecasting (2025) — link
  23. BiCuDNNLSTM-1dCNN: a hybrid deep learning predictive model for stock price prediction (2022)
  24. Stock market index prediction using deep transformer model (2022)
  25. StockMixer : A Simple yet Strong MLP-based Architecture for Stock Price Forecasting (2024)
  26. FactorVAE : A Probabilistic Dynamic Factor Model based on VAE for Predicting Cross-Sectional Stock Returns (2022)
  27. Diffusion Variational Autoencoder for Tackling Stochasticity in Multi-Step Regression Stock Price Prediction ( D-Va ) (2023)
  28. MERA : Mixture-of-Experts with Retrieval-Augmented representation for stock prediction (2025) — link
  29. Strictly Proper Scoring Rules, Prediction, and Estimation (2007)
  30. A Test for Superior Predictive Ability (2005)
  31. Multifactor prediction model for stock market analysis based on deep learning techniques (2025)
  32. A novel interval dual convolutional neural network method for interval-valued stock price prediction (2024)
  33. " Why should i trust you?" Explaining the predictions of any classifier (2016)
  34. A graph-based CNN-LSTM stock price prediction algorithm with leading indicators (2023)
  35. A novel hybrid model based on recurrent neural networks for stock market timing (2020)
  36. An improved DenseNet model for Prediction of Stock Market using Stock Technical Indicators (2023)
  37. A CNN-LSTM Stock Prediction Model Based on Genetic Algorithm Optimization (2023)
  38. A Deep Learning based Approach to Stock Market Price Prediction using Technical indicators (2023)
  39. A hybrid two-stage financial stock forecasting algorithm based on clustering and ensemble learning (2020)
  40. An adaptive feature selection schema using improved technical indicators for predicting stock price movements (2022)
  41. Neural network systems with an integrated coefficient of variation-based feature selection for stock price and trend prediction (2023)
  42. ModAugNet: A new forecasting framework for stock market index value with an overfitting prevention LSTM module and a prediction LSTM module (2018)
  43. BiCuDNNLSTM-1dCNN—A hybrid deep learning-based predictive model for stock price prediction (2022)
  44. Stock market index prediction using deep Transformer model (2022)
  45. Stock market index prediction based on reservoir computing models (2021)
  46. StockNet—GRU based stock index prediction (2022)
  47. Stock market forecasting with super-high dimensional time-series data using ConvLSTM, trend sampling, and specialized data augmentation (2020)
  48. McVCsB: A new hybrid deep learning network for stock index prediction (2023)
  49. Forecasting stock market returns using deep learning and time series techniques: a comparative and empirical study using technical indicators (2024)
  50. A hybrid model integrating deep learning with investor sentiment analysis for stock price prediction (2021)
  51. Two-Stage Deep Ensemble Paradigm Based on Optimal Multi-scale Decomposition and Multi-factor Analysis for Stock Price Prediction (2024)
  52. Predicting stock price using two-stage machine learning techniques (2021)
  53. Forecasting multistep daily stock prices for long-term investment decisions: A study of deep learning models on global indices (2024)
  54. Deep Learning-based Integrated Framework for stock price movement prediction (2023)
  55. A novel hybrid model for stock price forecasting integrating encoder forest and informer (2023)
  56. Predicting stock market index using fusion of machine learning techniques (2015)
  57. Predicting stock and stock price index movement using trend deterministic data preparation and machine learning techniques (2015)
  58. The random character of stock market prices (1964)
  59. Efficient capital markets: A review of theory and empirical work (1970)
  60. Efficient market hypothesis and forecasting (2004)
  61. Deep learning for stock prediction using numerical and textual information (2016)
  62. Evaluating multiple classifiers for stock price direction prediction (2015)
  63. A deep learning framework for financial time series using stacked autoencoders and long-short term memory (2017)
  64. Computational intelligence and financial markets: A survey and future directions (2016)
  65. Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks (1990)
  66. Stock market trend prediction using ARIMA-based neural networks (1996)
  67. A hybrid modeling approach for forecasting the volatility of S&P 500 index return (2012)
  68. Deep learning networks for stock market analysis and prediction: Methodology, data representations, and case studies (2017)
  69. Stock market index prediction using artificial neural network (2016)
  70. Financial time series prediction using hybrids of chaos theory, multi-layer perceptron and multi-objective evolutionary algorithms (2017)
  71. Learning to diagnose with LSTM recurrent neural networks (2015) — link
  72. A LSTM-based method for stock returns prediction: A case study of China stock market (2015)
  73. Stock closing price prediction based on sentiment analysis and LSTM (2020)
  74. Investment strategies applied to the Brazilian stock market: a methodology based on sentiment analysis with deep learning (2021)
  75. Application of deep reinforcement learning in stock trading strategies and stock forecasting (2020)
  76. EAN: Event attention network for stock price trend prediction based on sentimental embedding (2019)
  77. Stock market forecast using multivariate analysis with bidirectional and stacked (LSTM, GRU) (2018)
  78. Hybrid deep learning model for stock price prediction (2018)
  79. Evaluation of bidirectional LSTM for short-and long-term stock market prediction (2018)
  80. A comparative analysis of forecasting financial time series using arima, lstm, and bilstm (2019) — link
  81. Stock price prediction using LSTM, RNN and CNN-sliding window model (2017)
  82. Twitter dataset and evaluation of transformers for Turkish sentiment analysis (2021)
  83. An improved deep learning model for predicting stock market price time series (2020)
  84. Application of artificial neural network for the prediction of stock market returns: The case of the Japanese stock market (2016)
  85. Stock price prediction using support vector regression on daily and up to the minute prices (2018)
  86. Stock market’s price movement prediction with LSTM neural networks (2017)
  87. Novel Stock Crisis Prediction Technique—A Study on Indian Stock Market (2021)
  88. CNNpred: CNN-based stock market prediction using a diverse set of variables (2019)
  89. LSTM-based Deep Learning Model for Stock Prediction and Predictive Optimization Model (2021)
  90. Artificial neural networks approach to the forecast of stock market price movements (2016)
  91. Applications of deep learning in stock market prediction: recent progress (2021)
  92. Stock price prediction on daily stock data using deep neural networks (2018)
  93. A CNN-LSTM-based model to forecast stock prices (2020)
  94. Deep learning-based feature engineering for stock price movement prediction (2019)
  95. Novel deep learning model with CNN and bi-directional LSTM for improved stock market index prediction (2019)
  96. Forecasting stock prices with a feature fusion LSTM-CNN model using different representations of the same data (2019)
  97. Stock prediction using convolutional neural network (2018)
  98. Which artificial intelligence algorithm better predicts the Chinese stock market? (2018)
  99. Evaluation of co-evolutionary neural network architectures for time series prediction with mobile application in finance (2016)
  100. Improving stock closing price prediction using recurrent neural network and technical indicators (2018)
  101. Convolutional neural network for stock trading using technical indicators (2022)
  102. Stock price prediction based on technical indicators with soft computing models (2020)
  103. Stock closing price prediction using machine learning techniques (2020)
  104. LSTM based decision support system for swing trading in stock market (2022)
  105. NSE stock market prediction using deep-learning models (2018)
  106. Time series prediction of stock price using deep belief networks with intrinsic plasticity (2017)
  107. Deep stock ranker: A LSTM neural network model for stock selection (2018)
  108. Stock prediction using deep learning (2017)
  109. An innovative neural network approach for stock market prediction (2020)
  110. Portfolio formation with preselection using deep learning from long-term financial data (2020)
  111. Stock market forecasting using machine learning models (2019)
  112. Using Deep Learning Neural Networks and Stacking Ensemble Learning to Predict CSI 300 Index (2022)
  113. A CNN-BiLSTM-AM method for stock price prediction (2021)
  114. Stock market prediction based on generative adversarial network (2019)
  115. Decomposition techniques and long short term memory model with black widow optimization for stock price prediction (2023)
  116. An Improved hybrid Salp Swarm Optimization and African Vulture Optimization Algorithm for Global Optimization Problems and Its Applications in Stock Market Prediction (2023)
  117. Stock market forecasting using a multi-task approach integrating long short-term memory and the random forest framework (2022)
  118. MG-Conv: A spatiotemporal multi-graph convolutional neural network for stock market index trend prediction (2022)
  119. Gradient-based learning applied to document recognition (1998)
  120. A CNN–LSTM model for gold price time-series forecasting (2020)
  121. A novel CNN-GRU-based hybrid approach for short-term residential load forecasting (2020)
  122. Stock price forecasting model based on modified convolution neural network and financial time series analysis (2019)
  123. A stock market trading system using deep neural network (2017)
  124. Forecasting the volatility of stock price index: A hybrid model integrating LSTM with multiple GARCH-type models (2018)
  125. A smart system for short-term price prediction using time series models (2019)
  126. Stock price prediction using the ARIMA model (2014)
  127. Forecasting volatility of oil price using an artificial neural network-GARCH model (2016)
  128. A comprehensive survey on deep neural networks for stock market: The need, challenges, and future directions (2021)
  129. A comprehensive survey on transfer learning (2020)
  130. A survey of transfer learning (2016)
  131. A novel transfer learning framework for time series forecasting (2018)
  132. Implementing transfer learning across different datasets for time series forecasting (2021)
  133. Stock price prediction with optimized deep LSTM network with artificial rabbits optimization algorithm (2023)
  134. TCLN: A Transformer-based Conv-LSTM network for multivariate time series forecasting (2023)
  135. Three stage fusion for effective time series forecasting using Bi-LSTM-ARIMA and improved DE-ABC algorithm (2022)
  136. GARCH models: structure, statistical inference and financial applications (2019)
  137. Multi-model generative adversarial network hybrid prediction algorithm (MMGAN-HPA) for stock market prices prediction (2022)
  138. Stock market forecasting using DRAGAN and feature matching — → 2301.05693
  139. Fine-tuned support vector regression model for stock predictions (2023)
  140. A turning point prediction method of stock price based on RVFL-GMDH and chaotic time series analysis (2021)
  141. A two-stage model for stock price prediction based on variational mode decomposition and ensemble machine learning method (2024)
  142. Stock market forecasting using deep learning with long short-term memory and gated recurrent unit (2024)
  143. Deep learning framework for stock price prediction using long short-term memory (2024)
  144. Predictive analysis of stochastic stock pattern utilizing fractional order dynamics and heteroscedastic with a radial neural network framework (2024)
  145. A hierarchical deep model integrating economic facts for stock movement prediction (2024)
  146. Prediction based mean-value-at-risk portfolio optimization using machine learning regression algorithms for multi-national stock markets (2023)
  147. Multi-sentiment fusion for stock price crash risk prediction using an interpretable ensemble learning method (2024)
  148. Optimal execution of portfolio transactions (2001)
  149. The Statistics of Sharpe Ratios (2002) — link
  150. Comparing predictive accuracy (2002)
  151. BiMT-TCN: A cutting-edge hybrid model for enhanced stock price prediction (2025)
  152. Galformer: a transformer with generative decoding and a hybrid loss function for multi-step stock market index prediction (2024)
  153. Deep learning for time series forecasting: a survey (2025)
  154. Stock market trend prediction using deep neural network via chart analysis: a practical method or a myth? (2025)
  155. Stock price prediction using deep learning and frequency decomposition (2021)
  156. Information fusion-based genetic algorithm with long short-term memory for stock price and trend prediction (2022)
  157. Predicting stock market index using LSTM (2022)
  158. Enhancing stock market Forecasting: A hybrid model for accurate prediction of S&P 500 and CSI 300 future prices (2025)
  159. An intelligent framework based on optimized variational mode decomposition and temporal convolutional network: Applications to stock index multi-step forecasting (2025)
  160. Predicting the highest and lowest stock price indices: A combined BiLSTM-SAM-TCN deep learning model based on re-decomposition (2024)
  161. The importance of data noise reduction–wavelet transformation in stock price forecasting with Bidirectional Long Short-Term Memory network (2026)
  162. Enhanced financial market forecasting using a hybrid deep learning prediction model with encoder-decoder architecture (2026)
  163. Fractals and scaling in finance: Discontinuity, concentration, risk. Selecta volume E (2013)
  164. Multifractality in asset returns: theory and evidence (2002)
  165. Selection of estimation window in the presence of breaks (2007)
  166. Advances in forecasting under instability (2013)
  167. A guide to wavelets for economists (2007)
  168. Adapting to unknown smoothness via wavelet shrinkage (1995)
  169. Unbalanced Haar technique for nonparametric function estimation (2007)
  170. Time series analysis: forecasting and control (2015)
  171. An introduction to wavelets and other filtering methods in finance and economics (2002)
  172. A new approach to the economic analysis of nonstationary time series and the business cycle (1989)
  173. Regime changes and financial markets (2012)
  174. A simple nonparametric test of predictive performance (1992)
  175. Asset allocation under multivariate regime switching (2007)
  176. The adaptive markets hypothesis: Market efficiency from an evolutionary perspective (2004)
  177. Robust statistical procedures (1996)
  178. The statistics of Sharpe ratios (2002)
  179. Enhancing financial time series forecasting: a comparative study of discrete wavelet transform and LSTM models for selected global indices: K. Zi 'o \l kowski, P. Wieczy 'n ski (2026)
  180. Frequency-domain MLPs are more effective learners in time series forecasting (2023)
  181. Filternet: Harnessing frequency filters for time series forecasting (2024)
  182. Nhits: Neural hierarchical interpolation for time series forecasting (2023)
  183. Multi-resolution time-series transformer for long-term forecasting (2024)
  184. Pathformer: Multi-scale transformers with adaptive pathways for time series forecasting (2024)
  185. WaveletMixer: a multi-resolution wavelets based MLP-mixer for multivariate long-term time series forecasting (2025)
  186. Non-stationary transformers: Exploring the stationarity in time series forecasting (2022)
  187. U-mixer: An unet-mixer architecture with stationarity correction for time series forecasting (2024)
  188. Deep learning with long short-term memory networks for financial market predictions (2018)
  189. Financial time series forecasting with multi-modality graph neural network (2022)
  190. Empirical asset pricing via machine learning (2020)
  191. A note on the validity of cross-validation for evaluating autoregressive time series prediction (2018)
  192. Forecast evaluation for data scientists: common pitfalls and best practices: H. Hewamalage et al. (2023)
  193. Is gold a safe haven? International evidence (2010)
  194. Testing the equality of prediction mean squared errors (1997)
  195. VertiFuseX: Generalizable Financial Forecasting via Multi-Stream Temporal Fusion (2026) — link
  196. WaVeFuse: Regime-Adaptive Equity Index Forecasting via Channel-Wise Wavelet Denoising and Vertical Attention Fusion (2026) — link
  197. A simple approximate long-memory model of realized volatility (2009)
  198. International comovement of stock market returns: A wavelet analysis (2009)

1 of 206 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.