---
title: Comparative Analysis of Financial Data Techniques
url: https://www.ml-quant.com/papers/ssrn/5268353/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
identifier: SSRN 5268353
source_url: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5268353
featured: 2025-05-30
citations: unknown
topic: Econometrics & Forecasting
---


# Comparative Analysis of Financial Data Techniques

The study contrasts the traditional method of calculating logarithmic returns with the fractional differencing method in data preparation for machine learning models, finding that fractional differentiation methods enhance predictive model forecasting performance.

- Source: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5268353
- Identifier: SSRN 5268353
- Released: 2025-05-25
- First featured: Quant Letter No. 99 (2025-05-30): https://www.ml-quant.com/issues/2025-05-30/
- Citations (Semantic Scholar): not tracked
- Published in: not yet
- Topic: Econometrics & Forecasting

## Related

- [TKAN: Temporal Kolmogorov-Arnold Networks](https://www.ml-quant.com/papers/ssrn/4825654/): The article presents Temporal Kolomogorov-Arnold Networks (TKANs), a new neural network design that merges the benefits of Recurrent Neural Networks and Long Short-Term Memory for improved multistep time series forecasting.
- [Robust agents learn causal world models](https://www.ml-quant.com/papers/arxiv/2402.10877/): The research suggests that intelligent agents must learn an approximate causal model to generalize to new domains, impacting fields like transfer learning and causal inference.
- [Hybrid Models for Financial Forecasting: Combining Econometric, Machine Learning, and Deep Learning Models](https://www.ml-quant.com/papers/arxiv/2505.19617/): ARIMA with SVM/LSTM: A study using econometric models, machine learning, and deep learning to predict financial trends for the S&P 500 and Bitcoin emphasizes the importance of well-constructed hybrid models for profitable trading strategies.
- [Deep Learning Enhanced Multivariate GARCH](https://www.ml-quant.com/papers/arxiv/2506.02796/): A new volatility modeling framework, LSTM-BEKK, is introduced, integrating deep learning into multivariate GARCH processes for improved robustness and forecasting in financial return data.
- [CATS: Clustering-Aggregated and Time Series for Business Customer Purchase Intention Prediction](https://www.ml-quant.com/papers/arxiv/2505.13558/): A new model, Clustering and Attention mechanism GRU (CAGRU), has been proposed for predicting customer buying intentions, using customer characteristics and a GRU neural network to provide more accurate predictions across different customer groups.
- [A Set-Sequence Model for Time Series](https://www.ml-quant.com/papers/arxiv/2505.11243/): The article presents a Set-Sequence model for financial predictions, eliminating the need for manually created features, learning a shared summary at each period and predicting outcomes, performing better than benchmarks on stock return prediction and mortgage behavior tasks.