Paper: arXiv 2512.15113

Authors: Mohit Beniwal

Abstract

Long-term price forecasting remains a formidable challenge due to the inherent uncertainty over the long term, despite some success in short-term predictions. Nonetheless, accurate long-term forecasts are essential for high-net-worth individuals, institutional investors, and traders. The proposed improved genetic algorithm-optimized support vector regression (IGA-SVR) model is specifically designed for long-term price prediction of global indices. The performance of the IGA-SVR model is rigorously evaluated and compared against the state-of-the-art baseline models, the Long Short-Term Memory (LSTM), and the forward-validating genetic algorithm optimized support vector regression (OGA-SVR). Extensive testing was conducted on the five global indices, namely Nifty, Dow Jones Industrial Average (DJI), DAX Performance Index (DAX), Nikkei 225 (N225), and Shanghai Stock Exchange Composite Index (SSE) from 2021 to 2024 of daily price prediction up to a year. Overall, the proposed IGA-SVR model achieved a reduction in MAPE by 19.87% compared to LSTM and 50.03% compared to OGA-SVR, demonstrating its superior performance in long-term daily price forecasting of global indices. Further, the execution time for LSTM was approximately 20 times higher than that of IGA-SVR, highlighting the high accuracy and computational efficiency of the proposed model. The genetic algorithm selects the optimal hyperparameters of SVR by minimizing the arithmetic mean of the Mean Absolute Percentage Error (MAPE) calculated over the full training dataset and the most recent five years of training data. This purposefully designed training methodology adjusts for recent trends while retaining long-term trend information, thereby offering enhanced generalization compared to the LSTM and rolling-forward validation approach employed by OGA-SVR, which forgets long-term trends and suffers from recency bias.

Complexity vs Empirical Score

  • Math Complexity: 5.0/10
  • Empirical Rigor: 7.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper applies standard machine learning models (SVR, LSTM) with a custom genetic algorithm for hyperparameter tuning, which involves moderate mathematical complexity, while the empirical rigor is high due to extensive backtesting on multiple global indices over several years with clear performance metrics and computational efficiency comparisons.

Research Flowchart

  flowchart TD
  A["Research Goal: <br>Robust Long-Term Price Forecasting<br>for Global Stock Indices"] --> B["Data: 5 Global Indices<br>(Nifty, DJI, DAX, N225, SSE)<br>2021-2024"]
  B --> C["Methodology: <br>IGA-SVR Model"]
  C --> D["Genetic Algorithm Optimization<br>Minimizes weighted MAPE<br>(Recent + Long-term Data)"]
  D --> E["Training SVR<br>with Optimal Hyperparameters"]
  E --> F["Forecasting & Evaluation<br>vs. LSTM & OGA-SVR"]
  F --> G["Outcome: <br>Superior Performance"]
  G --> H["IGA-SVR Outcomes:<br>- 19.87% lower MAPE vs LSTM<br>- 50.03% lower MAPE vs OGA-SVR<br>- 20x faster execution than LSTM"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes an improved genetic algorithm‑optimized support vector regression (IGA‑SVR) model for long‑term price prediction of global stock indices.

Source sentence
The proposed improved genetic algorithm-optimized support vector regression (IGA-SVR) model is specifically designed for long-term price prediction of global indices.

Assumes 5

  • The genetic algorithm selects SVR hyperparameters by minimizing MAPE over the full training set and the most recent five years. model
    Source sentence
    The genetic algorithm selects the optimal hyperparameters of SVR by minimizing the arithmetic mean of the Mean Absolute Percentage Error (MAPE) calculated over the full training dataset and the most recent five years of training data.
  • Recent five years of close prices are given higher weight in training. model
    Source sentence
    The Improved Genetic Algorithm-Optimized Support Vector Regression (IGA-SVR) model is trained on the entire training dataset, with higher weightage assigned to the most recent five years' close prices.
  • Yahoo Finance data are accurate and complete for the indices. data
    Source sentence
    The data was acquired from Yahoo Finance.
  • The number of GA generations is limited to 30 to control computation. model
    Source sentence
    Based on the literature, the number of generations is restricted to 30 [95].
Show 1 more
  • SVR hyperparameters are bounded by logical restrictions to avoid excessive computation. model
    Source sentence
    Therefore, logical restrictions are imposed on the hyperparameters of the SVR model and the number of generations.

Data

Backtest sample: global stock indices · equities · daily

  • Indices data for Nifty, DJI, DAX, N225, SSE · Yahoo Finance market-prices public
    April 2008 to December 2024 · Top five GDP countries (US, China, Germany, Japan, India)
    Source sentence
    The data was acquired from Yahoo Finance. The indices data for the top five GDP countries

Recognised sources: Yahoo Finance

Costs

Not applicable The paper reports no trading or execution result.

Finds 2

  • IGA‑SVR reduces MAPE by 19.87% compared to LSTM and 50.03% compared to OGA‑SVR for long‑term daily price forecasting of global indices. predictive-accuracy
    Source sentence
    Overall, the proposed IGA-SVR model achieved a reduction in MAPE by 19.87% compared to LSTM and 50.03% compared to OGA-SVR, demonstrating its superior performance in long-term daily price forecasting of global indices.
  • Execution time of LSTM is approximately 20 times higher than that of IGA‑SVR. performance
    Source sentence
    The execution time for LSTM was approximately 20 times higher than that of IGA-SVR, highlighting the high accuracy and computational efficiency of the proposed model.

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 21 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.