Paper: arXiv 2310.16845

Authors: Veli Safak

Abstract

Kardemir Karabuk Iron Steel Industry Trade & Co. Inc., ranked as the 24th largest industrial company in Turkey, offers three distinct stocks listed on the Borsa Istanbul: KRDMA, KRDMB, and KRDMD. These stocks, sharing the sole difference in voting power, have exhibited significant price divergence over an extended period. This paper conducts an in-depth analysis of the divergence patterns observed in these three stock prices from January 2001 to July 2023. Additionally, it introduces an innovative training set selection rule tailored for LSTM models, incorporating a rolling training set, and demonstrates its significant predictive superiority over the conventional use of LSTM models with large training sets. Despite their strong correlation, the study found no compelling evidence supporting the efficiency of dual-class stocks as predictors of each other’s performance.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 7.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper applies advanced mathematical techniques like continuous wavelet transform (CWT) and LSTM architecture with detailed mathematical formulations, showing significant math density. Empirically, it uses a long historical dataset (2001-2023), introduces a novel rolling training set methodology for LSTM, and compares approaches with clear quantitative evaluation, indicating strong backtest-readiness.

Research Flowchart

  flowchart TD
  A["Research Goal: Assess if dual-class stocks are effective predictors of each other"] --> B["Data Acquisition<br>3 Stocks KRDMA/KRDMB/KRDMD<br>Jan 2001 - Jul 2023"]
  B --> C["Methodology: LSTM Models with Rolling vs. Fixed Training Sets"]
  C --> D{"Computational Analysis"}
  D --> E["Price Divergence Analysis"]
  D --> F["Predictive Accuracy Comparison"]
  E --> G["Key Findings: Significant Price Divergence"]
  F --> H["Key Findings: Rolling Set Superior to Fixed Set"]
  H --> I["Key Findings: No Evidence of Dual-Class Stock Predictive Efficiency"]
  G --> I

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 a rolling training set selection rule for LSTM models to improve price prediction of dual‑class stocks.

Source sentence
introduces an innovative training set selection rule tailored for LSTM models, incorporating a rolling training set

Assumes 2

  • Daily mid‑prices will consistently remain below 100 Turkish Lira, justifying the scaling transformation. data
    Source sentence
    This scaling choice is based on the assumption that the daily mid‑prices will consistently remain below 100 Turkish Lira (TRY).
  • Min‑max scaling is inappropriate because the range of validation‑set prices is unknown. model
    Source sentence
    the range of these mid‑prices is considered unknown since they are the very values we aim to forecast.

Data

  • Daily mid‑prices of KRDMA, KRDMB, KRDMD · Borsa Istanbul market-prices public
    January 2001 to July 2023 · Three dual‑class stocks of Kardemir Karabuk Iron Steel Industry
    Source sentence
    These stocks, sharing the sole difference in voting power, have exhibited significant price divergence over an extended period. This paper conducts an in‑depth analysis of the divergence patterns observed in these three stock prices from January 2001 to July 2023.

Costs

Not applicable The paper reports no trading or execution result.

Finds 3

  • The rolling training set approach yields significantly better predictive performance than conventional large training sets. predictive-accuracy
    Source sentence
    demonstrates its significant predictive superiority over the conventional use of LSTM models with large training sets.
  • Dual‑class stocks do not serve as reliable predictors of each other's future price movements. negative
    Source sentence
    the study found no compelling evidence supporting the efficiency of dual‑class stocks as predictors of each other's performance.
  • Shorter training periods outperform longer ones for LSTM forecasting of these stocks. performance
    Source sentence
    substantial empirical evidence supporting the practice of favoring shorter training periods over extended ones

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 19 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.