Paper: arXiv 2307.13422
Authors: Ivan Letteri
Abstract
Volatility-based trading strategies have attracted a lot of attention in financial markets due to their ability to capture opportunities for profit from market dynamics. In this article, we propose a new volatility-based trading strategy that combines statistical analysis with machine learning techniques to forecast stock markets trend. The method consists of several steps including, data exploration, correlation and autocorrelation analysis, technical indicator use, application of hypothesis tests and statistical models, and use of variable selection algorithms. In particular, we use the k-means++ clustering algorithm to group the mean volatility of the nine largest stocks in the NYSE and NasdaqGS markets. The resulting clusters are the basis for identifying relationships between stocks based on their volatility behaviour. Next, we use the Granger Causality Test on the clustered dataset with mid-volatility to determine the predictive power of a stock over another stock. By identifying stocks with strong predictive relationships, we establish a trading strategy in which the stock acting as a reliable predictor becomes a trend indicator to determine the buy, sell, and hold of target stock trades. Through extensive backtesting and performance evaluation, we find the reliability and robustness of our volatility-based trading strategy. The results suggest that our approach effectively captures profitable trading opportunities by leveraging the predictive power of volatility clusters, and Granger causality relationships between stocks. The proposed strategy offers valuable insights and practical implications to investors and market participants who seek to improve their trading decisions and capitalize on market trends. It provides valuable insights and practical implications for market participants looking to.
Complexity vs Empirical Score
- Math Complexity: 7.0/10
- Empirical Rigor: 6.0/10 (capped from 7.5: transaction costs only partially modelled; no out-of-sample test; see penalty flags)
- Quadrant: Holy Grail — high math complexity, high empirical rigor (effective empirical rigor capped from 7.5)
Why this score: The paper employs advanced mathematical concepts including volatility estimators (Parkinson, Garman-Klass, Rogers-Satchell, Yang-Zhang), Granger Causality tests, and k-means clustering, which require a solid mathematical foundation to understand and implement. Empirically, it is highly rigorous, featuring a detailed backtesting framework with specific metrics (Maximum Drawdown, Sharpe, Sortino, Calmar ratios), uses real market data from NYSE and NasdaqGS, and includes a dedicated backtesting module (AitaBT), making it highly actionable for implementation.
Research Flowchart
flowchart TD
A["Research Goal<br>Forecast Stock Trend via<br>Volatility-based Trading Strategy"] --> B
subgraph B ["Methodology"]
direction TB
B1["Data Collection<br>NYSE/NasdaqGS Stocks"] --> B2["Clustering<br>K-means++ on Volatility"]
B2 --> B3["Granger Causality Test<br>Identify Predictive Pairs"]
B3 --> B4["Trading Signal Logic<br>Buy/Sell based on Predictor"]
end
B --> C["Backtesting & Evaluation<br>Performance Metrics"]
C --> D{"Key Findings"}
D --> D1["Strategy Reliability & Robustness"]
D --> D2["Profitable Opportunities Captured"]
D --> D3["Volatility Clusters are Effective Predictors"]
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 new volatility-based trading strategy that combines statistical analysis with machine learning techniques to forecast stock market trends.
Source sentence
we propose a new volatility-based trading strategy that combines statistical analysis with machine learning techniques to forecast stock markets trend.
Assumes 6
- The true volatility of the asset is proportional to the logarithm of the high/low price ratio. model
Source sentence
It is derived from the assumption that the true volatility of the asset is proportional to the logarithm (ln) of the ratio of the high x(h) t and low x(l) t prices of N observations.
- Price movements are log‑normally distributed for the Garman‑Klass estimator. model
Source sentence
This estimator assumes that price movements are log-normally distributed, which may not always be the case in practice.
- The range of prices within the interval is a good proxy for volatility in the Rogers‑Satchell estimator. model
Source sentence
RS assumes that the range of prices within the interval is a good proxy for the volatility of the asset, additionally, the estimator may be sensitive to outliers and extreme price movements.
- Granger causality significance level is set at α = 0.05. statistical
Source sentence
Step 2. F-statistic comparison with the critical value from the F-distribution where the significance level has α = 0.05.
Show 2 more
- Trading starts with an initial budget of $1000 per stock and uses a compounded budgeting mode. execution
Source sentence
... the managing of the budget is set in compounded mode, so the full amount is reused for each trade.
- Only the time window from 1st May 2020 to 1st May 2023 is used to avoid the March 2020 anomaly. data
Source sentence
we decide to use only the time window in the period after instead to remove it, starting from 1st May 2020 to 1st May 2023.
Data
- Historical Volatility Clustering dataset market-prices unknown
Source sentence
... we use this interval as the dataset, and then from the intermediate cluster, the candidate assets selected are TSLA with the highest, AMZN and META in the middle, with QCOM and IBM with the lowest values, respectively.
Costs
Partial
fixed commission of $9 per trade
Source sentence
considering the fixed commission of 9$ per trade.
Finds 3
- The strategy generated a total gain of $231.77 in 40 days, increasing the portfolio to $3,231.77 (7.725% return). performance
Source sentence
resulted in a total gain of 231.77$ in 40 days of market opening
- Trades of META bought following the AMZN trend yielded a PnL of 1.281% and a return of 9.721%, outperforming a buy‑and‑hold return of 6.684%. performance
Source sentence
The trades of META bought following the AMZN trend resulted in a Profit and Loss (PnL) of 1.281%, with a return of 9.721%. This return outperforms the B&H strategy, which would have yielded a return of 6.684%.
- Trades of QCOM bought following the META trend yielded a PnL of 2.774% and a return of 12.866%, beating the B&H return of 9.235%. performance
Source sentence
trades of QCOM bought following the META trend showed a PnL of 2.774%, with a return of 12.866% compared to the B&H return of 9.235%.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 17 related papers). Coverage is partial, so this is neither replication nor refutation.
3 papers in the archive
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.