Testing leverage-based trading strategies under an adaptive-expectations agent-based model

Commodity exchange-traded funds (ETFs) are a significant part of the rapidly growing ETF market. They have become popular in recent years as they provide investors access to a great variety of commodities, ranging from precious metals to building materials, and from oil and gas to agricultural products. In this article, we analyze the tracking performance of commodity leveraged ETFs and discuss the associated trading strategies. It is known that leveraged ETF returns typically deviate from their tracking target over longer holding horizons due to the so-called volatility decay. This motivates us to construct a benchmark process that accounts for the volatility decay, and use it to examine the tracking performance of commodity leveraged ETFs. From empirical data, we find that many commodity leveraged ETFs underperform significantly against the benchmark, and we quantify such a discrepancy via the novel idea of realized effective fee. Finally, we consider a number of trading strategies and examine their performance by backtesting with historical price data.

The performance of volatility-based trading strategies depends, among other factors, on the asset selection and the associated risk preference. For this study, we conducted a representative survey for Germany to determine the asset preferences of individuals with lower-risk and higher-risk preference. These two types of behavioral modified asset allocations (lower-risk and higher-risk) form the basis for testing our volatility-based trading strategy with different risk and loss levels. The tests are based on historical asset price data over a period of nearly the last eleven years. The goal was to historically outperform the broad market by changing various factors, such as the initial asset allocation, the asset reallocation, and the risk and loss level underlying the trading strategy. We achieve this by using the riskier initial asset allocation and applying our trading strategy with a risk and loss level of 10% each. In this case, a historical return of 326% could have been achieved with our trading strategy over the period under review.

The aim of this research was to measure the risk of the IHSG stock data using the Value at Risk (VaR). IHSG stock index data typically indicates a jump. However, Geometric Brownian Motion (GBM) model can not catch any of the jumps. To view the jumps, it is necessary that the model was then developed into a Geometric Brownian Motion (GBM) model with Jumps. On the GBM model with Jumps, returns the data are discontinuous. To determine the value of VaR, the value of return to perform the simulation model of GBM with Jumps is required. To represent processes that contain jumps, discontinuous Poisson process using the Peak-Over Threshold is required. To determine the parameters of model, calibration of historical data using the Maximum Likelihood Estimation (MLE) method is performed. VaR value for GBM model with Jumps with a 95% and 99% confidence level are -0,0580 and -0,0818 while VaR value for GBM model with a 95% and 99% confidence level are -0,0101 and -0,0199. VaR for GBM model with Jumps with a confidence level of 95% and 99% show greater than the model VaR for GBM.

This study explores the relationship between stock volatility, measured by beta & price movement patterns using the Fibonacci series in the context of low beta stocks. Low beta stocks, typically more stable than high beta ones, were analyzed using Fibonacci retracement & extension levels to identify key support & resistance points. Historical price data from Nifty 50 low beta were examined under various market conditions including bullish, bearish & stable. The findings from our research indicate that the low beta stocks adhere to Fibonacci levels, hence it offers reliable long-term guidance. The study highlights that while Fibonacci retracement levels are useful, their effectiveness depends on stock volatility & market conditions, which helps investors in optimizing their trading strategies.

Investing in the Malaysian stock market can be overwhelming due to the abundance of options, which necessitates informed decision-making to navigate the volatile market. This study addresses a common problem faced by investors venturing into the stock market, where instability and fluctuations pose significant risks, leading to financial losses stemming from inadequate knowledge about suitable stocks for investment. Unlike many studies that focus on long-term forecasting methods, this research adopts the Geometric Brownian Motion (GBM) model for short-term investment analysis. The study aims to identify the most effective volatility measurement model, develop a forecasting model using GBM based on the chosen volatility model, and evaluate the accuracy of the GBM model using Mean Square Error (MSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Deviation (MAD). Four volatility models, which include simple, log, high-low, and high-low-closed volatility are analysed to determine the most effective volatility measurement model. Four months of daily stock data were collected to ensure accuracy excluding factors such as seasonality, politics, natural disasters, and wars. Findings indicate that the simple volatility model is the most suitable for forecasting stock market trends using the GBM model, demonstrating high accuracy based on MSE, MAPE and MAD. These results suggest that employing the simple volatility model within GBM model can offer a practical and accurate approach for short-term market analysis in Malaysia, potentially aiding investors in mitigating risks and optimizing their trading strategies.

Traditional financial models often rely solely on historical price data, failing to account for real-time market complexity and sentiment. This research addresses this gap by proposing a quantum-inspired reinforcement learning framework that integrates sentiment indicators and market conditions for optimized multi-stock trading decisions. A quantum-inspired financial modeling methodology is used to maximize trading decisions, with Rewards and Sentiment indicators based on Convolutional Revolution Reinforcement Learning (RMS-CRRL) and Quantum Finance Theory (QFT). This research suggests a multi-stock trading strategy with Dow Jones 50 component stocks, where the financial information is obtained from the Yahoo Finance and Alpha Vantage APIs. The model leverages data preprocessing methods such as missing value interpolation, outlier detection, and consistency enhancements to provide high-quality data. The novelty of the model lies in its integration of market conditions and consumer sentiment, setting it apart from conventional models that rely solely on historical price data. The RMS-CRRL structure utilizes convolutional neural networks for stock price forecasting to minimize uncertainty in the financial market. The revised revolution optimization algorithm is applied to model stability and the best choice of hyperparameters. In addition, the interpretability of the convolutional neural networks-based predictions is guaranteed using rectangular constraints for local interpretable model-agnostic explanations providing explainability in trade decisions. Comparisons for performance show that the suggested model outperforms baseline methods such as traditional DDPG and A3C models, achieving cumulative returns of 22.4%, which reflects a 7.2% improvement in annual returns. The RMS-CRRL approach also demonstrates an average portfolio value benefit of 10-15% compared to the conventional approach.

As an important part of structured products, Barrier Reverse Convertibles also occupies a large market share and attracts many investors, which makes an accurate valuation of it very significant. This paper does an application of Monte Carlo Simulation in Pricing of Barrier Reverse Convertibles. The author intends to verify the feasibility of Geometric Brownian Motion (GBM) and Monte Carlo Simulation on Barrier Reverse Convertibles linked to high technology companies. In this paper, taking 15.00% USD Barrier Reverse Convertible linked to Intel Corporation as an example, GBM and Monte Carlo Simulation are used to estimate final stock prices of intel corporation. By payoff function and present value function, stock prices are transformed into product prices. This article uses three years Intel corporation stock adjusted price, which is from 20 May 2022 to 20 May 2025 as historical data and 2025 2 years U.S. Treasury Bond coupon rate 0.0388 as risk free rate. The estimated price on fixing day is $ 983.21, from which a distinct overpricing in real market can be seen. At most of time points and stock price levels, Deltas and Gammas are relatively high. Given the above results, this article proves that it is feasible to use GBM and Monte Carlo Simulation in valuation of Barrier Reverse Convertibles, but obvious errors still exist. This article provides the future development direction of valuation methods of Barrier Reverse Convertibles.

Stock investment is in great demand by investors because it can provide large profits with large risks or losses, in accordance with the investment principle of low risk low return, high risk high return. Stock prices that fluctuate in a very short time make it difficult for investors to predict stock prices in the future, so investors must pay more attention and gather as much information as possible regarding the shares to be bought or sold. This study aims to create a data mining model using a Linear Regression algorithm that can predict daily stock closing prices to provide information that supports investors in stock transactions. The data used is historical data on daily stock prices for 10 companies in the last 8 years for the period 25 February 2013 – 25 February 2021. Historical stock price data will be prepared using the Noving Average method and create a data mining model using the linear regression method to generate stock price prediction models. The resulting model can be used to predict stock prices well enough to assist investors in making investment decisions to obtain large profits with low risk.

- Predicting stock market movements is always a challenging task due to how unpredictable and constantly changing the market can be. This study aims to improve the accuracy of stock price forecasts by combining traditional time series models with technical indicators. Time series models like ARIMA and Exponential Smoothing are great for spotting patterns and trends in historical data, but they often miss short-term changes. In contrast, technical indicators such as the Relative Strength Index (RSI), MACD, and Bollinger Bands help identify market momentum and shifts in investor behavior. To build a more reliable prediction model, this research uses a combination of these tools. We analyzed historical stock price data from the National Stock Exchange (NSE) of India, covering the years 2016 to 2024. The focus was on major companies from sectors like banking, IT, and FMCG to ensure that the approach works across different industries. Python was used for data cleaning, analysis, and modeling, using libraries like pandas, statsmodels, and ta. The results show that when technical indicators are added to time series models, the predictions become more accurate. This hybrid method works especially well during market ups and downs, making it a strong and flexible forecasting tool. Overall, the study offers a practical and data-driven solution for investors and analysts who want to make better-informed decisions in the stock market. It also helps bridge the gap between traditional statistical methods and real-time market behavior. Key Words: Stock market forecasting, time series analysis, ARIMA, LSTM, technical indicators, data analytics, financial modeling.

Stock prices are a non-linear, long-term series of data. The research topic of this paper is to use the Long Short-Term Memory Network (LSTM) model to analyze the historical stock price data of BYD (BYD) to predict the future stock price trend, and to provide intelligent reference for investors by evaluating the model performance, aiming to improve the accuracy of stock price prediction. The accuracy and consistency of stock price prediction can be increased by using this method, which successfully takes advantage of the time characteristics in historical data. After explaining the idea and architecture of LSTM, this article creates a stock price prediction model based on it, using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) as evaluation metrics. At the same time, network structures with different layers were designed to verify the effectiveness of the model, and the experimental results were shown. The advantages and limitations of the model are highlighted, and the feasibility of deep learning in the stock financial market and the credibility and advantages of strategic decision-making are demonstrated.

Against the backdrop of an increasingly large stock market and a growing number of investors, many individuals have faced bankruptcy due to blindly following trends in stock investment. This situation has created a demand for methods to predict stock prices in advance. To address this issue, researchers have previously proposed various approaches, including traditional analytical methods based on historical data, statistical analysis methods, machine learning and deep learning methods, etc. The research focus of this paper is to investigate the specific conditions under which the Backpropagation Neural Network model delivers superior performance in stock price prediction. This study utilizes 44 years of historical stock price data from Apple Inc. (AAPL), encompassing key features such as Date, Opening Price, Highest Price of the Day, Lowest Price of the Day, Closing Price, Adjusted Closing Price, and Trading Volume. Specifically, the study focuses on selecting the Opening Price, Highest Price of the Day, Lowest Price of the Day, Closing Price, and Trading Volume as input features. The research ultimately achieved relatively accurate results. These findings are then leveraged to extend insights into effective configuration strategies for a wider range of other prediction methods.

Forecasting gold prices is crucial for investors, enabling them to determine the optimal timing for investment and to understand market trends. Various deep learning methods have been employed for this purpose using historical price data. This research employs an LSTM model to forecast gold prices by integrating features derived from news sentiment analysis. Sentiment analysis is conducted through text mining techniques using the IndoBERT model. The resulting sentiment score, based on the polarity and keywords in daily gold news, is designed to represent market trends. LSTM is particularly effective at recognizing complex and varied patterns within time-series data. This study compared models for forecasting gold prices, such as a base case that relies solely on historical data and scenarios that integrate sentiment features over specific periods. This comparative analysis aimed to identify the optimal predictive model. The findings indicate that incorporating sentiment information at a lag of t − 2 yields superior performance compared to the other predicted scenarios. More precisely, this model attained a Mean Absolute Percentage Error (MAPE) of 1.51% and a Root Mean Squared Error (RMSE) of $60.92. These findings confirm that sentiment scores significantly influence gold price predictions and enhance the predictive capability of the LSTM model.

Forecasting of electricity price in the competitive power market has become very important for the market participants in order to support their planning and operations activities. In addition to the change in prices caused by the demand levels (reflected in normal interaction between supply and demand), numerous other factors affect the electricity price including the sudden surges in prices which are frequently caused by unexpected outages. This paper attempts to represent the price at any one period as a random variable which may be treated statistically. Historical price data from Singapore market (USEP) are used to study the behavior of the market price in order to identify the possible statistical distributions and then hypothesis tests are formulated and carried out to determine the theoretical distributions which are best supported by the historical data. It is shown that the raw price data may not be directly amenable to statistical representation. However, suitable processing of the data such as filtering out the extreme values or suitable regrouping of the data may be carried out to bring out the underlying statistical characteristics of the price behavior. It is shown that the market price behavior may support more than one theoretical distribution although the log normal distribution was found to fit the historical data best.

We impose dynamically, a shortfall constraint in terms of Tail Conditional Expectation on the portfolio selection problem in continuous time, in order to obtain optimal strategies. The financial market is assumed to comprise n risky assets driven by geometric Brownian motion and one risk-free asset. The method of Lagrange multipliers is combined with the Hamilton-Jacobi-Bellman equation to insert the constraint into the resolution framework. The constraint is re-calculated at short intervals of time throughout the investment horizon. A numerical method is applied to obtain an approximate solution to the problem. It is found that the imposition of the constraint curbs investment in the risky assets.

Exact Replication of the Best Rebalancing Rule in Hindsight