Price Movements Research Articles

Media Content, African Swine Fever, a Trade War, and U.S. Lean Hog Futures

Abstract In 2018 and 2019, China’s outbreak of African swine fever (ASF) and the U.S.–China trade war captured media headlines worldwide. This research uses a unique data set of media headlines and sentiments to estimate the impact of media on U.S. lean hog futures prices for nearby and distant expiration contracts. Findings suggest futures prices are influenced by news media content, with results differing by time to contract expiration and sentiment of the headline. International headlines with positive and negative connotations toward ASF and trade war have more significant effects, indicating sensationalist media creates the greatest price movements compared to neutral headlines.

Unpacking the relation between media sentiment and house prices: A topic modeling approach

This study uses a topic modeling approach to investigate the relation between news media sentiment and house price movements. By examining real estate related articles published in local newspapers of 16 major cities in Canada and Australia, we find that housing media sentiment has significant relation with future house price movements in the Canadian market but not in the Australian market. We identify the specific topics discussed in news media related to the housing market and report differences in their themes and media sentiment's predictive power between Canada and Australia. This analysis presents novel inferences of qualitative and hard-to-quantify information related to the housing market in two different countries.

Shipping Market Forecasting Considering the Individual Investor Emotions: A Novel Twofold Partial Swarm Optimization Based Stacked Long Short-Term Memory Model

ABSTRACT Purpose The general purpose of shipping market prediction is to help shipping analysts design a strategy to make shipping strategies. Previous studies have pointed out human emotions has an obvious impact on markets, however, few research focus on utilizing human emotions to improve the shipping market forecasting issue. This paper fills the gap in the literature of considering industrial individual emotions to shipping market prediction issue Methodology We proposed a conceptual framework of an emotion-based shipping market prediction (ESMP) system focused on considering the multidimensional emotions of individual investors. To make it precisely, a two-fold Particle Swarm Optimization (TFPSO) algorithm is proposed conduct the shipping forecasting under the framework of ESMP, with advantages of automatically finding the architecture and hyperparameters of the Deep Long Short-Term Memory (DLSTM) simultaneously in the shipping market forecasting the values of the Baltic Dry, Dirty Tanker and Container indices of ocean transportation in the world ocean during the crisis period 2010–2022 Findings The prediction accuracy of the ESMP was higher than that of the models using the conventional factors in all learning periods of the study, and it was determined that the results complement the sentiment indicator employed to predict the shipping indices. Additionally, the accuracy of the proposed method is superior to conventional neural network models in all used error metrics. Additional Mann–Whitney U test on MSE difference between TFPSO-DLSTM and compared models demonstrates that the significant advantages boosted by TFPSO-DLSTM Originality Shipping sentiment has provided further proof that shipping decisions are significantly driven by emotions. our study shows that individual shipping investors’ sentiment are able to improve shipping rate forecasts significantly, although the magnitudes of the improvements are relatively small from an economic point of view Research implications This study undertook the initiative to procure BDI, BDTI, BCTI, COVID-19 data and shipping sentiment index data through in-depth interviews to provide first-hand perspectives into whether shipping sentiments impact the shipping market trend. It builds upon existing literature on the present stance of deep learning models, which largely relies non-sentiment factors. The study also extends prior literature on hyperparameter searching methods by highlighting the structure of searching method. Besides this, the contribution of this study also aligns with the prior discussions on evolving methodologies in temporal forecasting research Practical implications This study implies that shipping sentiments enables to make better shipping marketing trends, and foster sustainable growth. Shipping fear index serves for measuring investors’ attitude and mood toward shipping markets in terms of the general and certain sectors or assets, possibly promoting the movement of price and providing long-term investors and active traders with trading or arbitrage opportunities. Due to the complexities and regulations of the shipping market, it is essential to collaborate exclusively with shipping sentiment indices with artificial intelligent model in shipping market forecasting research

Advancements in AI for Predicting Price Movements: An In-Depth Literature Survey

Advancements in AI for Predicting Price Movements: An In-Depth Literature Survey

Wavelet Based Financial Forecast Ensemble Featuring Hybrid Quantum-Classical LSTM Model

One of the most sought-after goals in the financial world is a reliable method by which investors can predict a stock price movement consistently. Advancements in stock prediction via the use of machine learning have improved the accuracy of such predictions and yielded better ideas about value investments in the stock market. However, with the addition of the state-of-art tool M-band wavelet transform as a preprocessing step, we can denoise our raw data set (prior stock prices) and refine it to make the forecast even more accurate. Following this preprocessing step, multiple machine learning techniques are deployed to construct a robust, non-parametric hybrid forecasting model. To demonstrate the novelty of our algorithm, we present a case study on stock price prediction employing a discrete 4-band wavelet transform integrated with a hybrid machine learning ensemble. Our results underscore the potential and importance of such ensemble methods in refining the accuracy and reliability of financial forecasts. Furthermore, in theory, quantum computing can further optimize these algorithms, potentially leading to more precise stock price forecasts. In particular, our ensemble will feature a hybrid quantumclassical LSTM Model to demonstrate the potential of both wavelets and quantum computing in stock forecasting.

Enhancing Agricultural Commodity Price Forecasting Using Generative Models: A Deep Learning Approach

Predicting stock market prices is a critical yet challenging task in finance. Technical analysis, a widely used methodology in investment theory, involves forecasting price movements by analyzing historical market data. Recently, deep learning has gained prominence for its exceptional ability to process complex data, making it a popular tool for financial applications such as stock prediction, portfolio optimization, financial information analysis, and trade execution strategies. In this study, we propose a novel deep learning architecture that integrates a Generative Adversarial Network (GAN) with a Convolutional Neural Network (CNN) as the discriminator and Gated Recurrent Units (GRU) as the generator. This framework generates distributions of daily stock prices through adversarial learning to predict stock closing prices. Using daily trading data from Ruchi Soya Industries Limited, an empirical analysis was performed across a broad time frame. Results show that the proposed GAN-based architecture significantly outperforms traditional deep learning models, including GRU, LSTM, and Bi-LSTM, in predicting stock closing prices. These findings demonstrate the potential of this novel approach for improving stock price forecasting accuracy.

Machine Learning-Based Framework for Intraday Stock Price Movement Prediction Using Open Prices and Historical Data

Machine Learning-Based Framework for Intraday Stock Price Movement Prediction Using Open Prices and Historical Data

Energy price and returns: a threshold heteroscedasticity model approach

ABSTRACT The energy market is characterized by its intricate dynamics, influenced by various factors that result in erratic behavior, making it challenging to model energy price returns and volatility. In energy market, the changes in Brent and West Texas Intermediate (WTI) crude oil price yields and volatility are the focus of great interest and attention. Due to the influence of different periods and some economic factors, the fluctuation of crude oil price shows asymmetric characteristics. For exploring the asymmetric characteristics of energy market, we introduce the application of the threshold double autoregressive (TDAR) model to handle the complexities of energy price returns and volatility. The TDAR model is able to capture nonlinearities and regime shifts in the oil market, making it particularly suitable for understanding the inherent complexities in crude oil price movements. The empirical results implied that TDAR model demonstrates superior performance in capturing the intricate dynamics of crude oil prices compared to traditional asymmetric GARCH-type models, making it a valuable tool for market practitioners and policymakers. Our findings reveal the subtle movements in Brent and WTI prices and offer valuable insights for risk management and investment strategies, which provide an alternative for market practitioners and policymakers.

Predictive crypto-asset automated market maker architecture for decentralized finance using deep reinforcement learning

This study proposes a quote-driven predictive automated market maker (AMM) platform with on-chain custody and settlement functions, alongside off-chain predictive reinforcement learning capabilities, to improve the liquidity provision of real-world AMMs. The proposed architecture augments Uniswap V3, a cryptocurrency AMM protocol, by using a novel market equilibrium pricing to reduce divergence and slippage losses. Furthermore, the proposed architecture involves a predictive AMM capability, for which a deep hybrid long short-term memory (LSTM) and Q-learning reinforcement learning framework is used. It seeks to improve market efficiency through obtaining more accurate forecasts of liquidity concentration ranges, where liquidity starts moving to expected concentration ranges prior to asset price movement; thus, liquidity utilization is improved. The augmented protocol framework is expected to have practical real-world implications through (1) reducing divergence loss for liquidity providers; (2) reducing slippage for crypto-asset traders; and (3) improving capital efficiency for liquidity provision for the AMM protocol. The proposed architecture is empirically benchmarked against the well-established Uniswap V3 AMM architecture. The preliminary findings indicate that the novel AMM framework offers enhanced capital efficiency, reduced divergence loss, and diminished slippage, which could potentially address several of the challenges inherent to AMMs.

The Dynamic Relationship Between Exchange Rate Volatility and Stock Prices in the Egyptian Real Estate Market and the Moderating Effect of Interest Rates

Among different channels of the financial market, the exchange rates and stock markets play a vital role in the domestic and international markets. This study aims to explore the dynamic relationship between the exchange rate (USD/EGP) and the real estate sector listed companies of the Egyptian Stock Exchange (EGX) from 2013 to 2023. Additionally, the study examines the moderating effect of interest rates on the relationship. To achieve this, the study employs several econometric methods such as Augmented Dicky Fuller (ADF) test, Granger Causality test to determine the direction of the relationship, Vector Autoregressive (VAR) model, and more. The results demonstrate that a long-run relationship exists between the variables. Furthermore, there is a significant unidirectional causality that runs from the exchange rate to real estate stock prices; thus, fluctuations in the exchange rate can significantly affect the profitability of investments in real estate stocks. Therefore, investors should closely monitor the movements in the exchange rate level to predict changes in real estate stock prices. When including the interest rate as a moderator, the outcome changes to bidirectional causality, meaning that the exchange rate and real estate stock price movements affect each other at a 1% significance level. Hence, interest rate could be used as an instrument to control demand in both the stock market and foreign exchange market. Finally, it is recommended that policymakers use the monetary policy while considering this bidirectional relationship in the presence of interest rate as a moderating variable.

Exchange Rate Pass-Through to Provincial Consumer Prices: A Geographical-Goods Level Analysis

This study investigates exchange rate pass-through (ERPT) to provincial consumer prices in Canada from a geographical and goods-specific perspective. ERPT refers to the extent in which changes in the exchange rate influence domestic price movements. For example, a depreciation of the Canadian dollar relative to the U.S. dollar may increase the price of imports into Canada from the U.S., thereby raising consumer prices in Canada. Using extensive datasets from 2000 to 2023, this research quantifies the magnitude and speed of ERPT across various provinces and consumer price index (CPI) categories. The methodology employs a regression model framework, building upon the approach of Savoie-Chabot and Khan (2015) to estimate the impact of short-run and long-run exchange rate fluctuations on consumer prices. The findings highlight significant regional and categorical variation in ERPT. For example, Alberta exhibits the highest short-run ERPT for energy consumer prices where a 1% increase in the exchange rate leads to a 0.60% increase in prices. In contrast, Quebec exhibits the lowest short-run ERPT for energy consumer prices where a 1% increase in the exchange rate leads to only a 0.18% increase in prices. Price responses in each of these provinces differs from the short-run ERPT of Canada which is 0.32%. Furthermore, energy prices exhibit the highest sensitivity to exchange rate fluctuations, while most often categories excluding energy show minimal impact. These results emphasize the importance of region-specific economic policies to address distinct provincial price responses to exchange rate fluctuations. This study offers valuable insights for policymakers to develop strategies that mitigate any adverse economic effects of exchange rate changes and promote regional economic stability and growth.

Information flow and investment cash flow sensitivity

ABSTRACT In this paper, we examine whether firms with greater information flow, i.e. which reveal more firm-specific information through their idiosyncratic stock price movements, are associated with lower investment cash flow sensitivity. Using a sample of 450 French non-financial firms over the period 2006–2022, we find results consistent with the idea that greater information flow decreases information asymmetry, thus allowing firms to carry out investments more efficiently as they depend less on the availability of internal cash flows.

Bitcoin Sentiment Analysis using Machine Learning Algorithms

The rapid rise of cryptocurrencies, particularly Bitcoin, has intensified research interest in understanding the factors influencing their price movements. Among these factors, social media sentiment has emerged as a crucial predictor, reflecting collective investor mood and market expectations. This paper provides a comprehensive survey of various sentiment analysis models applied in cryptocurrency markets, with a specific focus on the relationship between social media sentiment and Bitcoin price fluctuations. The study identifies the Aigents model as the most effective, showing significant improvements in predictive accuracy following fine-tuning. Findings reveal a predictive association between sentiment measures and price changes, typically with a latency of one to two days. The paper offers insights into the capabilities and limitations of existing Natural Language Processing (NLP) models in the context of cryptocurrency sentiment analysis, presenting practical implications for investors and analysts in navigating the volatile cryptocurrency markets. Key Words: cryptocurrency, Bitcoin, social media sentiment, natural language processing (NLP), sentiment analysis, price prediction, machine learning models, Aigents model, financial forecasting, Twitter, Reddit, artificial neural networks (ANN), Long Short-Term Memory (LSTM), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Random Forest, Naive Bayes, Extreme Gradient Boosting (XGBoost), predictive analytics, market behavior, interpretable AI.

Applied Random Forest Algorithm for News and Article Features on The Stock Price Movement: An Empirical Study of The Banking Sector in Vietnam

Applied Random Forest Algorithm for News and Article Features on The Stock Price Movement: An Empirical Study of The Banking Sector in Vietnam

Can ChatGPT Predict Stock Market Price Movements?

Can ChatGPT Predict Stock Market Price Movements?

CRYPTOCURRENCY PRICE PREDICTION USING LSTM

Cryptocurrency, a digital form of currency, has emerged as a prominent asset class with unique characteristics such as decentralization, security, and global accessibility. As the adoption of cryptocurrencies continues to grow, the need for accurate price prediction using machine learning (ML) algorithms becomes crucial for various stakeholders in the financial ecosystem. This paper presents a comprehensive approach to cryptocurrency price prediction, focusing on the following key aspects: The data for this study is collected from the Binance platform, a leading cryptocurrency exchange known for its extensive market data and liquidity. The dataset includes historical price data across different time frames, including 1-hour, 4-hour, and daily intervals. Prior to analysis, the collected data undergoes thorough preprocessing steps to ensure data quality and consistency. This process includes handling missing values, removing outliers, and standardizing data formats for further analysis. Long Short-Term machine learning model algorithm is employed for price prediction. This model is chosen for its ability to capture complex patterns and dynamics in cryptocurrency price movements. The prediction phase involves training and testing the ML model using the preprocessed data. Performance evaluation metrics such as R-squared, Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), and Mean absolute percentage error (MAPE) are utilized to assess the accuracy and robustness of the prediction model across different time frames. Accurate cryptocurrency price prediction is essential for various stakeholders, including investors, traders, businesses, and regulators. It facilitates informed decision- making, risk management, market analysis, trading strategy development, business planning, and regulatory compliance in the dynamic cryptocurrency market. By addressing these key elements, this study aims to contribute to the advancement of cryptocurrency price prediction methodologies using ML techniques, thereby enhancing decision-making processes and fostering a more efficient and transparent digital asset market ecosystem.

ANALYZING FINANCIAL RATIOS' IMPACT ON STOCKS RETURNS: CASE STUDY OF LQ45 INDEX 2014-2023

This study examines the correlation between fundamental financial metrics and stock returns of Indonesia's LQ45 Index over a 10-year period (2014-2023). The study aims to provide actionable insights for investors and companies by identifying the most critical financial metrics in predicting stock price movements. By analyzing data from 16 consistently listed Indonesian companies, the study focuses on stocks that have consistently maintained their position in the index over the past decade, a gap that has been underexplored in prior research. The study employs various statistical tests to validate the regression model, with hypothesis testing conducted through t-test and F-tests to determine the significance of financial ratios on stock returns. The analysis shows that Return on Equity, Price to Earnings Ratio (PER), Price to Book Value (PBV), Free Cash Flow per Share to Price (PFCF), and Dividend Payout Rate (DPR) have a significant positive effect on stock return. The research recommends that retail investors prioritize the PFCF ratio when evaluating potential stock investments. Transparency in financial reporting is essential, with companies encouraged to provide accurate and comprehensive financial statements, including clear reporting of key financial metrics, to build investor confidence.

Integrating PCA with deep learning models for stock market Forecasting: An analysis of Turkish stocks markets

Financial data such as stock prices are rich time series data that contain valuable information for investors and financial professionals. Analysis of such data is critical to understanding market behaviour and predicting future price movements. However, stock price predictions are complex and difficult due to the intense noise, non-linear structures, and high volatility contained in this data. While this situation increases the difficulty of making accurate predictions, it also creates an important area for investors and analysts to identify opportunities in the market. One of the effective methods used in predicting stock prices is technical analysis. Multiple indicators are used to predict stock prices with technical analysis. These indicators formulate past stock price movements in different ways and produce signals such as buy, sell, and hold. In this study, the most frequently used ten different indicators were analyzed with PCA (Principal Component Analysis. This study aims to investigate the integration of PCA and deep learning models into the Turkish stock market using indicator values and to assess the effect of this integration on market prediction performance. The most effective indicators used as input for market prediction were selected with the PCA method, and then 4 different models were created using different deep learning architectures (LSTM, CNN, BiLSTM, GRU). The performance values of the proposed models were evaluated with MSE, MAE, MAPE and R2 measurement metrics. The results obtained show that using the indicators selected by PCA together with deep learning models improves market prediction performance. In particular, it was observed that one of the proposed models, the PCA-LSTM-CNN model, produced very successful results.

A Study on Stock Price Prediction Based on Attention GCN-BiLSTM Modeling

It is of great significance to accurately predict stock price movements. By combining the R-Vine-Copula structure with the relationship between the conceptual sectors to which the stocks belong, the price correlation between stocks is measured and the adjacency matrix is constructed, and the model AGLSTM, which is constructed by combining the temporal attention mechanism, Graph Convolutional Neural Network and Bi-directional Long and Short-Term Memory Neural Network, is proposed to predict the price of 46 stocks from the constituent stocks of the SSE 50 by modeling them. The experimental results show that, compared with the baseline, the AGLSTM can predict the stock price. The experimental results show that compared with the baseline model, the MAE and RMSE results of the AGLSTM model predicted one step forward outperform all the comparison models, and the MAPE results are close to the best baseline model results. The results of MAPE are close to those of the best baseline model. Moreover, good results are also achieved in the experiment of multi-step forward prediction, which demonstrates the ability of AGLSTM model in long-term prediction and can provide some reference for investors to make investment decisions.

Bitcoin Return Prediction: Relevant Factors Determining Return

Cryptocurrencies with Bitcoin being the most well-known, are volatile in price. Being able to predict or speculate price movements in Cryptocurrencies would then be meaningful for researchers and lucrative for investors. This paper hypothesized relevant factors that may determine return of Bitcoin, namely Bitcoin price data, google trend and precious metal price data, and collects data for the past 6 years. After regrouping and processing the data, a regression model is used to test the strengths of predictability of each possible factor. The results reveal that Bitcoin price data is a valid predictor of daily Bitcoin return, google trend or search volume is valid only for weekly return predictions, and precious metal prices do not have a significant relationship with Bitcoin return.