Daily Price Research Articles

Comparative Analysis of Machine Learning, Hybrid, and Deep Learning Forecasting Models: Evidence from European Financial Markets and Bitcoins

This study analyzes the transmission of market uncertainty on key European financial markets and the cryptocurrency market over an extended period, encompassing the pre-, during, and post-pandemic periods. Daily financial market indices and price observations are used to assess the forecasting models. We compare statistical, machine learning, and deep learning forecasting models to evaluate the financial markets, such as the ARIMA, hybrid ETS-ANN, and kNN predictive models. The study results indicate that predicting financial market fluctuations is challenging, and the accuracy levels are generally low in several instances. ARIMA and hybrid ETS-ANN models perform better over extended periods compared to the kNN model, with ARIMA being the best-performing model in 2018–2021 and the hybrid ETS-ANN model being the best-performing model in most of the other subperiods. Still, the kNN model outperforms the others in several periods, depending on the observed accuracy measure. Researchers have advocated using parametric and non-parametric modeling combinations to generate better results. In this study, the results suggest that the hybrid ETS-ANN model is the best-performing model despite its moderate level of accuracy. Thus, the hybrid ETS-ANN model is a promising financial time series forecasting approach. The findings offer financial analysts an additional source that can provide valuable insights for investment decisions.

Analyzing the influence of geopolitical risks on European power prices using a multiresolution causal neural network

This paper presents a rigorous investigation into the multiresolution cross-correlation and causality between European energy markets and geopolitical risk (GPR). Using daily electricity spot prices from January 2015 to January 2023, the study employs the variational mode decomposition (VMD) method to uncover the underlying patterns and reveal the interaction between European power prices and GPR. The paper further develops a VMD-assisted multi-scaled causal neural network (VMD-M-CNN) as a multivariable forecasting approach, benchmarking it against other competing models. The results demonstrate a strong dependence between electricity markets and GPR across different investment time-scale horizons. Additionally, the relationship between the two has taken different forms over the crisis period in contrast with calm periods. The proposed VMD-M-CNN approach proves to be more effective than other benchmark models, providing valuable insights for market participants and policymakers to make informed decisions. The rigorous methodology employed and the novel approach developed enhance the credibility and robustness of the study, contributing to the growing body of literature on energy markets and GPR.

Electricity price forecasting in New Zealand: A comparative analysis of statistical and machine learning models with feature selection

In this study, we present an empirical comparison of statistical models and machine learning models for daily electricity price forecasting in the New Zealand electricity market. We demonstrate the effectiveness of GARCH and SV models and their t-distribution variants when paired with feature selection techniques, including LASSO, mutual information, and recursive feature elimination. A key aspect of our study is the inclusion of a diverse set of explanatory variables in all models. We compare these models against a range of popular machine learning models, including LSTM, GRU, XGBoost, LEAR, and a four-layer DNN, where the latter two are considered benchmarks. Our results reveal that GARCH and SV models, particularly their t variants, perform exceptionally well when paired with feature selection techniques and explanatory variables. In most scenarios considered, these models outperform machine learning models when coupled with LASSO feature selection. This contribution provides a comprehensive evaluation of the performance of different models and feature selection techniques for electricity price forecasting in the New Zealand electricity market. Our best-performing model improves the symmetric mean absolute percentage error (sMAPE) and mean absolute scaled error (MASE) by 2% to 3% over the LEAR benchmark model, highlighting the practical relevance of our findings.

Is the Markowitz Model Still Effective in Forming an Optimal Stock Portfolio? Empirical Evidence from the IDX ESG Leaders Index

The current portfolio calculation was basically developed by Markowitz (1952) who gave rise to the Modern Portfolio Theory. However, along with the development of science, other models emerged in portfolio calculation. This study tried to provide a flashback to whether the Markowitz portfolio model is still effective in forming an optimal portfolio compared to other portfolio calculation models. This study used descriptive research with a quantitative approach. This study used secondary data in the form of daily stock prices of companies, including the IDX ESG Leaders Index on the Indonesia Stock Exchange for the period of December 2020 to March 2022. The results showed that 11 stocks out of 25 stock samples were included in the optimal portfolio in both the Markowitz model and the Single Index Model but with different stock compositions. In terms of portfolio performance, the Markowitz model is still superior to the Single Index Model. This can be seen from the calculation results of the three indexes used, including the Sharpe Index, Treynor Index, and Jensen Index. This study provided an important contribution that the Markowitz model portfolio formation can be used and has proven to perform well in the formation of a stock portfolio.

Dynamic hedging strategies across assets and commodities – a wavelet analysis

PurposeThe aim of this research is to explore multiscale hedging strategies among cryptocurrencies, commodities, and GCC stocks. Particularly, this is done by evaluating the connectedness among these asset classes covering a period with COVID-19 implications. Using the wavelet approach, the present study aims to recommend whether there exist different time horizon-based hedging abilities across the asset classes.Design/methodology/approachThe approach used in this study is a multiscale decomposition of time series based on wavelets of daily prices of 13 asset classes. Since the wavelet analysis allows to decompose the time series into its frequency components at different time scales by a filtering process the study covered 1-day, 8-day, and 64-day time horizons to examine the hedging properties across those asset classes.FindingsThe results of this study show that hedging effectiveness differs among stock markets over time. In some cases, cryptocurrencies may keep their hedging properties across time while in others they switch from safe haven to hedge devices. In almost all cases, the three main cryptocurrencies showed diversifying properties as was observed by the multiscale correlation and hedge ratio estimations. In a competing sense, gold showed safe haven properties across time than cryptocurrencies except at an 8-day time scale where hedge ratios were low, positive and statistically different from zero that could be interpreted as a good hedge device in the medium term.Research limitations/implicationsThough this research has considered a set of thirteen asset classes, it was limited to a period in which most cryptocurrencies started trading for the first time which reduces the number of observations compared to Bitcoin prices and stable coins such as Ethereum, Ripple, and Bitcoin Cash. Also, the research was focused on the GCC stock markets which may have different results as compared to other regional markets of Asia or Latin America. A comparative analysis in future could be another area of research in future.Practical implicationsThis study has some significant policy implications. The cryptocurrency market is severely affected by demand and risk shocks to crude oil prices during the COVID-19 period. From the investor's point of view, diversification benefits can be obtained by combining cryptocurrencies along with oil-related products during episodes of financial turmoil and COVID-19 pandemic. The GCC region is constantly endeavoring to adopt more scientific tools and mechanisms of investment, and therefore, this study's results will provide some useful directions to the government, policymakers, financial institutions, and investors.Originality/valueThe current study covers a big bunch of 13 assets spanning across financial and real assets. This is based on literature gap and hence, will be a significant addition to the existing literature. Moreover, the GCC region is emerging as a global investment hub and this study will provide investors dynamic hedging strategies across these asset classes.

Impacts of trading restrictions on price volatilities and speculative activities: Evidence from CSI 300 futures

Commonly trading restrictions have various impacts on asset prices and investors’ speculative activities. This study uses the unique trading restrictions in China from 2015 to 2017 to study how it affects the price volatilities and speculative activities of China Securities Index 300 (CSI 300) futures. To capture the daily price volatilities precisely, we apply four different measures incorporate opening, high, low, and closing prices. Speculative and hedging activities are calculated by using volume and open interest data. We find interesting evidence that trading restrictions decrease the price volatilities and the trading activities of speculators. However, though the trading restrictions have been eased since 2017, the speculators are still unwilling to participate in futures market immediately. Overall, the trading restrictions affect the price volatilities and speculative activities and provide the protection for the majority of investors, however, it also hinders the price discovery function of the futures market.

The Chinese oil futures volatility: Evidence from high-low estimator information

This paper mainly investigates whether the high-low estimator has valuable information to predict the Chinese oil futures volatility. The results show that the high-low estimator constructed based on daily prices contains useful information to predict the Chinese oil futures volatility. Moreover, adding regime switching to the models is helpful to improve forecasting accuracy, especially combining regime switching and the high-low estimator. This paper tries to provide new evidence for Chinese oil futures volatility prediction.

Healthy Eating for All? The Challenge of Adhering to Dietary Guidelines for Low-Income Groups in China.

The Chinese Dietary Guidelines propose a balanced diet for healthy living, but the affordability of this diet needs to be considered, especially for low-income households. To investigate the affordability of a healthy diet, this study analyzes the daily retail prices of 46 food items in 36 Chinese cities from 2016 to 2021. This study compares expenditure, diet composition, and nutritional status in two scenarios aligned with the guidelines. The results show that the mean minimum cost of a balanced diet exceeds the current per capita food expenditure for at least 182.85 million urban households. This suggests that low-income people would need to increase their expenditure by at least 20-121% to achieve the recommended diets. This study also identifies affordable and nutrient-dense foods such as standard flour, eggs, black beans, and cabbage, which policymakers should focus on when monitoring food prices. The findings recommend a combination of social and food system policies to reduce prices and make healthy diets accessible. This study identifies the gaps in the Chinese Dietary Guidelines for ensuring accessibility for vulnerable groups and provides a template for policymakers and researchers to track diet affordability using available food price data in China, contributing to China's 2030 Health Plan and the UN's Sustainable Development Goals.

Does investor sentiment create value for asset pricing? An empirical investigation of the KOSPI‐listed firms

AbstractThis paper proposes the development of an improved investor sentiment index (ISI) to apply on the Korea Composite Stock Price Index (KOSPI) and assess the vitality of sentiment‐based factor for explaining critical equity market anomalies in asset pricing in Korea. We follow the methodology of Huang et al. (2015), the align sentiment index, and employ the partial least squares method to overcome the drawbacks of the pioneering BM index of Baker and Wurgler (2006, 2007). Based on the daily trading and price data for individual companies from 2006 to 2021, we construct a novel ISI, which has robust predicting ability for the aggregate stock market return, in comparison to other popular measures of sentiment in the contemporary finance literature. Furthermore, the sentiment‐based factor in this paper captures the small firm effect that the asset pricing modelling, containing the more topical Fama–French five factor modelling (5F‐FF), has struggled to illuminate completely. Given that our results have shown Korean stock market as fairly well‐organised in terms of the availability of the market intelligence, we speculate our results to have important managerial implications for financial regulators in Korea and countries holding similar economic features.

The Influence Of Employer Pension Funding Size, Operational Costs, Active Management and Age On Financial Performance

This study aims to analyze the effect of size, operating costs, active management, and the age of Employer Pension Funds on the financial performance of Employer Pension Funds in Indonesia. The population used is the Pension Fund of Employers in Indonesia for the 2017-2021 period by taking samples through purposive sampling. The research method used is a quantitative approach by testing using multiple linear regression analysis in the SPSS program. The source of the data was obtained internally from the OJK and the published results of the JCI daily closing price. The type of data used is secondary data based on the financial reports of Employer Pension Funds and the JCI daily closing price. The results of the study show that the size, operational costs and active management have a negative effect on the financial performance of Employer Pension Funds in Indonesia. Meanwhile, the age of Employer Pension Funds has no effect on the financial performance of Employer Pension Funds in Indonesia

New horizon in fuzzy distributions: statistical distributions in continuous domains generated by Choquet integral.

In this paper, some statistical properties of the Choquet integral are discussed. As an interesting application of Choquet integral and fuzzy measures, we introduce a new class of exponential-like distributions related to monotone set functions, called Choquet exponential distributions, by combining the properties of Choquet integral with the exponential distribution. We show some famous statistical distributions such as gamma, logistic, exponential, Rayleigh and other distributions are a special class of Choquet distributions. Then, we show that this new proposed Choquet exponential distribution is better on daily gold price data analysis. Also, a real dataset of the daily number of new infected people to coronavirus in the USA in the period of 2020/02/29 to 2020/10/19 is analyzed. The method presented in this article opens a new horizon for future research.

Electricity Price spirals now or later: An emerging market experience

Time series models have been employed to identify the structural breaks and the existence of volatility based on the behavior and direction of electricity prices. The models were tuned and validated with dataset consisting of the daily spot electricity prices of 13 bid areas of the Indian exchange-traded market for ten years. The Indian exchange-traded market is a diversified market divided into 13 bid areas, each consisting of states with unique trading characteristics. Hence it is vital to conduct an in-depth analysis of each to identify the problematic areas, key reasons behind it and further tame it, which has never been done before. The results revealed that the bid areas S1 and S2 depicts negative leverage effect predominantly due to the elections and transmission constraints. Whereas rest of the areas shows positive result with less price spirals due to changes in regulation, weather and new operational plants. Insights from the research encourages power market participants to further apply forecasting techniques and prepare a price risk mitigation model and reach out at a global level.

Cryptocurrency Price Prediction Using Enhanced PSO with Extreme Gradient Boosting Algorithm

Abstract Due to the highly volatile tendency of Bitcoin, there is a necessity for a better price prediction model. Only a few researchers have focused on the feasibility to apply various modelling approaches. These approaches may prone to have low convergence issues in outcomes and acquire high computation time. Hence a model is put forward based on machine learning techniques using regression algorithm and Particle Swarm Optimization with XGBoost algorithm, for more precise prediction outcomes of three cryptocurrencies; Bitcoin, Dogecoin, and Ethereum. The approach uses time series that consists of daily price information of cryptocurrencies. In this paper, the XGBoost algorithm is incorporated with an enhanced PSO method to tune the optimal hyper-parameters to yield out better prediction output rate. The comparative assessment delineated that the proposed method shows less root mean squared error, mean absolute error and mean squared error values. In this aspect, the proposed model stands predominant in showing high efficiency of prediction rate.

Explainable artificial intelligence modeling to forecast bitcoin prices

Forecasting cryptocurrency behaviour is an increasingly important issue for investors. However, proposed analytical approaches typically suffer from a lack of explanatory power. In response, we propose for cryptocurrency pricing an explainable artificial intelligence (XAI) framework, including a new feature selection method integrated with a game-theory-based SHapley Additive exPlanations approach and an explainable forecasting framework. This new approach, extendable to other uses, improves both forecasting and model generalizability and interpretability. We demonstrate that XAI modeling is capable of predicting cryptocurrency prices during the recent cryptocurrency downturn identified as associated in part with the Russian-Ukraine war. Modeling reveals the critical inflection points of the daily financial and macroeconomic determinants of the transitions between low and high daily prices. We contribute to financial operating systems research and practice by introducing XAI techniques to enhance the transparency and interpretability of machine learning applications and to support various decision-making processes.

Bitcoin Price Prediction

Abstract: In this paper, we have proposed to predict the value of Bitcoin accurately considering the various parameters that affect the value of Bitcoin. Collecting information from various references and using papers in real time, I discovered the advantages and disadvantages of predicting the price of Bitcoin. Each paper has its own set of ways to predict the price of Bitcoin. Many papers have an accurate value but some do not, but time complexity is high in those predictions, so to reduce the complexity of the time here in this paper we use an algorithm linked to artificial intelligence called LASSO (at least one opt-out operator. Some papers have used different algorithms such as SVM (support vector machine), GLM, CNN (Convolutional Neural Networks), and RNN (Recurrent neural networks) which do not have good time management, but LASSO acquisition of results on a larger website is faster and faster so for this purpose, we find comparisons between other algorithms and the LASSO algorithm, this test paper helps future researchers to make an impact on their papers. The process takes place in the first paper moment of research, we aim to understand and discover everyday trends in the Bitcoin market while gaining insight into the relevant features surrounding Bitcoin price. Our data set contains various features related to the value of Bitcoin and the payment network throughout the years, which are recorded daily. We previously processed data, using other data mining techniques to reduce data noise. Then for the second minute of our study, we use available information, and we will predict the daily price change signal with the highest accuracy.

A Stochastic Analysis of the Effect of Trading Parameters on the Stability of the Financial Markets Using a Bayesian Approach

The purpose of this study was to identify and measure the impact of the different effects of entropy states over the high-frequency trade of the cryptocurrency market, especially in Bitcoin, using and selecting optimal parameters of the Bayesian approach, specifically through approximate Bayesian computation (ABC). ABC corresponds to a class of computational methods rooted in Bayesian statistics that could be used to estimate the posterior distributions of model parameters. For this research, ABC was applied to estimate the daily prices of the Bitcoin cryptocurrency from May 2013 to December 2021. The findings suggest that the behaviour of the parameters for our tested trading algorithms, in which sudden jumps are observed, can be interpreted as changes in states of the generated time series. Additionally, it is possible to identify and model the effects of the COVID-19 pandemic on the series analysed in the research. Finally, the main contribution of this research is that we have characterised the relationship between entropy and the evolution of parameters defining the optimal selection of trading algorithms in the financial industry.

Do the US president's tweets better predict oil prices? An empirical examination using long short-term memory networks

ABSTRACT The price of oil is highly complex to predict as it is impacted by global demand and supply, geopolitical events, and market sentiment. The accuracy of such predictions, however, has far-reaching implications for supply chain performance, portfolio management, and expected stock market returns. This paper contributes to the oil price prediction literature by evaluating the predictive impact of the US President's communication on Twitter, while benchmarking various Natural Language Processing (NLP) techniques, including Term Frequency-Inverse Document Frequency (TF-IDF), Word2Vec, Doc2Vec, Global Vectors for Word Representation (GloVe), and Bidirectional Encoder Representations from Transformers (BERT). These techniques are combined with a deep neural network Long Short-Term Memory (LSTM) architecture using a five-day lag for both the oil price and the textual Twitter data. The data was collected during the term of US President Donald Trump, resulting in 1449 days of crude oil price prediction and a total of 16,457 tweets. The study is validated for Brent and West Texas Intermediate blends, using the daily price of a barrel of crude oil as the target variable. The results confirm that including the US President's tweets significantly increases the predictive power of oil price prediction models, and that an LSTM architecture with BERT as NLP technique has the best performance.

China mainland new energy index price forecasting with the neural network

For policymakers and investors, forecasting prices of energy indices has always been an important task. The present work focuses on the Chinese market and explores the daily price forecasting problem for the new energy index in the mainland during the period spanning January 4, 2016 – December 31, 2020. Our analysis is facilitated through the nonlinear autoregressive neural network model and one hundred and twenty model settings are tested in the fields of the algorithm for training the model, the number of hidden neurons utilized, the number of delays utilized, and the ratio utilized for segmenting the price series into different phases. Analysis here leads to the construction of a rather simple model based upon four delays and two hidden neurons with the Levenberg-Marquardt algorithm for training, which generates accurate and stable forecast results, with relative root mean square errors below 1.80% and mean absolute percentage errors below 1.30% across training, validation, and testing phases and for the overall sample. This constructed model also leads to statistically significantly better forecasting performance than a linear autoregressive model at the 1% significance level based upon the modified Diebold-Mariano test. The model built could be used as part of policy analysis for policymakers and decision making for investors. The forecasting results might also benefit design of similar energy indices by offering reference information in terms of price paths projected through the model.

Forecasting the Daily Stock Prices of the Philippine Stock Exchange Index During the Opening and Closing of the Market

While there are many factors and other alternative methods to forecast the future performance of the Philippine Stock Market Index, the researchers have elected to utilize a quantitative approach based on technical market factors available which consists mainly of intra-day daily trading levels of the index.

Speculation or currency? Multi-scale analysis of cryptocurrencies—The case of Bitcoin

This paper proposes a novel two-stage VMD-based multi-scale regression to analyze various cryptocurrency attributes that are still unclear in the existing literature. In the first stage, Variational Mode Decomposition (VMD) is used to decompose the cryptocurrency prices into low, medium and high frequency modes with different attributes. In the second stage, the VMD-based multi-scale regression is proposed for these modes with selected explanatory variables. Using the proposed framework, we focus on analyzing the multiple attributes of daily Bitcoin price data as a case study. Empirical results indicate that the low-frequency mode has specific currency or long-term investment characteristics, unlike the short/medium-term investment attributes for the medium-frequency mode, while the high-frequency mode represents some speculation. Some events merely affect a single frequency mode, but others impact all frequency modes. The results of events analysis based on VMD could enhance the identification of the multiple attributes of Bitcoin. Our findings are insightful for future regulation and management of virtual currencies.