Extreme Price Research Articles

Novel Custom Loss Functions and Metrics for Reinforced Forecasting of High and Low Day-Ahead Electricity Prices Using Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) and Ensemble Learning

Day-ahead electricity price forecasting (DAEPF) is vital for participants in energy markets, particularly in regions with high integration of renewable energy sources (RESs), where price volatility poses significant challenges. The accurate forecasting of high and low electricity prices is particularly essential, as market participants seek to optimize their strategies by selling electricity when prices are high and purchasing when prices are low to maximize profits and minimize costs. In Japan, the increasing integration of RES has caused day-ahead electricity prices to frequently fall to almost zero JPY/kWh during periods of high RES output, creating significant profitability challenges for electricity retailers. This paper introduces novel custom loss functions and metrics specifically designed to improve the forecasting accuracy of extreme prices (high and low prices) in DAEPF, with a focus on the Japanese wholesale electricity market, addressing the unique challenges posed by the volatility of RES. To implement this, we integrate these custom loss functions into a Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) model, augmented by an ensemble learning approach and multimodal features. The proposed custom loss functions and metrics were rigorously validated, demonstrating their effectiveness in accurately predicting high and low electricity prices, thereby indicating their practical application in enhancing the economic strategies of market participants.

A robust electricity price forecasting framework based on heteroscedastic temporal Convolutional Network

Electricity price forecasting (EPF) is a complex task due to market volatility and nonlinearity, which cause rapid and unpredictable fluctuations and introduce heteroscedasticity in forecasting. These factors result in varying prediction errors over time, making it difficult for models to capture stable patterns and leading to poor performance. This study introduces the Heteroscedastic Temporal Convolutional Network (HeTCN), a novel Encoder-Decoder framework designed for day-ahead EPF. HeTCN utilizes a Temporal Convolutional Network (TCN) to capture long-term dependencies and cyclical patterns in electricity prices. A key innovation is the heteroscedastic output layer, which directly represents variable uncertainty, enhancing performance under fluctuating market conditions. Additionally, a multi-view feature selection algorithm identifies crucial factors for specific periods, improving forecast precision. The framework employs an improved loss function based on maximum likelihood estimation (MLE), which adjusts for the heteroscedastic nature of electricity prices by predicting both the mean and variance of the price distribution. This approach mitigates the impact of extreme price spikes and reduces overfitting, resulting in robust and reliable predictions. Comprehensive evaluations demonstrate HeTCN’s superiority over existing solutions such as DeepAR and the Temporal Fusion Transformer (TFT), with average improvements of 25.3%, 24.9%, and 17.4% in the mean absolute error (MAE), symmetric mean absolute percentage error (sMAPE), and the root of mean squared error (RMSE) compared to DeepAR, and 17.6%, 14.4%, and 13.6% relative to TFT across five distinct electricity markets. These results underscore HeTCN’s effectiveness in managing volatility and heteroscedasticity, marking a significant advancement in electricity price forecasting.

Exploring the Linkage Effects between Coking Coal Futures and Carbon Emission Rights Prices under the Dual-Carbon Framework

Coking coal is a vital energy resource for economic development, with fluctuations in coking coal futures prices significantly guiding spot prices. Additionally, variations in carbon emission rights prices directly affect China's energy conservation and emission reduction efforts. This paper examines the daily trading prices of carbon emission rights and coking coal futures. It utilizes various tests, including the ADF test and cointegration test, and constructs a two-dimensional vector autoregression (VAR) model with a two-period lag. The paper also performs an impulse response function analysis and variance decomposition. Empirical results reveal: First, the daily trading prices of coking coal futures have exhibited a fluctuating upward trend over the past six years; Second, there is a long-term cointegration relationship between coking coal futures prices and carbon emission rights prices; Third, there is no Granger causality between the two; Fourth, fluctuations in carbon emission rights prices have a stronger guiding effect on coking coal futures prices than vice versa. Recommendations include: (1) Strengthening oversight of the coking coal futures market and leveraging policy guidance to prevent extreme prices; (2) Systematically including more participants in the carbon market, optimizing the trading structure, and promoting healthy market development; (3) When developing policies for carbon emission rights, the government should mitigate the impact of price fluctuations on other sectors.

Cryptocurrency and Its Role in Portfolio Diversification

Purpose: This study sought to explore cryptocurrency and its role in portfolio diversification. Methodology: The study adopted a desktop research methodology. Desk research refers to secondary data or that which can be collected without fieldwork. Desk research is basically involved in collecting data from existing resources hence it is often considered a low cost technique as compared to field research, as the main cost is involved in executive’s time, telephone charges and directories. Thus, the study relied on already published studies, reports and statistics. This secondary data was easily accessed through the online journals and library. Findings: The findings reveal that there exists a contextual and methodological gap relating to cryptocurrency and its role in portfolio diversification. Preliminary empirical review revealed that incorporating cryptocurrencies into investment portfolios offered promising diversification benefits due to their low correlation with traditional assets, despite their high volatility and regulatory uncertainties. It highlighted the significant risk management challenges posed by cryptocurrencies' extreme price fluctuations and the evolving regulatory landscape. The study emphasized the importance of careful, limited allocation to cryptocurrencies, robust risk management practices, and continuous market monitoring. Ultimately, it suggested that cryptocurrencies could enhance portfolio performance when strategically used alongside traditional diversification methods. Unique Contribution to Theory, Practice and Policy: The Modern Portfolio Theory, Efficient Market Hypothesis and Behavioural Finance Theory may be used to anchor future studies on portfolio diversification. The study recommended a cautious yet strategic inclusion of cryptocurrencies in investment portfolios to enhance diversification, emphasizing the importance of ongoing research, robust risk management, and proactive monitoring due to their high volatility and regulatory uncertainties. It called for clear and consistent regulatory frameworks to protect investors while fostering market growth, and highlighted the need for collaboration between academia, industry, and regulatory bodies to improve financial literacy and market stability. These recommendations aimed to contribute to theoretical, practical, and policy aspects of cryptocurrency investments.

A survey on cryptocurrency price prediction

A type of digital currency known as a cryptocurrency allows all transactions to be completed online. There is no hard cash version of this soft currency. We highlight that a decentralized currency differs from a centralized currency in the any user of a virtual currency can purchase services without the need for third parties to get involved. Due to its extreme price volatility, using these cryptocurrencies has an impact on trade and international relations. Moreover, the constantly fluctuating oscillations indicate the urgent need for a more precise method of predicting this price. Deep learning techniques that use effective learning models for training data, including the LSTM, GRU, and Feedback Neural Network, can be used to do this. Benchmark datasets are used to test the suggested strategy. That brings us to the neural network, one of the clever data mining technologies that researchers in many domains have been using for the past ten years. In the current economy, stock market data is essential. There are two types of forecasting methodologies: nonlinear models (ARCH, GARCH, Neural Network) and linear models (AR, MA, ARIMA, ARMA). To forecast a company's stock price based on past prices, we employed the Box Jenkins Model also known as ARIMA, and Long Short-Term Memory (LSTM), and Feedback Neural Network also known as RNN.

A comparison of deflation methods for carbon footprint calculations using Japanese data

An embodied CO2 emissions intensity database estimated from input–output data at current prices is now available in Japan. This study compares two non-survey methods: The double-deflation (DD) method and an elaborated approach combining the DD method with the generalized RAS method to deflate an input–output table (IOT). We used these methods to estimate new datasets of embodied CO2 emission intensity data in Japan, utilizing Japan's IOTs for 2005, 2011, and 2015 at constant prices in 2015. Furthermore, we compare the estimated data of 367 commodity sectors for a specific year (2005 or 2011). The results reveal that the intensity for certain aggregated sector groups, such as non-ferrous metals and finance and insurance, exhibits higher uncertainty owing to the extreme price homogeneity assumption associated with the DD method. Consequently, we recommend that life-cycle-assessment practitioners use an open database when analyzing changes in the carbon footprint of products over time.

Extremely stablecoins

In this research, I explore the factors driving extreme fluctuations in stablecoin prices concerning financial and cryptocurrency market indices. Through quantile models, I reveal the directional predictability of diverse financial variables, capturing extreme price variations and return distribution fluctuations. The outcomes highlight asymmetric pricing reactions and the impact of stablecoin stability mechanisms. The varying effects observed in stablecoin and cryptocurrency behavior imply a broader market link and potential fragility. Notably, intermediary constraints and stablecoin liquidity exhibit predictive power for subsequent price shifts in both tails, providing insights into the complex dynamics within cryptocurrency and stablecoin markets.

Cryptocurrency assets valuation prediction based on LSTM, neural network, and deep learning hybrid model

Cryptocurrency, a digital currency managed by decentralized networks, has gained immense popularity since the inception of Bitcoin. These digital assets, often characterized by extreme price volatility, have generated substantial interest from investors. Traditional financial models struggle to account for the unique dynamics and complexities of cryptocurrencies, prompting the adoption of deep learning techniques. This study investigates the use of Long Short-Term Memory (LSTM), Neural Networks, and Deep Learning (CNN) in predicting cryptocurrency prices. These deep learning models leverage various data sources, such as technical indicators and sentiment analysis, to gain a comprehensive understanding of cryptocurrency markets. The research evaluates the performance of these models using Root Mean Squared Error (RMSE) as the primary metric. The results demonstrate that the hybrid model, combining LSTM, Neural Networks, and Deep Learning, exhibits the highest predictive accuracy across multiple cryptocurrencies, including Bitcoin (BTC), Ethereum (ETH), and Binance Coin (BNB). However, challenges persist, such as model adaptability to unforeseen market events and data noise. Future developments may involve incorporating external factors and interdisciplinary collaboration to create more holistic valuation models. Despite these challenges, the study underscores the potential of hybrid deep learning models in enhancing cryptocurrency valuation accuracy and their relevance in risk management strategies for investors and traders.

Optimal liquidation using extended trading close for multiple trading days

The extended trading close (ETC) provides institutional investors an opportunity to trade at the closing price after the regular trading session (RTS) and disclosing the order imbalances to other market participants. ETCs exist in the Nasdaq, the SSE STAR, the SZSE ChiNext and the TWSE. To help a risk-averse institutional investor take advantage of the RTS and the ETC for liquidation, we develop a multistage dynamic programming model including the ETC, and derive recursive solutions for the multiple trading days scenario with closed-form solutions for the scenario with only two trading days. We also verify that the ETC is able to mitigate extreme price movements caused by fast liquidation, which is also a goal of the ETC set out by the SSE STAR and the SZSE ChiNext. Finally, we derive three results. First, an institutional investor can reduce execution costs after the introduction of the ETC. Second, a critical trading day exists, and to avoid prematurely revealing trading intentions, the investor should not trade in the ETC until such day. Third, even though the ETC orders submitted by the investor are unfilled, implementation of the ETC encourages the investor to change the liquidation strategy in the RTS, which reduces extreme price movements. In summary, the practical implications of this paper are that the investor should not trade during the ETC on the front few days to avoid prematurely revealing the investor’s trading intention by unfilled orders in the ETC and that introducing the ETC can reduce liquidation costs and extreme price movements.

Navigating Gene Therapy Access: The Case of Bulgaria in the Context of the EU Regulatory Landscape.

Gene therapies (GTs) have recently emerged as revolutionary personalized therapeutic options. Despite their promising potential, challenges such as uncertainty regarding long-term health benefits and safety, along with extreme price tags, pose significant obstacles to patient access. Within the EU, the European Medicines Agency plays a pivotal role with regards to GT market authorization. However, national authorities are responsible for pricing and reimbursement, which results in fragment patient access within the EU. This study aimed to provide an overview of the complex landscape of post-market authorization accessibility for GT products in Bulgaria, comparing it with neighboring EU countries. We applied a mixed-methods approach, including desk research, public data requests, and list price comparisons. As of 1 April 2023, 14 GTs had a valid market authorization at the EU level. In Bulgaria, Kymriah® was the only GT included in the Positive Drug List (PDL), with an official list price of EUR 335,636.94. Similar results were found in Romania, whereas five GTs were included in Greece's PDL. Additionally, Zolgensma® was found accessible in Bulgaria through an alternative individual access scheme at an estimated price of EUR 1,945,000.00. In conclusion, this study emphasized targeted policy interventions to address health inequalities and to ensure timely access to GTs within the EU.

State-dependent intertemporal risk-return tradeoff: Further evidence

We suggest that the intertemporal risk-return tradeoff is not necessarily positive but rather state dependent. We further explore the state dependent risk-return relation by examining how the positive risk-return relation is distorted in response to various market conditions, including extreme price changes, differing levels of investor sentiment, the introduction of stock options, and throughout business cycles. The tendency for uninformed investors to be optimistic (pessimistic) in response to good (bad) market news cause overpricing (underpricing), and the resulting trade activity of arbitrageurs that distorts the positive risk-return tradeoff, is documented consistently across these environments. We find that the attenuation (reinforcement) of the positive risk-return relation under investors’ optimistic (pessimistic) expectations is stronger in high (low) sentiment periods, in the presence of extreme returns, in the period after stock options became available, and during expansionary periods. We argue that the asymmetric intertemporal risk-return relation is a consequence of rational arbitrageurs’ trading to exploit mispricing through the selling of overpriced stocks conditional on good news and buying underpriced stocks conditional on bad news.

Evaluating the dynamic connectedness of financial assets and bank indices during black-swan events: A Quantile-VAR approach

This study examines whether precious metals, industrial metals, energy and agricultural commodities, or cryptocurrencies form trustworthy safe havens against extreme price volatility of major global bank stock indices during black-swan events such as the COVID-19 pandemic and the Russia–Ukraine conflict. Using daily data and applying Quantile-VAR dynamic pairwise and extended joint connectedness methodologies, we investigate dynamic connectedness between major financial assets and major bank indices during exceptional crises. Findings provide evidence that crude oil and both Ethereum and Bitcoin present evidence of propagating significant shocks towards bank stock indices during crises, but other large-cap cryptocurrencies present no evidence of any specific influence. Further, gold, natural gas, and wheat are identified as the main absorbers of spillovers from banking indices during analysed crises, with more pronounced effects identified during exceptional phases of volatility. Such findings suggest that risk in the banking sector can be efficiently hedged by traditional safe havens such as gold and counterbalanced by highly outperforming assets such as natural gas and wheat. The study significantly contributes to understanding the interplay between banking sectors and various financial assets during crises and the subsequent strategies available for managing systemic risks, providing valuable insights for policymakers, regulators, and investors alike.

The Role of USDA Reports on Extreme Price Comovements in Agricultural Economics

The Role of USDA Reports on Extreme Price Comovements in Agricultural Economics

Energy price bubbles and extreme price movements: Evidence from China's coal market

This study investigates the factors behind the extreme price movements in China's coal market, with a particular focus on the impact of climate risk and energy transition in recent years. The Generalized Sup Augmented Dickey-Fuller (GSADF) method is employed to detect coal price bubbles, and a dynamic model averaging (DMA) approach is then used to analyze the causes of these price bubbles. The findings reveal that price bubbles in the Chinese coal market are mainly triggered by fluctuations in international energy prices. The extreme prices are rooted in supply-demand imbalances resulting from energy transition, economic development, and geopolitical conflicts. Policies aimed at adjusting coal supplies can effectively mitigate abnormal coal price fluctuations in China, while normal coal price fluctuations are significantly influenced by changes in energy demand driven by macroeconomic development. During the green transition towards renewable energy, the current high prices of fossil energy present challenges to China's energy supply security but also offer opportunities for the development of the renewable energy market. However, energy transition has facilitated the spread of price bubbles across coal, natural gas, and renewable energy markets, potentially leading to contagion effects.

‘There Is No Value’

This article draws on narratives from middle-class households in Beirut to examine their disenfranchisement amid the country’s financial collapse. A common expression, “There is no value,” is often used in allusion to extreme price fluctuation that has punctuated everyday life since the end of 2019. The phrase also speaks to the undoing of a middle-class lifestyle that emerged with the pegging of the Lebanese lira to the US dollar in the 1990s. Reflecting on how value is conceptualized at the level of the everyday leads into exploring some of the complex ways in which this middle-class milieu accessed credit following Lebanon’s civil war, and how everything has changed during the current dollar shortage.

Can inflation predict energy price volatility?

Fluctuations in energy prices impact production costs and inflation. This study examines whether inflation data can predict volatility in energy markets. Both inflation and energy market volatility exhibit complex behaviour over time, including structural shifts due to demand and supply shocks. Accounting for differences in data frequencies, we use an extended GARCH model (MIDAS) with Laguerre polynomials for time-varying parameters. The empirical results demonstrate that including low-frequency inflation data enhances energy model predictability particularly during periods of high volatility and extreme price fluctuations. Considering inflation improves forecasting for energy market models, benefiting portfolio management and helping policymakers manage inflation.

Measuring the competitiveness of commodity markets using price signals and information theory

Technological advancements, abrupt changes in market conditions, and political reforms, among other things, necessitate strong regulatory oversight, and accurate measurement of performance related indicators. The more accurate, information rich, and transparent these measurements/signals, the lower the level of uncertainty felt by value chain participants, who are thus able to recognize and observe whether the market’s state is efficient. Its lack, may lead to indecisiveness, translating into false interpretations that could lead to wrong policy directions. This paper provides an ex-post evaluation tool intending to deliver additional insights or quality information that would aid the regulator in assessing the state of the market. The tool is applied to the UK wholesale natural gas market for the period between 2011 and 2020, assessing and testing the market’s weak-form efficiency. It claims that today’s gas prices reflect a specific type of information, primarily past gas prices, and that only new information can help predict future prices. In this manuscript, based solely on a limited and available untapped dataset (day-ahead price time series), and working under the assumption that gas prices are the result of market processes, a variety of information metrics (gas price randomness, distribution of extreme prices, ability to predict prices - based on historical sets) is extracted with the use of suitable mathematical statistical models. A weighted entropy index is then computed, and measures the state of the commodity market. The results indicate that the analysis has helped gain information, thus reducing uncertainty (relative to a pre-analysis) by 86.5 %. Additionally, there is sufficient evidence that the UK natural gas prices are weak-form efficient.

Does price limit improve price discovery? Evidence from IPO market in a quasi-natural experiment

The two major Chinese stock exchanges imposed a new trading rule with a price-move limit of 44% on the first trading day for IPOs after 2013. Following this quasi-natural experiment, market sentiment on IPOs drastically rises, resulting in extreme price run-up in initial trading days. Newly listed stocks under the trading rule experience greater illiquidity, higher turnover and volatility than earlier ones. Account-level data from the Shanghai Stock Exchange shows that following the new rule, net buyers of IPO change from institutional to retail investors. The evidence suggests that this price limit delays price discovery by arousing sentiment and crowding out informed investors.

Assessing jump and cojumps in financial asset returns with applications in futures markets

Price jumps and extreme dependence in financial asset prices are both very important issues in Finance since both would distort the decisions in asset allocation and risk management. Cojumps, as joint occurrence of extreme price movements, share even relevant financial implications about systematic risk. This paper aims to unravel the empirical relationship between cojumps and tail dependence. We first propose a simple approach for identifying jumps and co-jumps in asset prices utilizing the recent realized measures of variation in high frequency finance. The framework is informative in testing the existence of jumps, dating jump days, and quantifying jump contributions to price variations. Our identification approach improves the over-alarm detection of jumps from previous formal tests. Applying the idea to test the presence of cojumps in realized covariances allows us to disentangle the nature of cojumps and the real source of extreme tail dependence. For the empirically investigated S&P 500 spot and futures indices, we find asymmetries in both the frequencies and magnitudes among those downward or upward abrupt comovements. Perhaps surprisingly, we find that it is covolatility instead of cojumps in asset prices that counter-intuitively contributes to the observed extreme tail dependence. However, when the tests are applied in the Taiwanese TAIEX futures and spot indices, a completely different picture emerged. We found evidence of prevailing cojumps, and moreover, it is these cojumps that triggered the marginal jumps within each index returns and contributed to the extreme returns. When further examined cross-sectionally with Taiwanese and Chinese future indices, we documented a pattern of cojump induced tail dependence in futures returns across the Taiwan straits.

A GARCH Model to Understand the Volatility of the Electricity Spot Price in Brazil

Electricity is sensitive to extreme price events and spot price volatility is an inherent characteristic of competitive electricity markets. The purpose of this article it to model the realized volatility of electricity spot price in Brazil. The Brazilian electricity industry presents unique characteristics and because of this price varies a lot in a short period. So, we developed a GARCH model using 862 weekly observations to understand the realized volatility in the four different market. We conclude that the spot price in Brazil presents high volatility that presents risk to agents. This high volatility is associated with institutional factors and the increase in the share of renewable energy in the electricity mix.