Portfolio Trading Research Articles

Buildings as Batteries

Aggregators are increasingly using portfolios of various end-user resources such as local renewable generation, batteries, EVs, and demand-side management to participate in the electricity markets as virtual power plants. Whilst the smart control of energy within buildings is well established for cost optimization, the prospect of an aggregator collaborating with the energy managers of buildings to trade electricity in the wholesale markets is under researched. This paper uses a state-of-the-art architectural simulator to model the thermal inertia of a typical high-rise building in London and represent the thermodynamics of its hourly occupancy and weather responses. The building is simulated as a virtual battery by manipulating the HVAC settings. Price arbitrage on the electricity wholesale market and contracting with the local distribution network for flexible reserve services are both feasible and profitable. It is concluded that aggregators can apparently use buildings as additional assets within their trading portfolios. JEL Classification: L85 and L94.

Channel and pricing strategies in a risk-averse dual-channel supply chain: portfolio trade credit financing and platform investment

ABSTRACT This paper considers a dual-channel supply chain where the supplier can distribute products through an online platform and a capital-constrained traditional offline retailer. The retailer needs to fund its business by a portfolio financing of trade credit and platform investment to enter the market. With the price-dependent market uncertain demand, the supplier faces choices among direct, traditional, and dual channels for product distribution. We use the CVaR criterion to formulate the retailer’s risk-averse behaviour. Then derive the optimal retail channel strategy in a Stackelberg game model where the platform leads by setting a revenue-sharing rate. Our results unveil a Pareto improvement region in revenue sharing rates within the portfolio financing scheme. As the slotting fee escalates, the supplier tends towards traditional channels. Conversely, with a modest slotting fee, a threshold emerges for the retailer’s risk aversion, delineating the optimal retail channel. High retailer risk aversion favours the dual channel, while lower aversion steers all members towards the direct channel. Theoretical findings are validated through numerical analysis.

A study on the efficacy of an attention mechanism-based strategy planning method for stock trading portfolios

In order to take full advantage of the fact that there is a large amount of information and data available in the field of stock trading, this paper develops a new type of stock trading strategy, which reduces investment risk. A stock trading portfolio strategy planning method based on an attention mechanism, as disclosed in the present invention, relates to the field of stock trading technology. The introduction of an attention mechanism allows the strategy to focus on those stocks with higher evaluation coefficients, thereby improving the selectivity and accuracy of the strategy and thus the effectiveness of the portfolio strategy. Furthermore, based on the prediction results of the deep spatio-temporal neural network model, it is possible to determine the stock's short-term trends and identify the optimal trading points. This information can be utilized as a reference for investors, assisting them in making more informed decisions regarding the timing of buying and selling, improving the success rate and yield of trading, enabling them to swiftly and accurately assess the performance of each stock, facilitating timely adjustments to their portfolios, and enabling them to devise the optimal trading strategy and timing for each stock, enhancing the accuracy and efficiency of their trading decisions.

Online Portfolio Selection of Fuzzy Mean Regression Strategy Considering Investor Sentiment Based on Text Data

Investors are often affected by emotion, cognition, and other psychological factors in stock trading when making decisions. At present, people can use machine learning and other technologies to obtain a massive amount of text data from the Internet to mine information related to investor behavior and sentiment. Building intelligent online portfolio trading strategies that consider investor sentiment has become an important topic and key challenge in the financial field. Therefore, this paper explores how to use text data to depict investor sentiment, fuzzifies historical stock price data, designs a new weight transfer equation, and finally obtains a novel fuzzy mean regression strategy that considers investor sentiment based on text data. We conduct empirical tests on this strategy by using the stock price data selected from CSI300 constituent stocks, as well as the text data of investors’ opinions on the internet. The results show that the strategy proposed in this study has a higher Calmar ratio than other mean regression strategies previously studied.

A novel portfolio optimization method and its application to the hedging problem

Abstract In this article we will propose a novel, self-financing, dynamic and path dependent portfolio trading strategy which is based on the well known principle “sell high – buy low”. Trading strategies are important also for the hedging problem selling/buying an option. The main problem of the writer of an option is how to invest the amount that she has received selling the option therefore the proposed trading strategy play an important role here. We will see that the hedging problem reduces to an optimization one and therefore the portfolio optimization and the hedging problem are closely related. We will also propose a deterministic portfolio selection method (i.e., without making any assumption-guess about the assets) and a notion of a deterministic fair price of an option.

ABD Merkez Bankası'nın Para Politikası Kurallarından Sapmaları ve Bunların Yayılma Etkileri

The aim of this study is to examine the impact of the FED's deviations from monetary policy rules on the Turkish economy. In other words, the spillover effect of the FED's monetary policy was examined. The dataset used in the study was divided into two parts as rule-based and discretionary periods with reference to Taylor (2012), and the interaction between the variables was examined using structural VAR models. The results show that deviations from monetary policy rules have quite significant effects on the Turkish economy in the period when the FED is closer to discretionary policy. Considering that the economies of developing countries such as Turkey are more affected by the monetary policies of the FED than the developed countries, it can be said that transparent and predictable monetary policies are an effective tool to be protected from these global effects. In addition, the diversification of the foreign trade portfolio and the strengthening of foreign exchange reserves are also extremely important in making the Turkish economy resilient to the shocks resulting from the FED's monetary policies.

Investment management approaches: A comprehensive review of equity trading simulators, methodological challenges, and future directions

This paper provides a comprehensive review of equity trading simulators, focusing on their performance in assuring pre-trade compliance and portfolio investment management. A systematic search was conducted that covered the period of January 2000 to May 2023 and used keywords related to equity trade simulators, portfolio management, pre-trade compliance, online trading, and artificial intelligence. Studies demonstrating the use of simulators and online platforms specific to portfolio investment management, written in English, and matching the specified query were included. Abstracts, commentaries, editorials, and studies unrelated to finance and investments were excluded. The data extraction process included data related to challenges in modern portfolio trading, online stock trading strategies, the utilization of deep learning, the features of equity trade simulators, and examples of equity trade simulators. A total of 32 studies were included in the systematic review and were approved for qualitative analysis. The challenges identified for portfolio trading included the subjective nature of the inputs, variations in the return distributions, the complexity of blending different investments, considerations of liquidity, trading illiquid securities, optimal portfolio execution, clustering and classification, the handling of special trading days, the real-time pricing of derivatives, and transaction cost models (TCMs). Portfolio optimization techniques have evolved to maximize portfolio returns and minimize risk through optimal asset allocation. Equity trade simulators have become vital tools for portfolio managers, enabling them to assess investment strategies, ensure pre-trade compliance, and mitigate risks. Through simulations, portfolio managers can test investment scenarios, identify potential hazards, and improve their decision-making process.

Filled and Killed: Forecast and Realized Trading Costs across Horizons from Global Equity and Fixed Income Portfolio Trades

Filled and Killed: Forecast and Realized Trading Costs across Horizons from Global Equity and Fixed Income Portfolio Trades

Analyzing the nature of fund selection measures: Stock picking or trading skill?

We analyze why fund selection measures, including daily six factor alpha, daily return volatility, and minimum churn rate of a fund within a quarter, predict fund performance. Among these three fund selection measures, we demonstrate that funds with lower return volatility generate better fund performance, portfolio holdings performance, and trading portfolio performance. This result suggests that funds with lower return volatility possess better both stock picking and trading skill. In addition, funds with higher alpha or higher minimum churn rate have better fund performance and better portfolio holdings performance, but worse trading portfolio performance. As we demonstrate that portfolio holdings performance rather than trading portfolio performance mainly contributes to fund performance, our findings suggest that fund investors should pay attention to funds’ stock picking skill more than their trading skill. Moreover, funds’ lack of trading skill helps explain why original funds may underperform copycat funds.

Implied volatility is (almost) past-dependent: Linear vs non-linear models

We explore and demonstrate a clear pattern of past-dependence in predicting implied volatility, which extends up to twenty days and is present in both linear and nonlinear models. Empirically, we find that linear models utilizing precedent values up to 20 days explain up to 80% of the variance in the S&P 500 index, while nonlinear models explain 78%. Furthermore, we observe that L2 regularization, as opposed to L1, enhances the performance of linear models and underscores the importance of past-dependence. As a result, a simple linear model with L2 regularization, incorporating precedent values up to the past 20 days, consistently outperforms both linear and nonlinear benchmark models when applied to the S&P 500 index option and SSE 50 ETF option over the past decades. Additionally, we uncover that the impact of all preceding values exists but consistently diminishes over time in all empirical studies involving past-dependence settings. Finally, we demonstrate that these forecasts yield profitable outcomes when implementing a simple long-short portfolio trading strategy. This study primarily investigates the ‘path-dependency’ of implied volatility in financial markets and further confirms that simple models can sometimes surpass complex nonlinear models in predicting volatility.

Ensemble Machine Learning in Algorithmic Trading: Balancing Performance and Risk

This research delves into the potential of ensemble machine learning in algorithmic trading, aiming to strike a balance between performance and risk management. In the increasingly intricate landscape of financial markets, employing ensemble approaches that amalgamate multiple models becomes imperative. The study elucidates how these techniques enhance the accuracy, resilience, and adaptability of algorithmic trading systems. It further emphasizes the significance of model variability, hyperparameter tuning, and the bagging ensemble learning technique for optimizing outcomes. Moreover, the research scrutinizes the benefits of ensemble learning for algorithmic trading portfolios, particularly in terms of tail risk management and risk diversification. By utilizing ensemble approaches, several advantages emerge, including mitigated model bias, heightened generalizability, and an improved Sharpe Ratio. Ensemble methods enable traders to navigate market complexities more effectively, adapting to changing conditions and reducing the impact of outliers. This comprehensive approach not only enhances trading performance but also fortifies risk management strategies, making it indispensable in today's dynamic financial environment. Keywords: Ensemble machine learning, algorithmic trading, risk management, performance, Stock market forecast

Examining the relationship between international digital trade, green technology innovation and environmental sustainability in top emerging economics

Ensuring preserving a sustainable environment is a crucial concern for individuals worldwide. In previous research, CO2 emissions have been used to measure environmental deterioration. However, in this study, we have expanded the scope to include carbon emissions and several other gases. This comprehensive measure is referred to as the ecological footprint (EFP). More significant international digital trade (IDT) has the potential to achieve several positive results, including reducing EFP (economic frictions and barriers), stimulating economic growth, and minimizing trade risk and volatility. These benefits can be realized by implementing structural reforms in significant production and development sectors. Green technology innovation (GTI) has the potential to make substantial progress in ecological quality and energy efficiency. Nevertheless, previous studies still need to adequately prioritize examining rising economies in terms of international trade diversification and GTI. This study examined the effects of IDT, GTI, and renewable energy consumption (REC) on EFP in BRICST countries. The study utilized data from the period between 1995 and 2022. The cross-sectionally augmented autoregressive distributed lag (CS-ARDL) model demonstrates that EFP negatively correlates with trade diversification, REC, and GTI in the long and short term. These countries have demonstrated a significant presence of eco-friendly products in their trade portfolios, and their manufacturing processes are shifting towards GTI. The objective is to enhance the REC sources and minimize EFP from consumption. Conversely, the increasing economic growth within this economic group has a compounding impact on the environment's decline since it amplifies the carbon emissions from increased consumption. To reduce the EFP level, the paper suggests increasing investment in GTI, promoting worldwide digital trade, and embracing renewable energy sources.

DEVELOPMENT OF FINANCIAL RISK HEDGING STRATEGIES

The purpose of the study is to develop the theory of hedging, generalize approaches to the formation of effective risk hedging strategies, and provide recommendations for the implementation of international banks' experience in Ukrainian practice. It is proved that in the process of choosing effective strategies, it is advisable to distinguish three types of financial risk hedging: a) operational hedging (balance sheet) to ensure operational flexibility; b) market hedging (external), which involves the use of derivatives as instruments to hedge profit volatility; c) contract hedging (contractual clauses, embedded derivatives. Based on the analysis of the reports of 250 banks, it was found that Bank of America, JPMorgan Chase & Co., Goldman Sachs, Morgan Stanley, and Citigroup are the market makers in the most developed US futures market. Horizontal analysis of the balance sheets of these banks showed that although the share of derivatives designated as hedging instruments is insignificant (less than 1%) relative to term contracts recognized as trading instruments, the hedging effect is significantly higher than the gains on the trading portfolio. It was found that in international practice, the dominant direction is the hedging of interest rate risk (the share of which varies between banks from 42.52% to 61.26%), in the structure of hedging instruments, interest rate swaps prevail. The share of currency derivatives varies from 18.67% to 41.45%. Interest rate risk is hedged by over-the-counter bilateral term contracts for active operations, currency risk hedging equally covers the risks of passive and active operations. The regression analysis revealed that private credit and trading assets have an inverse effect on regional hedging exposure, and debt instruments have a direct effect. The study provides a basis for assessing the potential of using innovative risk-hedging instruments by Ukrainian banks.

A Framework for Sustainable and Green Finance Through Effective Trading Portfolio Management

This research paper responds to the growing global demand for environmentally and socially responsible financial practices by outlining a strong framework for incorporating sustainable and green finance into effective trading portfolio management. The study acknowledges the current difficulties of reconciling financial goals with sustainability criteria and uses a methodological approach that includes risk-sensitive asset allocation, mean-variance optimization, and strategic maximization of the Sharpe ratio. By carefully examining and analyzing this research, it explores the complex dynamics of sustainable finance, thus providing a holistic understanding of how financial success is related to environmental and social responsibility. The findings of this study provide important insights into ongoing discussions on responsible investment strategies, thereby giving investors and policymakers a guide on how to align their financial objectives with sustainable development imperatives.

Filled and Killed: Forecast and Realized Trading Costs Across Horizons from Global Equity and Fixed Income Portfolio Trades

Filled and Killed: Forecast and Realized Trading Costs Across Horizons from Global Equity and Fixed Income Portfolio Trades

Моделювання системи інтелектуального прогнозування торгів на фондових біржах

The article highlights the reasons for changes in the price quotations of financial assets on stock exchanges. The article models the process of a situation when a trader fixes the period of holding his trading position. It defines periods of buying and selling and, taking into account that high-frequency stock trading on ultra-short intervals shows low profitability, introduces an important condition that allows a stock trader to freely open and close trading positions during the entire period of buying and selling with consideration of the proposed restrictions. The article offers modelling of the trader's strategy of carrying out trading actions aimed at maximisation of profit. Taking into account the liquidity constraints and quantitative limitations for trading orders, the article proposes to determine the optimal high-frequency trading strategy for buying and selling by a trader, which can be formulated as the task of minimising the cost of trading orders. Based on the number of available exchange trade orders and the values relative to the respective trade order at specific moments, determining the optimal high-frequency trading strategy for buying and selling a trader can be reduced to solving a simple cost minimisation problem under the given conditions of liquidity constraints for each trade order, completion of the trading portfolio without active positions before the end of the period and the total number of exchange buy and sell transactions. The key phases in building the structure on which the stock trading strategy itself is based are described. The need to determine what data will be entered into the algorithm of the artificial neural network based on the input data and to determine which algorithm will be used for a particular task is established. The structure of the software model of the system for intelligent forecasting of trading on stock exchanges is designed. The complex of the automated trading system includes the development of a graphical display of quotes and a tool for visual analysis. At the same time, information about proven trading strategies can be stored in a database that can be added and deleted by traders in the developed intelligent system for forecasting trading on stock exchanges.

A portfolio trading system using a novel pixel graph network for stock selection and a mean-CDaR optimization for portfolio rebalancing

Due to the complexity and dynamic nature of financial markets, portfolio management tasks require continuous adaptation with market intelligence. Furthermore, to make profitable trading decisions, it is essential to predict the future performance of the market. To that end, this study proposes a dynamic portfolio trading system in two stages: stock trend prediction and portfolio optimization. In the first stage, although many researchers proposed powerful classification models to predict BUY/SELL trading signals, some scholars only focused on converting time series' features into multichannel images without considering any spatial relationships between features. Considering the relational capacity of graphs, this study proposes a novel image-based classification model, called Pixel Graph Network (PGN), to analyze the constructed images with propagating information from all neighboring pixels, build up a representation of the images as special graphs, and get advantages of a graph-based learning algorithm. In the second stage, stocks with predicted BUY trading signals in the following days, are fed into the Mean–Conditional Drawdown at Risk (M-CDaR) optimization model to prevent any significant drawdown of the investment portfolio. Moreover, a hierarchical feature selection algorithm has been proposed, combined with a continuous trend labeling method to improve the efficiency of the training process. To demonstrate the superiority of our proposed PGN model in comparison with benchmarks, 18 stocks from the New York Stock Exchange (NYSE) are used for numerical experiments. The obtained results show the improvement and effectiveness of the proposed model in both stock trend prediction and portfolio optimization. In terms of financial performance, the proposed PGN+M-CDaR portfolio-based trading system achieves an average annual return of 112%, an average annual Sortino ratio of 10.78, and an average annual Sharpe ratio of 2.79 which are promising to achieve substantial profits while mitigating downside risk.

Systemic Risk in the Global Energy Sector: Structure, Determinants and Portfolio Management Implications

We examine the dynamics of tail dependence across returns of 105 global energy firms from 26 countries covering the regions of America, Asia Pacific and Europe. A partial correlation-based approach is used to quantify the dependence structure and level of systemic risk under relatively stable and extremely bearish and bullish market conditions. The dependence network of energy stock returns is constructed based on the novel triangulated maximally filtered graph (TMFG). The results reveal a high degree of tail dependence and role played by geographical proximity. The strongest links are found under extreme bearish market conditions. American and European energy firms are more interconnected and contribute more to systemic risk than Asian-Pacific companies. The dependence intensifies during periods of market turmoil, especially during the COVID-19 pandemic. A higher instability in the dependence structure is observed during extremely bearish market circumstances. A simple portfolio trading strategy based on the dependence ranking of energy firms outperforms a naïve equally-weighted buy-and-hold portfolio strategy.

The Impact of Different Trading Strategies on the Same Portfolio

The frequent occurrence of political events, the spread of the epidemic and changes in policies of various countries have made the market more volatile. Due to the instability of the overall economic environment, investors are increasingly focusing on using strategies to improve returns, so this article selects several hedge fund trading strategies, analyzes the same trading portfolio, and compares the performance of different trading portfolios. Different effects of the same combination give spirits to investors when making transactions. The market neutral trading strategy is effective for the trading combination of F and NVDA. Based on different data, the long-short equity trade strategies are not completely effective, and the credit arbitrage trade strategy is not suitable for this trading combination. Market neutral trading strategies and long-short equity trade strategies basically meet the combination of F and NVDA, but due to the possibility of credit rating fluctuations, credit arbitrage trade strategy does not meet this combination. These different effects reflect some of the underlying factors in the market, so the analysis in this article will improve investors' understanding of investment strategies and change their previous trading thoughts. Taking these effects into consideration can help investors effectively avoid potential risks, make better choices, or obtain more returns.

Prediction of clinical trials outcomes based on target choice and clinical trial design with multi-modal artificial intelligence.

Drug discovery and development is a notoriously risky process with high failure rates at every stage including disease modeling, target discovery, hit discovery, lead optimization, preclinical development, human safety and efficacy studies. Accurate prediction of clinical trial outcomes may help significantly improve the efficiency of this process by prioritizing therapeutic programs that are more likely to succeed in clinical trials and ultimately benefit patients. Here we describe inClinico, a transformer-based artificial intelligence software platform designed to predict the outcome of Phase II clinical trials. The platform combines an ensemble of clinical trial outcome prediction (CTOP) engines that leverage generative artificial intelligence and multimodal data, including omics, text, clinical trial design, and small molecule properties. inClinico was validated in retrospective, quasi-prospective, and prospective validation studies internally and with pharmaceutical companies and financial institutions. The platform achieved 0.88 ROC AUC in predicting the Phase II to Phase III transition on a quasi-prospective validation dataset. The first prospective predictions were made and placed on date-stamped preprint servers in 2016. To validate our model in a real-world setting, we published forecasted outcomes for several Phase II clinical trials achieving 79% accuracy for the trials that have read out. We also present an investment application of inClinico using date stamped virtual trading portfolio demonstrating 35% 9-month return on investment (ROI).