Hedging in Jump Diffusion
The study applies a jump-diffusion risky asset model to calculate the hedging strategy for a European call option, using a decision tree, table of values, and figures.
6 sharesSource ↗
Quant LetterNo. 62
156 items across 10 sections, as sent to readers on 21 August 2024. Paper titles open their ML-Quant page; ↗ goes to the source.
Quantitative-finance and ML-for-finance preprints from arXiv.
15 items
The study applies a jump-diffusion risky asset model to calculate the hedging strategy for a European call option, using a decision tree, table of values, and figures.
6 sharesSource ↗
The research introduces two new indexes for optimal portfolio selection in fluctuating financial markets, proving their superior accuracy and performance in both simulated and real markets.
6 sharesSource ↗
The paper presents an extended approach to solve large-scale robust portfolio optimization problems, showing robust trading performance and significantly reduced computational time.
6 shares1 citation todaySource ↗
The article investigates the causes and effects of periodic trading activities in equity markets, establishing a unique open-loop Nash equilibrium and providing new insights into market dynamics.
4 shares6 citations todaySource ↗
The article explores the importance and challenges of using infinite-mean models in economics and finance, particularly when dealing with heavy-tailed datasets.
3 shares16 citations todaySource ↗
A novel framework for financial time series forecasting is presented, using causality models to improve prediction accuracy, especially in unstable markets.
2 shares8 citations todaySource ↗
The paper studies optimal insurance solutions using the Lambda-Value-at-Risk model, revealing that a truncated stop-loss indemnity is ideal under certain conditions and discusses the effect of model uncertainty.
2 shares7 citations todaySource ↗
The research analyzes the effect of environmental transition on asset value, using a decision-making model to determine optimal divestment decisions.
2 shares2 citations todaySource ↗
A study reveals that a deep neural network can predict human decisions in two-player matrix games better than existing theories, with human response varying based on game complexity.
4 shares22 citations todaySource ↗
The study highlights the crucial role of open labeled datasets like CIFAR-10 in the rise of Deep Learning, emphasizing its impact on computer vision, object recognition, and its importance in teaching machine learning techniques.
2 shares2 citations todaySource ↗
The research uses a simulator with AI agents to study the effects of tax credits on various households, suggesting a new distribution method to lessen inequality.
3 shares4 citations todaySource ↗
Deep-MacroFin, a new framework that employs deep learning to solve complex equations in economics, is introduced, providing a more user-friendly alternative to existing tools.
2 shares2 citations todaySource ↗
The research investigates the application of unsupervised clustering in investment management, showing that accurate classification is achievable with appropriate feature selection and distance metric.
2 shares5 citations todaySource ↗
The study presents a mathematical model for informational persuasion, offering a theoretical foundation for examining AI's influence on industries and explaining why people can be persuaded even when all information is publicly accessible.
2 shares4 citations todaySource ↗
Explore-then-Commit Algorithm: The article introduces a new algorithm for online learning in decentralized two-sided matching markets. This algorithm doesn't need prior knowledge of preference rankings or agent communication and effectively minimizes player regret.
4 shares8 citations todaySource ↗
Working papers in finance and economics from SSRN.
29 items
The article discusses a new method for enhancing malware detection and defense mechanisms by combining a Knowledge Graph with Deep QNetworks to predict future cyber attack patterns.
24 sharesSource ↗
The study examines the effects of central bank balance sheet policies on financial stability, concluding that while they help stabilize the economy during financial stress, they also prolong and increase the likelihood of such episodes.
15 sharesSource ↗
The research indicates that the growth of passive investing increases stock correlations and market volatility, reducing the advantages of diversification and increasing market risk during crises.
3 sharesSource ↗
The study introduces machine learning strategies informed by physics to predict sea ice velocity and concentration in the Arctic Ocean, performing better than purely data-driven models, particularly during rapid melting and freezing periods.
2 sharesSource ↗
The paper presents a machine learning-based probabilistic prediction model to estimate the failure modes of reinforced concrete columns under earthquake load, taking into account the uncertainty due to randomness in column members and dynamic loads.
2 sharesSource ↗
The paper presents a machine learning system that uses past user login data to detect and block suspicious login attempts.
2 shares2 citations todaySource ↗
The article suggests a numerical method that merges different techniques to improve the estimation of the optimal portfolio.
2 sharesSource ↗
The study reveals that cryptocurrency returns are significantly affected by fluctuations in advanced country currencies and oil prices, but not by emerging country currencies or global markets.
2 sharesSource ↗
The research shows that domestic economic policy uncertainty greatly affects sovereign credit risk, but has a minor impact on sovereign bond yields and consumer confidence.
4 sharesSource ↗
The study employs a semi-supervised machine learning model to assess the social value of Chinese state-owned enterprises, showing an upward trend and notable differences across sectors and years.
2 sharesSource ↗
The study finds that in the Korean stock market, speculative sentiment in leverage exchange-traded funds (ETFs) negatively correlates and predicts market returns after 3 months.
4 sharesSource ↗
The research shows that machine learning and cloud computing significantly enhance scalability and reduce testing time in software regression testing.
2 sharesSource ↗
The paper highlights the importance of understanding stock market volatility clusters for informed investment decisions and risk management.
2 sharesSource ↗
The article underscores the growing significance of Sentiment Analysis in business, enabling companies to leverage customer opinions for growth.
2 sharesSource ↗
The study reveals a correlation between extreme investor sentiment from traditional financial markets and a negative Bitcoin basis during unexpected inflation and deflation periods.
2 sharesSource ↗
The article explores the use of Identity Access Management (IAM) and machine learning to improve data security in cloud-based fleet management applications.
2 shares5 citations todaySource ↗
The paper discusses the advantages and challenges of data-driven marketing, and how it can be used to improve customer experience and strengthen business relationships.
2 sharesSource ↗
The study investigates the importance of online employee reviews in academic research fields like finance and economics, and their impact on understanding company performance and workplace culture.
2 sharesSource ↗
The article examines the use of artificial intelligence in transforming fleet financing, with a focus on fair pricing and improving decision-making in real-time marketplaces.
2 shares7 citations todaySource ↗
The paper addresses the issue of thread deadlocks in cloud-based applications, suggesting an AI and machine learning system to predict and manage these deadlocks, improving system reliability.
2 shares8 citations todaySource ↗
Alternative data is vital for investment strategies, providing unique insights and competitive advantages, but requires constant adaptation due to its ever-changing nature.
4 shares2 citations todaySource ↗
A study proposes using clean cryptocurrencies to create mimicking portfolios as a cost-effective hedging tool for ESG investors, compared to Bitcoin.
2 sharesSource ↗
Research on 56 frontier and emerging markets reveals that global financial conditions and debt-to-GDP ratio significantly influence access to international capital markets, with IMF reforms playing a key role.
3 sharesSource ↗
A Norwegian study reveals widespread crypto tax evasion, even among investors on exchanges that share data with tax authorities, indicating the need for effective and affordable enforcement strategies.
3 sharesSource ↗
The Information Ratio Churn (IRC) metric is introduced to calculate a fair fee for active management, with data indicating that higher IRC leads to lower realized Information Ratio, and 8% of assets are invested in funds with high IRC.
2 sharesSource ↗
The article presents a model that explains how traders' strategies interact with uneven market closures, leading to spikes in trading volume and return volatility due to liquidity trading accumulation.
2 sharesSource ↗
The study finds that institutions are more likely than individuals to strategically offer liquidity during price jumps, based on data from the Taiwan Stock Exchange.
2 sharesSource ↗
The paper reveals that retail market making in the German equity market is highly profitable due to reduced exposure to adverse selection and inventory risk, based on regulatory data.
2 sharesSource ↗
The study rejects standard corporate bond factor models in favor of a model featuring the global corporate bond market, a global maturity spread factor, and a global liquidity spread factor, based on research in the six largest international markets.
2 sharesSource ↗
Economics working papers from RePEc's NEP field reports.
29 items
The research shows the effectiveness of modeling compositional volatility, using German political party support and US income shares data as examples.
22 sharesSource ↗
Fintech lending platforms like Funding Circle and LendingClub have successfully provided loans to high-risk areas, filling the credit gap left by traditional lenders.
22 sharesSource ↗
A new hedge strategy has been developed to reduce carbon risk in diversified portfolios without significant losses in returns, making it a feasible option for investors and fund managers.
21 sharesSource ↗
The PCA and LSTM methods are proven to be effective in predicting Poland's yield curve, with PCA-LSTM being especially accurate in short-term forecasting of long-term interest rates.
16 sharesSource ↗
The research finds a negative link between distress risk and corporate profitability in Vietnam's stock market, which vanishes after bankruptcy regulations are implemented.
15 sharesSource ↗
The study reveals significant changes in volatility and day-of-the-week effects on cryptocurrency returns, especially during the COVID-19 pandemic.
15 sharesSource ↗
The paper introduces a new method for determining arbitrage-free conditions for a parametric yield curve considering market prices of risk, and presents bond portfolio optimization as a stochastic control issue.
14 sharesSource ↗
The research shows that political instability notably increases stock return volatility in BRICS countries and Turkey, with Turkey being particularly affected.
13 sharesSource ↗
The research introduces a machine learning model that uses multicriteria optimization to reduce bias in data fitting, tested on digit classification.
25 sharesSource ↗
The research shows that simple machine learning methods like the KNN model can predict bank failures effectively, especially when used with PCA.
20 sharesSource ↗
The research uses machine learning to study the impact of personality traits and financial literacy on Private Pension System participation in Turkey.
15 sharesSource ↗
The research suggests a new method for predicting the severity of marine accidents using machine learning, with the Light Gradient Boosting Machine model performing the best.
14 sharesSource ↗
The research reviews studies on using Google Search Volume Index to measure investor attention and stock market trends, finding a correlation between increased investor attention and market volatility.
12 sharesSource ↗
Research shows machine learning models using unconventional data are better at predicting credit losses and defaults, particularly during economic crises.
30 sharesSource ↗
A new binary crayfish optimization algorithm (IBCOA) has been created to improve feature selection in data mining and machine learning, enhancing classification accuracy.
19 sharesSource ↗
A study indicates that pessimistic future narratives, found through text mining of news articles, affected the distribution of Greek bonds during its debt crisis from 2009 to 2015.
17 sharesSource ↗
The article advocates for the use of machine learning in international business research to handle complexity and aid theory development.
16 sharesSource ↗
The study introduces a machine learning-based method for creating property price indices, which offers better accuracy but requires bias reduction measures.
15 sharesSource ↗
The paper suggests a machine learning model for smart ports to dynamically reschedule truck appointments using real-time data.
14 sharesSource ↗
The study introduces variable selection with random forests, a machine learning method, and compares it with linear models, providing practical usage recommendations.
14 sharesSource ↗
The research compares risk tolerance of institutional investors in China and the US, finding that US investors are more risk-averse.
15 sharesSource ↗
The study reveals that long-term exposure to high volatility leads to underestimation of volatility, which can be exploited for stock return predictability.
13 sharesSource ↗
Consequences: The research suggests that 1.23% of global GDP is laundered annually, negatively impacting economic and financial indicators, except inflation rates.
10 sharesSource ↗
The study concludes that long-term country equity premium forecasts are more accurately predicted using a cross-sectional global factor model than time-series prediction models.
10 sharesSource ↗
Insider trading can reveal the value of all securities held by an individual, with unsold stocks performing better than unbought ones, indicating that even sales driven by liquidity needs can offer valuable insights.
8 sharesSource ↗
Investment strategies that incorporate momentum and valuation styles in leveraged loans yield significant returns, implying that credit managers not using these methods are missing potential profits.
6 sharesSource ↗
The study explores the influence of AI and digitalization on the macroeconomics of EU countries, including the relationship between AI usage and GDP per capita, labor productivity, and the proportion of IT professionals in employment.
1 sharesSource ↗
Companies with higher agency costs tend to hire lower-tier auditors, but this trend decreases when the board has more financial experts; however, the quality of the internal control system doesn't affect this correlation.
1 sharesSource ↗
Factors such as generational differences, gender, education, and location significantly affect financial inclusion in Kenya, with Generation Y having more access to financial services than Generation X, and women and rural residents generally having less access.
1 sharesSource ↗
The general machine-learning papers the letter carried in 2023-25.
19 items
The article presents Transfusion, a new method for training multi-modal models on both discrete and continuous data, which performs better in terms of scalability and generation of images and text.
313 shares480 citations todaySource ↗
The study introduces a scaling law for predicting the loss of neural language model training at any step and learning rate, aiding researchers in choosing optimal learning rate schedules.
53 shares31 citations todaySource ↗
The paper introduces BAM, a new method for training Mixture of Experts models, which optimizes the use of specialized dense models and improves performance in perplexity and downstream tasks.
47 shares18 citations todaySource ↗
The authors present LongVILA, a solution for long-context vision-language models, which includes a system for Multi-Modal Sequence Parallelism, a five-stage training pipeline, and large-scale pre-training datasets, improving performance on long videos.
45 shares345 citations todaySource ↗
The research evaluates the ability of large language models to understand symbolic graphics programs, introducing a benchmark for semantic understanding and a method called Symbolic Instruction Tuning to enhance this capability.
37 shares42 citations todaySource ↗
The article presents SpaRP, a new method for 3D reconstruction and camera pose estimation from sparse-view images, offering superior quality, accuracy, and efficiency.
19 shares46 citations todaySource ↗
NeuRodin, a two-stage neural surface reconstruction framework, is introduced, providing high-quality surface reconstruction while maintaining the optimization flexibility of density-based methods.
18 shares20 citations todaySource ↗
The research investigates the use of four deep neural networks for atmospheric tracer transport modeling, with the SwinTransformer demonstrating excellent emulation capabilities for multi-year forward runs.
15 shares2 citations todaySource ↗
The article presents a new variable importance algorithm, Shapley Marginal Surplus for Strong Models, which surpasses other feature importance methods in inferential abilities.
14 sharesSource ↗
Financial Regularities: The article introduces PLUTUS, a pre-trained transformer-based model for financial time series modeling, setting a new benchmark in the field with its superior performance.
13 shares3 citations todaySource ↗
Large language models (LLMs) have shown a high probability of generating accurate causal arguments, outperforming existing methods, and can assist experts in causal analysis, despite occasional unpredictable failures.
1,546 shares571 citations todaySource ↗
A new framework has been introduced to integrate symmetry into machine learning models, enabling the enforcement of known symmetries, discovery of unknown ones, and promotion of symmetry during training.
796 shares37 citations todaySource ↗
A program using a modern language model, GPT-4, has demonstrated the ability to write code that can self-improve, showcasing the potential of language-model-infused scaffolding.
404 shares144 citations todaySource ↗
The Idea to Image system, utilizing GPT-4V(ision), allows for efficient conversion of abstract ideas into text-to-image prompts, resulting in images with superior semantic and visual qualities.
183 shares37 citations todaySource ↗
Research suggests that Large Language Models (LLMs) can be compromised by benign data, proposing a bi-directional anchoring method to identify such data and maintain model safety.
149 shares124 citations todaySource ↗
The creation of deep learning architectures can be streamlined using an end-to-end design pipeline, which uses synthetic tasks to predict scaling laws and find optimal architectures.
130 shares66 citations todaySource ↗
Multilingual Medical LLM: A multilingual medical dataset and benchmark have been created to expand medical AI to non-English speakers, with the Apollo models showing the best performance.
104 shares59 citations todaySource ↗
An autonomous agent has been developed to enhance the zero-shot reasoning capabilities of large language models, achieving top performance on various tasks.
93 shares46 citations todaySource ↗
Multimodal Chemistry Model: ChemVLM, a chemical multimodal large language model, has been introduced to manage visual information in the chemical field, showing competitive performance in different tasks.
58 shares103 citations todaySource ↗
Papers that shipped their code, from the Papers with Code feed (2023-25).
11 items
Automated Scientific Discovery: The article explores the use of AI in scientific research to solve complex problems.
4,522 shares
10K Word Generation from LLMs: The piece details the successful expansion of model output length to over 10,000 words without compromising quality.
315 shares
The paper presents Selective Context, a technique that enhances the efficiency of LLMs by eliminating redundancy.
261 shares
Theorem Proving Model: The article showcases DeepSeekProverV1.5, a language model designed for theorem proving in Lean 4.
133 shares
Neural Radiance Field Fruit Counter: The article presents FruitNeRF, a new framework that uses view synthesis to count any type of fruit in 3D.
116 shares
The paper presents a new method, 1.5Pints, for pretraining a Language Model that surpasses top models in instruction-following tasks using a specially compiled pretraining dataset of 57 billion tokens.
116 shares
The article explores the difficulties researchers encounter in staying updated with new developments in their fields due to the fast-paced expansion of scientific literature.
97 shares
The article emphasizes the recent advancements in neural information retrieval models, enhancing their performance in different information retrieval tasks.
75 shares
Despite the rise of pretrained and large language models, the BM25 search algorithm remains crucial in information retrieval.
61 shares
The model performs comparably to other open-source models that are trained with hundreds of thousands of hours of labeled speech data.
56 shares
The paper introduces PeriodWaveTurbo, a high-quality waveform generation model created using adversarial flow matching optimization.
48 shares
Repositories the letter featured.
10 items
Packt has released a guidebook on using Python for algorithmic trading.
35 shares
GPU-Accelerated Order Book Simulator: JAXLOB is a GPU-boosted simulator for limit order book, improving large scale trading reinforcement learning.
94 shares
Stock Market Prediction Library: Boar bear is a Python tool that employs deep learning to predict and model stock market trends.
2,018 shares
Jupyter notebooks are utilized to aid in creating graph data science blog posts on httpsbratanictomaz.medium.
1,182 shares
The Prometheus system is a tool for monitoring and storing time series data.
54,335 shares
OpenHands provides a platform that simplifies coding and increases productivity.
30,226 shares
The article talks about a library designed for creating high-quality Text-to-Speech models.
3,761 shares
The piece investigates the application of genetic programming in Python using a scikitlearn-inspired API.
1,578 shares
BAML presents a new templating language for creating typed LLM functions, as shown in the promptfiddle.
611 shares
Industry news: funds, hiring, markets and regulation.
20 items
Pamalican Asset Management, a new hedge fund focusing on equity capital markets, has been licensed by Hong Kong's Securities and Futures Commission.
7 shares
A MarketWatch report reveals that megacap tech stocks remain the main focus of the top 20 equity positions of hedge funds.
5 shares
Preqin's Investor Outlook H2 2024 report indicates that more institutional investors plan to reduce their hedge fund portfolio exposure.
5 shares
Strategic Vision Investment, a Hong Kong-based hedge fund, has shut down its primary Value Multiplier Fund that focused on Chinese equities.
4 shares
GoldenTree Asset Management has appointed Avineet Punhani as Principal, the firm manages approximately 55bn in assets.
4 shares
Mars is buying Kellanova for $35.9bn in cash after negotiating with TOMS Capital Investment Management.
4 shares
Digital asset investment funds saw a minor inflow of $30m last week due to predictions of the US Federal Reserve not cutting interest rates.
3 shares
US public pension plans, including the Virginia Retirement System and Texas County and District Retirement System, have profited from hedge fund strategies.
3 shares
Hedge funds outperformed the S&P500 in July due to strong returns from the financial sector and diversification away from megacap tech stocks.
3 shares
Trend-following commodity trading adviser hedge funds have shifted from a bearish outlook to buying Japanese equities again last week.
2 shares
Paloma Partners has hired Pei Ng, a former executive at Brevan Howard Asset Management, as its new CFO.
2 shares
Citigroup Inc has noted a change in hedge funds' carry trade strategies, with a preference for borrowing US dollars over Japanese yen for investment in emerging market currencies.
2 shares
Bitwise Asset Management has purchased ETC Group, a London-based firm, expanding its portfolio with nine European-listed crypto ETPs.
2 shares
Bridgewater Associates has continued to decrease its investment in Chinese stocks, marking the seventh consecutive quarter of selling off US-listed Chinese holdings in Q2 2024.
1 shares
The article predicts a potential decline in Leetcode's dominance in the near future.
1 shares
ExodusPoint Capital Management's assets under management dropped by $1bn to $11.04bn in the first half of 2024.
1 shares
Elliott Management is initiating a proxy fight to gain control of 10 out of 15 board seats at Southwest Airlines.
1 shares
Trian Fund Management has sold approximately 3.8m shares in Unilever, reducing its stake by nearly £181m ($232m).
1 shares
Episodes on markets, quant methods and economics.
10 items
Market analyst Mike Silva discusses technical and intramarket analysis, risk management during volatile periods, and the relationship between oil prices, 10-year yields, and geopolitical risks on his YouTube channel.
19 shares
Katie Stockton from Fairlead Strategies talks about the recent increase in market volatility, the shift from high beta growth stocks to value stocks, and the potential resurgence of the energy and materials sectors.
16 shares
Cristian deRitis of Moody’s Analytics discusses the evolution of stress testing, current trends, and challenges for banks and regulators, including the need for wider regulatory stress tests and the use of AI models.
9 shares
Brian Peltonen, from Fidelity and Parcosm, shares his journey from game programmer to data analytics expert, and his vision for his new startup.
8 shares
JP. Morgan's Phoebe White and Mike Hanson discuss the July CPI report, future Fed policy, and their views on rates and inflation markets after recent volatility.
8 shares
Investor Danny Moses highlights the significance of gold in economic instability, the effects of increasing oil prices on global economies, and potential investment opportunities in online gambling and cannabis sectors.
7 shares
Arindam Sandilya, Ben Jarman, and James Nelligan explore the connection between central bank easing cycles and currencies, especially in light of anticipated Federal Reserve cuts.
6 shares
JC Parets, founder of All Star Charts, is recognized for his influence as a technical analyst and his contributions to investing, particularly his Chartered Market Technician designation.
6 shares
AntiBubbles: Diego Parilla, CIO of Quadriga, shares his perspective on the equity market, inflation, commodities, and precious metals in a discussion with MacroVoices' Erik Townsend and Patrick Ceresna.
5 shares
Lincoln Archibald, a managing director at Fund Launch Partners, stresses the need to maintain low operating costs when starting an alternative investment fund, and discusses other factors that lead to the success of new hedge funds.
4 shares
Posts from quant and economics blogs and newsletters.
6 items
The article offers advice on improving forex trading strategies for all levels of traders.
5 shares
The piece stresses the need to comprehend different trading strategies in the financial market.
3 shares
The study presents a novel technique using autoencoders to improve signal to noise ratio in financial data, underlining the importance of group structures in financial data analysis.
3 shares
The article introduces a novel technique to enhance the signal to noise ratio in financial data through autoencoders, offering a new way to detect patterns in financial timeseries.
3 shares
The research explores the framing effect in trading bias, a mental bias that can influence traders' choices by altering the presentation of information.
2 shares
The guide offers trading advice for the CADCHF pair, featuring live rates, major influencing elements, and latest performance.
2 shares
Posts from quant researchers on X.
7 items
AlphaSimplex's white paper highlights the diminishing role of bonds as a hedge due to increased volatility and a positive correlation with stocks.
5 shares
Lee and colleagues propose that a long/short portfolio based on shifts in media narratives can generate substantial alpha, as investors often underreact to these changes.
3 shares
The article explores the fluctuation of order book liquidity and its influence on the market at the micro-order level.
3 shares
New research on quantitative investing, discussing topics like cryptocurrency, foreign exchange predictability, and market forecasting, is now available.
2 shares
An updated version of the research paper on dynamic asset allocation with specific regime forecasts has been published.
2 shares
A new guide on replicating trend following managed futures has been released.
1 shares
A new article examining recent studies on recession predictability and the effectiveness of recession indicators for market timing has been published.
0 shares