---
title: Quant Letter No. 33: January 2024, Week 3
url: https://www.ml-quant.com/issues/2024-01-17/
site: ML-Quant (https://www.ml-quant.com)
updated: 2026-09-26
license: Summaries CC BY 4.0; links go to the original sources
index: https://www.ml-quant.com/llms.txt
issue_date: 2024-01-17
---


# Quant Letter No. 33: January 2024, Week 3

Sent 2024-01-17. 62 items.

## arXiv

### Finance

- __[Cash management models for data fitting](http://dx.doi.org/10.1016/j.cor.2018.04.007)__: The article introduces a novel method for cash management models using stochastic and linear programming, showing that a small random data sample can effectively fit these models. (2024-01-16, shares: 9) · https://www.ml-quant.com/papers/doi/10-1016-j-cor-2018-04-007/
- __[Super-hedging pricing and Immediate-Profit arbitrage](http://arxiv.org/abs/2401.05713)__: The study explores the super-hedging pricing valuation issue in different financial contexts, emphasizing the effect of changes in prior information and the growth of super-hedging prices under uncertainty. (2024-01-11, shares: 5) · https://www.ml-quant.com/papers/arxiv/2401.05713/
- __[Efficiency of graph databases for financial analysis](http://arxiv.org/abs/2401.07483)__: The study contrasts SQL, No-SQL, and graph databases in terms of efficiency and performance, concluding that ESG's Graph database is superior for extended analytics in business and investment. (2024-01-15, shares: 4) · https://www.ml-quant.com/papers/arxiv/2401.07483/
- __[Quantum probability asset return modeling](http://arxiv.org/abs/2401.05823)__: The article suggests a new approach to quantum finance, connecting quantum probability's mathematical structure to traders' decisions and market behaviors, and formulating a Schrödinger-like trading equation to describe the multimodal distribution of asset returns. (2024-01-11, shares: 4) · https://www.ml-quant.com/papers/arxiv/2401.05823/
- __[SpotV2Net: Intraday Volatility Forecasting](http://arxiv.org/abs/2401.06249)__: Intraday Volatility Forecasting: The article introduces SpotV2Net, a new model for predicting intraday spot volatility using a Graph Attention Network, which has shown better accuracy in predicting Dow Jones Industrial Average index prices. (2024-01-11, shares: 4) · https://www.ml-quant.com/papers/arxiv/2401.06249/
- __[Equity Auction Dynamics: Liquidity Models](http://arxiv.org/abs/2401.06724)__: Liquidity Models: The study applies the latent/revealed order book framework to equity auctions, showing no indicative price predictability and providing accurate model parameter measurements. (2024-01-12, shares: 3) · https://www.ml-quant.com/papers/arxiv/2401.06724/
- __[Dynamic Portfolio Selection with Disappointment Aversion](http://arxiv.org/abs/2401.08323)__: The research discusses portfolio selection under generalized disappointment aversion (GDA), finding that investment in the stock market is consistently lower under GDA than under traditional Expected Utility theory. (2024-01-16, shares: 3) · https://www.ml-quant.com/papers/arxiv/2401.08323/
- __[Longstaff Schwartz Monte Carlo Approach to Game Option Pricing](http://arxiv.org/abs/2401.08093)__: The article suggests a two-step Longstaff Schwartz Monte Carlo method for pricing game options, which provides more reliable results than the original method. (2024-01-16, shares: 2) · https://www.ml-quant.com/papers/arxiv/2401.08093/
- __[Deep Minimizing Movement Method for Option Pricing](http://arxiv.org/abs/2401.06740)__: The paper introduces a deep learning method for pricing European basket options using Artificial Neural Networks and two methods for discretizing the integral operator, focusing on assets with jump-diffusion dynamics. (2024-01-12, shares: 2) · https://www.ml-quant.com/papers/arxiv/2401.06740/

### Miscellaneous

- __[Enhancing Financial Sentiment Analysis with Heterogeneous LLM Agents](http://arxiv.org/abs/2401.05799)__: A study suggests using large language models without fine-tuning for financial sentiment analysis, offering a design framework that enhances accuracy. (2024-01-11, shares: 4) · https://www.ml-quant.com/papers/arxiv/2401.05799/
- __[Analyzing Herd Behavior in Investment](http://arxiv.org/abs/2401.07183)__: A study introduces the concept of average deviation to measure the difference between two agents' investment decisions, studying the effect of herd behavior on these decisions. (2024-01-14, shares: 2) · https://www.ml-quant.com/papers/arxiv/2401.07183/

### Crypto & Blockchain

- __[Backrun Auctions & Trader Protection](http://arxiv.org/abs/2401.08302?utm_source=dlvr.it&utm_medium=twitter)__: The study presents a new laminated queueing model for batched trading on decentralized exchanges, aiming to improve transaction infrastructure and examining the potential for price manipulation by arbitrageurs. (2024-01-16, shares: 5) · https://www.ml-quant.com/papers/arxiv/2401.08302/
- __[Transformer-Based ETH Price Prediction](http://arxiv.org/abs/2401.08077?utm_source=dlvr.it&utm_medium=twitter)__: The research uses a transformer-based neural network to forecast Ethereum prices, indicating a strong correlation with other cryptocurrencies and sentiments, and suggests a theory on sentiment-driven illusion of causality in cryptocurrency price movements. (2024-01-16, shares: 4) · https://www.ml-quant.com/papers/arxiv/2401.08077/
- __[Analysis of Impermanent Loss in DEX](http://arxiv.org/abs/2401.07689)__: The paper explores the issue of impermanent loss in decentralized exchanges through Monte Carlo simulations, indicating that price changes don't always result in losses for liquidity providers and that an arbitrage-friendly environment is beneficial for them. (2024-01-15, shares: 3) · https://www.ml-quant.com/papers/arxiv/2401.07689/
- __[Liquidity Provision on DEX](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4694683)__: Decentralized exchanges' infrastructure can lead to arbitrage losses for liquidity providers, with design changes offering limited reduction in these losses, as shown in a study using the Silicon Valley Bank collapse. (2021-03-17, shares: 2) · https://www.ml-quant.com/papers/ssrn/4694683/

### Historical Trending

- __[Handling Missing Values in ML Portfolios](https://arxiv.org/abs/2207.13071)__: The research analyzes missing data in return predictors, concluding that simple imputation methods are effective and complex ones can underperform if misused. (2022-07-21, shares: 42) · https://www.ml-quant.com/papers/arxiv/2207.13071/
- __[Deep Signature Algorithm for Options](https://arxiv.org/abs/2211.11691)__: The research expands the backward scheme for state-dependent FBSDEs with reflections to path-dependent FBSDEs, demonstrating the convergence of the numerical algorithm and providing examples of its use. (2022-11-21, shares: 14) · https://www.ml-quant.com/papers/arxiv/2211.11691/
- __[Curriculum and Imitation Learning for Time-Series Control](https://arxiv.org/abs/2311.13326)__: The study finds that curriculum learning enhances performance in control tasks over highly stochastic time-series data, while imitation learning should be used carefully. (2023-11-22, shares: 19) · https://www.ml-quant.com/papers/arxiv/2311.13326/
- __[Kernel Hilbert Space Approach to Volatility Models](https://arxiv.org/abs/2203.01160)__: The paper presents a new regularization approach for solving the singular McKean-Vlasov equation, commonly used in financial models, using the reproducing kernel Hilbert space technique. (2022-03-02, shares: 17) · https://www.ml-quant.com/papers/arxiv/2203.01160/

## SSRN

### Quantitative

- __[SpotV2Net: Intraday Spot Volatility Forecasting](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692194)__: Intraday Spot Volatility Forecasting: SpotV2Net, a new forecasting model based on Graph Attention Network architecture, enhances the accuracy of intraday spot volatility predictions for financial assets. (2024-01-11, shares: 2) · https://www.ml-quant.com/papers/ssrn/4692194/
- __[Model Averaging & Double Machine Learning](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4691169)__: The article presents two new stacking methods for double-debiased machine learning (DDML), showing its robustness against unknown functional forms, with software available in Stata and R. (2024-01-11, shares: 3) · https://www.ml-quant.com/papers/ssrn/4691169/
- __[Data Preparation for Code Smell Detection: A Literature Review](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4693778)__: A Literature Review: The review examines data preparation techniques in deep learning-based code smell detection, suggesting ways to prepare high-quality data and emphasizing the need for data diversity, standardization, and accessibility. (2024-01-13, shares: 2) · https://www.ml-quant.com/papers/ssrn/4693778/
- __[Real estate valuation with prototype-based models and machine learning](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4695079)__: The article introduces a novel real estate valuation model that uses prototype-based learning, a method that compares properties to similar ones, unlike traditional machine learning methods. (2023-12-22, shares: 3) · https://www.ml-quant.com/papers/ssrn/4695079/
- __[DeepTraderX: Disrupting trading strategies with deep learning](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692622)__: Disrupting trading strategies with deep learning: The paper presents DeepTraderX, a Deep Learning-based trader that learns from market prices, and demonstrates its successful performance in a multithreaded market simulation. (2023-10-20, shares: 2) · https://www.ml-quant.com/papers/ssrn/4692622/
- __[Flood Risk Pricing: Geo-Hierarchical Deep Learning](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692475)__: Geo-Hierarchical Deep Learning: A new deep learning framework enhances flood risk modeling, providing more accurate pricing and reducing capital requirements, as shown in a Mississippi River case study. (2022-01-13, shares: 2) · https://www.ml-quant.com/papers/ssrn/4692475/
- __[Effectiveness of Forex Intervention: Role of Domestic Fundamentals](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692676)__: Role of Domestic Fundamentals: Foreign exchange intervention can stabilize currencies in emerging markets under conditions like low volatility and high inflation, as per a study of 20 emerging economies. (2023-12-28, shares: 2) · https://www.ml-quant.com/papers/ssrn/4692676/
- __[Probabilistic Electricity Price Forecasting with Trading Applications](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4695159)__: A new method for electricity price forecasting using artificial neural networks is less costly and performs similarly to benchmarks, offering potential trading strategies for investors. (2023-08-02, shares: 2) · https://www.ml-quant.com/papers/ssrn/4695159/

### Financial

- __[Gamma Risk and Volatility Propagation in Trading](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692190)__: A study reveals that short-term options trading does not increase market volatility, but rather has an inverse relationship with intraday volatility. (2024-01-11, shares: 27) · https://www.ml-quant.com/papers/ssrn/4692190/
- __[Option Flows and Market Instability](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4695776)__: The speculative use of call options can cause price instability in the underlying asset's market, even with advanced volatility estimators, as per a study using the MinMaSS stability measure. (2024-01-15, shares: 8) · https://www.ml-quant.com/papers/ssrn/4695776/
- __[Extreme Liquidity in Asset Modeling](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4694674)__: A study using crypto assets indicates that jumps in asset prices are signs of extreme liquidity and can be effectively modeled using autoregressive models adjusted with liquidity. (2024-01-13, shares: 2) · https://www.ml-quant.com/papers/ssrn/4694674/
- __[Carbon Footprint Reduction in Index Portfolios](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692326)__: The article suggests that investors can lower the carbon footprint of their index-based portfolio by over 50% by focusing on low carbon emission stocks and limiting high-emission companies. (2024-01-12, shares: 2) · https://www.ml-quant.com/papers/ssrn/4692326/
- __[Extended Overlaps: Optimal Trading Hours in Europe](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4693290)__: Optimal Trading Hours in Europe: The article reveals that extended trading hours between North America and Europe during Daylight Saving Time enhances market liquidity and price efficiency, contributing to the discussion on optimal trading hours in Europe. (2024-01-12, shares: 3) · https://www.ml-quant.com/papers/ssrn/4693290/
- __[Boosting Fund Performance](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4697785)__: Mutual funds that match their investments with similar benchmark peers like the S&P 500 index yield higher returns and experience less volatility. (2023-03-11, shares: 2) · https://www.ml-quant.com/papers/ssrn/4697785/
- __[Financial Market Developments and Employee Welfare](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4690550)__: Equity options and credit default swaps trading benefits company employees by reducing short-term managerial focus and improving information efficiency. (2023-04-19, shares: 2) · https://www.ml-quant.com/papers/ssrn/4690550/
- __[Global Volatility and Capital Flows](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4691941)__: During high volatility periods, institutional investors globally reduce their equity allocations, while retail investors shift from small-cap to large-cap stocks. (2022-04-13, shares: 2) · https://www.ml-quant.com/papers/ssrn/4691941/
- __[Fear in Finance: FoMO Impact](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692691)__: FoMO Impact: The Fear of Missing Out (FoMO) effect in financial markets boosts equity and cryptocurrency prices and lowers market volatility due to reduced investor disagreement. (2021-12-08, shares: 2) · https://www.ml-quant.com/papers/ssrn/4692691/
- __[Yelp Sentiment & Asset Pricing](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4691145)__: A sentiment index based on Yelp restaurant reviews can predict stock market reversals and mispricing, with pessimism being a key predictive factor. (2022-08-04, shares: 2) · https://www.ml-quant.com/papers/ssrn/4691145/
- __[Deep Calibration for Stochastic Volatility](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4692741)__: A new method using neural networks to calibrate stochastic volatility models has proven to be robust and efficient, as confirmed by empirical and Monte Carlo experiments. (2023-06-25, shares: 2) · https://www.ml-quant.com/papers/ssrn/4692741/
- __[Climate-Optimized Investment Portfolios](https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4696468)__: The Climate Capital Efficiency Ratio (CER) ranks companies based on their carbon emission savings per dollar spent, offering a useful tool for climate-focused investing. (2023-11-28, shares: 2) · https://www.ml-quant.com/papers/ssrn/4696468/

## Machine learning

### Recently Published

- __[Easy Training Data](https://arxiv.org/abs/2401.06751)__: The study suggests that current language models can effectively generalize from easy to hard data, implying that scalable oversight may be less challenging than previously believed. (2024-01-12, shares: 94) · https://www.ml-quant.com/papers/arxiv/2401.06751/
- __[Transformers as RNNs](https://arxiv.org/abs/2401.06104)__: The research shows that decoder-only transformers can be seen as infinite multi-state RNNs and introduces a new policy, TOVA, which performs better in long-range tasks and uses less memory. (2024-01-11, shares: 152) · https://www.ml-quant.com/papers/arxiv/2401.06104/
- __[TOFU Unlearning for LLMs](https://arxiv.org/abs/2401.06121)__: The study presents TOFU, a new benchmark for understanding unlearning in large language models, revealing that existing unlearning algorithms are not effective. (2024-01-11, shares: 22) · https://www.ml-quant.com/papers/arxiv/2401.06121/
- __[PANDORA: Parallel Dendrogram Construction Algorithm: Parallel Dendrogram Construction Algorithm PANDORA](https://arxiv.org/abs/2401.06089)__: Parallel Dendrogram Construction Algorithm: Parallel Dendrogram Construction Algorithm PANDORA: Pandora, a new parallel algorithm, has been introduced for efficient dendrogram construction in hierarchical clustering, offering significant speed improvements on CPUs and GPUs. (2024-01-11, shares: 15) · https://www.ml-quant.com/papers/arxiv/2401.06089/

## GitHub

### Finance

- __[ViTST: Time Series as Images Transformer](https://github.com/Leezekun/ViTST)__: Time Series as Images Transformer: NeurIPS 2023 introduces a paper discussing the use of Vision Transformer for analyzing irregularly sampled time series data. (2023-02-23, shares: 51)
- __[Stockformer: Swing Trading with STL Decomposition and Self-Attention](https://github.com/Eric991005/Stockformer)__: Swing Trading with STL Decomposition and Self-Attention: A paper suggesting a swing trading strategy using STL Decomposition and Self-Attention Networks is being reviewed for publication in Neurocomputing. (2023-11-09, shares: 12)
- __[Piepilot: Portfolio Optimizer](https://github.com/ranaroussi/piepilot)__: Portfolio Optimizer: A straightforward tool designed for optimizing investment portfolios. (2024-01-13, shares: 4)

### Trending

- __[Fast AI Gateway](https://github.com/Portkey-AI/gateway)__: The article explores a high-speed AI Gateway capable of managing 100 LLMs via a single, user-friendly API. (2023-08-23, shares: 1370)
- __[ChatGPT Web UI](https://github.com/ollama-webui/ollama-webui)__: The piece presents a web user interface client for Ollama, modeled after ChatGPT. (2023-10-06, shares: 3316)
- __[Draggable Streamlit Dashboard](https://github.com/okld/streamlit-elements)__: The article provides a guide on building a customizable Streamlit dashboard with tools like Material UI widgets, Monaco editor, Visual Studio Code, Nivo charts, etc. (2021-04-15, shares: 451)

## Videos

### Quantitative

- __[Mastering Python Finance](https://www.youtube.com/watch?v=qkKAJwg4ohQ)__: The Certificate in Python for Finance Program provides in-depth knowledge on financial data science, asset management, algorithmic trading, and computational finance using Python and AI. (2024-01-11, shares: 5)
- __[AI and Finance Future](https://www.youtube.com/watch?v=T-AyBUMcWeg)__: The AFA Panel on AI discusses the influence of AI on the financial sector with panelists from various universities. (2024-01-13, shares: 11)
- __[RAW AI in Finance Workshop 3](https://www.youtube.com/watch?v=5w99hUczjyY)__: A screen recording from the Workshop on AI in Finance at Texas State University San Marcos is accessible on GitHub. (2024-01-11, shares: 0)
- __[AFA Business Meeting & Awards](https://www.youtube.com/watch?v=BYZntg0228I)__: The AFA Business Meeting and Presidential Address involves discussions on the future of finance from professionals and academics. (2024-01-14, shares: 9)

## X / Twitter

### Quantitative

- __[Phidata's Python Library for Autonomous AI](https://twitter.com/carlcarrie/status/1747057820841701592)__: Phidata has launched a new Python library for Autonomous AI that utilizes LLM function calling, with resources accessible on Streamlit FastAPI. (2024-01-16, shares: 0)
- __[IMF Report on AI's Impact on Employment](https://twitter.com/carlcarrie/status/1747041277370065352)__: The IMF has published a report analyzing the effects of Generative AI brands on job markets. (2024-01-15, shares: 0)
- __[BCGs Report: From Potential to Profit with GenAI](https://twitter.com/carlcarrie/status/1746916291619836145)__: From Potential to Profit with GenAI: According to a BCG report, companies investing in GenAI could expect over 10% cost savings, potentially amounting to 1 billion in savings. (2024-01-15, shares: 0)

### Miscellaneous

- __[The Rise of Diffusion Models](https://twitter.com/carlcarrie/status/1746643702896796040)__: The article explores the increasing use of diffusion models in timeseries forecasting, detailing 11 specific versions, their theoretical basis, effectiveness on various datasets, and comparisons between them. (2024-01-14, shares: 0)
- __[Marimo: Reinventing Jupyter](https://twitter.com/carlcarrie/status/1745999782659723373)__: Reinventing Jupyter: The article presents Marimo, a revamped version of the Jupyter Python notebook, designed to be a reproducible, interactive, and shareable Python program, as opposed to an error-prone JSON scratchpad. (2024-01-13, shares: 0)
- __[HBR on Fabrications and LLM Problems](https://twitter.com/carlcarrie/status/1745962927629185407)__: The article reviews the Harvard Business Review's perspective on plausible fabrications and other issues related to LLM in the context of productivity transformation led by LLM. (2024-01-13, shares: 0)

## Reddit

### Quantitative

- __[Choosing the Right Field: Quantitative Finance Journey](https://www.reddit.com/r/quant/comments/196oru9/how_did_you_know_that_the_quant_field_was_right/)__:  (2024-01-14, shares: 41)
- __[Trouble at Jump Trading](https://www.reddit.com/r/quant/comments/193vnrs/trouble_at_jump_trading/)__:  (2024-01-11, shares: 70)
- __[Ito's Lemma Query](https://www.reddit.com/r/quant/comments/196fqrp/question_regarding_itos_lemma/)__:  (2024-01-14, shares: 20)