arXiv subjects

Kentaro Imajo

Publications and source records attributed to Kentaro Imajo.

17 recordsLinked to original sources

arXiv2026

Rethinking Synthetic Scenario Realism: Compatibility, Not Fidelity, Drives Hedging Performance

Deep hedging is a data-driven approach to learn hedging strategies. It relies on synthetic price paths generator, as real market data is often limited for training. Existing approaches primarily evaluate such generators based on realism, i.e., how well they capture statistical properties of real markets, but the relationship between realism and hedging performance remains unclear. In this work, we introduce a decision-centric perspective on synthetic data for deep hedging based on the notion of compatibility. Compatibility measures the extent to which strategies trained on synthetic scenarios remain effective in the true market. We theoretically show that 1) hedging performance decomposes into learning error and a compatibility gap, and 2) realism and compatibility can diverge. Empirically, we find that hedging performance is governed not by realism alone, but by the alignment between the generator and the hedger, together with task structure. Taken together, this work provides a principled basis for designing synthetic data in finance aligned with decision tasks.

q-fin.CP↗

arXiv2026

Uncovering Residual Factors in Financial Time Series via PCA and MTP2-constrained Gaussian Graphical Models

Financial time series are commonly decomposed into market factors, which capture shared price movements across assets, and residual factors, which reflect asset-specific deviations. To hedge the market-wide risks, such as the COVID-19 shock, trading strategies that exploit residual factors have been shown to be effective. However, financial time series often exhibit near-singular eigenstructures, which hinder the stable and accurate estimation of residual factors. This paper proposes a method for extracting residual factors from financial time series that hierarchically applies principal component analysis (PCA) and Gaussian graphical model (GGM). Our hierarchical approach balances stable estimation with elimination of factors that PCA alone cannot fully remove, enabling efficient extraction of residual factors. We use multivariate totally positive of order 2 (MTP2)-constrained GGM to capture the predominance of positive correlations in financial data. Our analysis proves that the resulting residual factors exhibit stronger orthogonality than those obtained with PCA alone. Across multiple experiments with varying test periods and training set lengths, the proposed method consistently achieved superior orthogonality of the residual factors. Backtests on the S&P 500 and TOPIX 500 constituents further indicate improved trading performance, including higher Sharpe ratios.

cs.CE↗

arXiv2025

PLaMo 2 Technical Report

In this report, we introduce PLaMo 2, a series of Japanese-focused large language models featuring a hybrid Samba-based architecture that transitions to full attention via continual pre-training to support 32K token contexts. Training leverages extensive synthetic corpora to overcome data scarcity, while computational efficiency is achieved through weight reuse and structured pruning. This efficient pruning methodology produces an 8B model that achieves performance comparable to our previous 100B model. Post-training further refines the models using a pipeline of supervised fine-tuning (SFT) and direct preference optimization (DPO), enhanced by synthetic Japanese instruction data and model merging techniques. Optimized for inference using vLLM and quantization with minimal accuracy loss, the PLaMo 2 models achieve state-of-the-art results on Japanese benchmarks, outperforming similarly-sized open models in instruction-following, language fluency, and Japanese-specific knowledge.

cs.CL↗

arXiv2025

A Judge-free LLM Open-ended Generation Benchmark Based on the Distributional Hypothesis

Evaluating the open-ended text generation of large language models (LLMs) is challenging because of the lack of a clear ground truth and the high cost of human or LLM-based assessments. We propose a novel benchmark that evaluates LLMs using n-gram statistics and rules, without relying on human judgement or LLM-as-a-judge approaches. Using 50 question and reference answer sets, we introduce three new metrics based on n-grams and rules: Fluency, Truthfulness, and Helpfulness. Our benchmark strongly correlates with GPT-4o-based evaluations while requiring significantly fewer computational resources, demonstrating its effectiveness as a scalable alternative for assessing LLMs' open-ended generation capabilities.

cs.CL↗

arXiv2024

Financial Fine-tuning a Large Time Series Model

Large models have shown unprecedented capabilities in natural language processing, image generation, and most recently, time series forecasting. This leads us to ask the question: treating market prices as a time series, can large models be used to predict the market? In this paper, we answer this by evaluating the performance of the latest time series foundation model TimesFM on price prediction. We find that due to the irregular nature of price data, directly applying TimesFM gives unsatisfactory results and propose to fine-tune TimeFM on financial data for the task of price prediction. This is done by continual pre-training of the latest time series foundation model TimesFM on price data containing 100 million time points, spanning a range of financial instruments spanning hourly and daily granularities. The fine-tuned model demonstrates higher price prediction accuracy than the baseline model. We conduct mock trading for our model in various financial markets and show that it outperforms various benchmarks in terms of returns, sharpe ratio, max drawdown and trading cost.

q-fin.CP↗

arXiv2024

Enhancing Financial Domain Adaptation of Language Models via Model Augmentation

The domain adaptation of language models, including large language models (LLMs), has become increasingly important as the use of such models continues to expand. This study demonstrates the effectiveness of Composition to Augment Language Models (CALM) in adapting to the financial domain. CALM is a model to extend the capabilities of existing models by introducing cross-attention between two LLMs with different functions. In our experiments, we developed a CALM to enhance the financial performance of an LLM with strong response capabilities by leveraging a financial-specialized LLM. Notably, the CALM was trained using a financial dataset different from the one used to train the financial-specialized LLM, confirming CALM's ability to adapt to various datasets. The models were evaluated through quantitative Japanese financial benchmarks and qualitative response comparisons, demonstrating that CALM enables superior responses with higher scores than the original models and baselines. Additionally, comparative experiments on connection points revealed that connecting the middle layers of the models is most effective in facilitating adaptation to the financial domain. These findings confirm that CALM is a practical approach for adapting LLMs to the financial domain.

cs.CL↗

arXiv2024

PLaMo-100B: A Ground-Up Language Model Designed for Japanese Proficiency

We introduce PLaMo-100B, a large-scale language model designed for Japanese proficiency. The model was trained from scratch using 2 trillion tokens, with architecture such as QK Normalization and Z-Loss to ensure training stability during the training process. Post-training techniques, including Supervised Fine-Tuning and Direct Preference Optimization, were applied to refine the model's performance. Benchmark evaluations suggest that PLaMo-100B performs well, particularly in Japanese-specific tasks, achieving results that are competitive with frontier models like GPT-4. The base model is available at https://huggingface.co/pfnet/plamo-100b.

cs.CL↗

arXiv2024

The Construction of Instruction-tuned LLMs for Finance without Instruction Data Using Continual Pretraining and Model Merging

This paper proposes a novel method for constructing instruction-tuned large language models (LLMs) for finance without instruction data. Traditionally, developing such domain-specific LLMs has been resource-intensive, requiring a large dataset and significant computational power for continual pretraining and instruction tuning. Our study proposes a simpler approach that combines domain-specific continual pretraining with model merging. Given that general-purpose pretrained LLMs and their instruction-tuned LLMs are often publicly available, they can be leveraged to obtain the necessary instruction task vector. By merging this with a domain-specific pretrained vector, we can effectively create instruction-tuned LLMs for finance without additional instruction data. Our process involves two steps: first, we perform continual pretraining on financial data; second, we merge the instruction-tuned vector with the domain-specific pretrained vector. Our experiments demonstrate the successful construction of instruction-tuned LLMs for finance. One major advantage of our method is that the instruction-tuned and domain-specific pretrained vectors are nearly independent. This independence makes our approach highly effective. The Japanese financial instruction-tuned LLMs we developed in this study are available at https://huggingface.co/pfnet/nekomata-14b-pfn-qfin-inst-merge.

cs.CL↗

arXiv2024

Construction of Domain-specified Japanese Large Language Model for Finance through Continual Pre-training

Large language models (LLMs) are now widely used in various fields, including finance. However, Japanese financial-specific LLMs have not been proposed yet. Hence, this study aims to construct a Japanese financial-specific LLM through continual pre-training. Before tuning, we constructed Japanese financial-focused datasets for continual pre-training. As a base model, we employed a Japanese LLM that achieved state-of-the-art performance on Japanese financial benchmarks among the 10-billion-class parameter models. After continual pre-training using the datasets and the base model, the tuned model performed better than the original model on the Japanese financial benchmarks. Moreover, the outputs comparison results reveal that the tuned model's outputs tend to be better than the original model's outputs in terms of the quality and length of the answers. These findings indicate that domain-specific continual pre-training is also effective for LLMs. The tuned model is publicly available on Hugging Face.

cs.CL↗

arXiv2023

Adversarial Deep Hedging: Learning to Hedge without Price Process Modeling

Deep hedging is a deep-learning-based framework for derivative hedging in incomplete markets. The advantage of deep hedging lies in its ability to handle various realistic market conditions, such as market frictions, which are challenging to address within the traditional mathematical finance framework. Since deep hedging relies on market simulation, the underlying asset price process model is crucial. However, existing literature on deep hedging often relies on traditional mathematical finance models, e.g., Brownian motion and stochastic volatility models, and discovering effective underlying asset models for deep hedging learning has been a challenge. In this study, we propose a new framework called adversarial deep hedging, inspired by adversarial learning. In this framework, a hedger and a generator, which respectively model the underlying asset process and the underlying asset process, are trained in an adversarial manner. The proposed method enables to learn a robust hedger without explicitly modeling the underlying asset process. Through numerical experiments, we demonstrate that our proposed method achieves competitive performance to models that assume explicit underlying asset processes across various real market data.

q-fin.CP↗

arXiv2023

Efficient Learning of Nested Deep Hedging using Multiple Options

Deep hedging is a framework for hedging derivatives in the presence of market frictions. In this study, we focus on the problem of hedging a given target option by using multiple options. To extend the deep hedging framework to this setting, the options used as hedging instruments also have to be priced during training. While one might use classical pricing model such as the Black-Scholes formula, ignoring frictions can offer arbitrage opportunities which are undesirable for deep hedging learning. The goal of this study is to develop a nested deep hedging method. That is, we develop a fully-deep approach of deep hedging in which the hedging instruments are also priced by deep neural networks that are aware of frictions. However, since the prices of hedging instruments have to be calculated under many different conditions, the entire learning process can be computationally intractable. To overcome this problem, we propose an efficient learning method for nested deep hedging. Our method consists of three techniques to circumvent computational intractability, each of which reduces redundant computations during training. We show through experiments that the Black-Scholes pricing of hedge instruments can admit significant arbitrage opportunities, which are not observed when the pricing is performed by deep hedging. We also demonstrate that our proposed method successfully reduces the hedging risks compared to a baseline method that does not use options as hedging instruments.

q-fin.CP↗

arXiv2022

Theoretically Motivated Data Augmentation and Regularization for Portfolio Construction

The task we consider is portfolio construction in a speculative market, a fundamental problem in modern finance. While various empirical works now exist to explore deep learning in finance, the theory side is almost non-existent. In this work, we focus on developing a theoretical framework for understanding the use of data augmentation for deep-learning-based approaches to quantitative finance. The proposed theory clarifies the role and necessity of data augmentation for finance; moreover, our theory implies that a simple algorithm of injecting a random noise of strength $\sqrt{|r_{t-1}|}$ to the observed return $r_{t}$ is better than not injecting any noise and a few other financially irrelevant data augmentation techniques.

cs.LG↗

arXiv2022

Uncertainty Aware Trader-Company Method: Interpretable Stock Price Prediction Capturing Uncertainty

Machine learning is an increasingly popular tool with some success in predicting stock prices. One promising method is the Trader-Company~(TC) method, which takes into account the dynamism of the stock market and has both high predictive power and interpretability. Machine learning-based stock prediction methods including the TC method have been concentrating on point prediction. However, point prediction in the absence of uncertainty estimates lacks credibility quantification and raises concerns about safety. The challenge in this paper is to make an investment strategy that combines high predictive power and the ability to quantify uncertainty. We propose a novel approach called Uncertainty Aware Trader-Company Method~(UTC) method. The core idea of this approach is to combine the strengths of both frameworks by merging the TC method with the probabilistic modeling, which provides probabilistic predictions and uncertainty estimations. We expect this to retain the predictive power and interpretability of the TC method while capturing the uncertainty. We theoretically prove that the proposed method estimates the posterior variance and does not introduce additional biases from the original TC method. We conduct a comprehensive evaluation of our approach based on the synthetic and real market datasets. We confirm with synthetic data that the UTC method can detect situations where the uncertainty increases and the prediction is difficult. We also confirmed that the UTC method can detect abrupt changes in data generating distributions. We demonstrate with real market data that the UTC method can achieve higher returns and lower risks than baselines.

q-fin.CP↗

arXiv2022

Power Laws and Symmetries in a Minimal Model of Financial Market Economy

A financial market is a system resulting from the complex interaction between participants in a closed economy. We propose a minimal microscopic model of the financial market economy based on the real economy's symmetry constraint and minimality requirement. We solve the proposed model analytically in the mean-field regime, which shows that various kinds of universal power-law-like behaviors in the financial market may depend on one another, just like the critical exponents in physics. We then discuss the parameters in the proposed model, and we show that each parameter in our model can be related to measurable quantities in the real market, which enables us to discuss the cause of a few kinds of social and economic phenomena.

physics.soc-ph↗

arXiv2021

No-Transaction Band Network: A Neural Network Architecture for Efficient Deep Hedging

Deep hedging (Buehler et al. 2019) is a versatile framework to compute the optimal hedging strategy of derivatives in incomplete markets. However, this optimal strategy is hard to train due to action dependence, that is, the appropriate hedging action at the next step depends on the current action. To overcome this issue, we leverage the idea of a no-transaction band strategy, which is an existing technique that gives optimal hedging strategies for European options and the exponential utility. We theoretically prove that this strategy is also optimal for a wider class of utilities and derivatives including exotics. Based on this result, we propose a no-transaction band network, a neural network architecture that facilitates fast training and precise evaluation of the optimal hedging strategy. We experimentally demonstrate that for European and lookback options, our architecture quickly attains a better hedging strategy in comparison to a standard feed-forward network.

q-fin.CP↗

arXiv2020

Trader-Company Method: A Metaheuristic for Interpretable Stock Price Prediction

Investors try to predict returns of financial assets to make successful investment. Many quantitative analysts have used machine learning-based methods to find unknown profitable market rules from large amounts of market data. However, there are several challenges in financial markets hindering practical applications of machine learning-based models. First, in financial markets, there is no single model that can consistently make accurate prediction because traders in markets quickly adapt to newly available information. Instead, there are a number of ephemeral and partially correct models called "alpha factors". Second, since financial markets are highly uncertain, ensuring interpretability of prediction models is quite important to make reliable trading strategies. To overcome these challenges, we propose the Trader-Company method, a novel evolutionary model that mimics the roles of a financial institute and traders belonging to it. Our method predicts future stock returns by aggregating suggestions from multiple weak learners called Traders. A Trader holds a collection of simple mathematical formulae, each of which represents a candidate of an alpha factor and would be interpretable for real-world investors. The aggregation algorithm, called a Company, maintains multiple Traders. By randomly generating new Traders and retraining them, Companies can efficiently find financially meaningful formulae whilst avoiding overfitting to a transient state of the market. We show the effectiveness of our method by conducting experiments on real market data.

q-fin.TR↗

arXiv2020

Deep Portfolio Optimization via Distributional Prediction of Residual Factors

Recent developments in deep learning techniques have motivated intensive research in machine learning-aided stock trading strategies. However, since the financial market has a highly non-stationary nature hindering the application of typical data-hungry machine learning methods, leveraging financial inductive biases is important to ensure better sample efficiency and robustness. In this study, we propose a novel method of constructing a portfolio based on predicting the distribution of a financial quantity called residual factors, which is known to be generally useful for hedging the risk exposure to common market factors. The key technical ingredients are twofold. First, we introduce a computationally efficient extraction method for the residual information, which can be easily combined with various prediction algorithms. Second, we propose a novel neural network architecture that allows us to incorporate widely acknowledged financial inductive biases such as amplitude invariance and time-scale invariance. We demonstrate the efficacy of our method on U.S. and Japanese stock market data. Through ablation experiments, we also verify that each individual technique contributes to improving the performance of trading strategies. We anticipate our techniques may have wide applications in various financial problems.

q-fin.PM↗