arXiv subjects

Qizhao Chen

Publications and source records attributed to Qizhao Chen.

10 recordsLinked to original sources

arXiv2025

A Framework for Measuring How News Topics Drive Stock Movement

In modern financial markets, news plays a critical role in shaping investor sentiment and influencing stock price movements. However, most existing studies aggregate daily news sentiment into a single score, potentially overlooking important variations in topic content and relevance. This simplification may mask nuanced relationships between specific news themes and market responses. To address this gap, this paper proposes a novel framework to examine how different news topics influence stock price movements. The framework encodes individual news headlines into dense semantic embeddings using a pretrained sentence transformer, then applies K-means clustering to identify distinct news topics. Topic exposures are incorporated as explanatory variables in an ordinary least squares regression to quantify their impact on daily stock returns. Applied to Apple Inc., the framework reveals that certain topics are significantly associated with positive or negative next-day returns, while others have no measurable effect. These findings highlight the importance of topic-level analysis in understanding the relationship between news content and financial markets. The proposed framework provides a scalable approach for both researchers and practitioners to assess the informational value of different news topics and suggests a promising direction for improving predictive models of stock price movement.

cs.CE↗

arXiv2025

Adaptive Alpha Weighting with PPO: Enhancing Prompt-Based LLM-Generated Alphas in Quant Trading

This paper introduces a reinforcement learning framework that employs Proximal Policy Optimization (PPO) to dynamically optimize the weights of multiple large language model (LLM)-generated formulaic alphas for stock trading strategies. Formulaic alphas are mathematically defined trading signals derived from price, volume, sentiment, and other data. Although recent studies have shown that LLMs can generate diverse and effective alphas, a critical challenge lies in how to adaptively integrate them under varying market conditions. To address this gap, we leverage a DeepSeek model to generate fifty alphas for ten stocks, and then use PPO to adjust their weights in real time. Experimental results indicate that the PPO-optimized strategy does not consistently deliver the highest cumulative returns across all stocks, but it achieves comparatively higher Sharpe ratios and smaller maximum drawdowns in most cases. When compared with baseline strategies, including equal-weighted, buy-and-hold, random entry/exit, and momentum approaches, PPO demonstrates more stable risk-adjusted performance. The findings highlight the importance of reinforcement learning in the allocation of alpha weights and show the potential of combining LLM-generated signals with adaptive optimization for robust financial forecasting and trading.

cs.CE↗

arXiv2025

Sentiment-Aware Mean-Variance Portfolio Optimization for Cryptocurrencies

Cryptocurrency markets are highly volatile and influenced by both price trends and market sentiment, making effective portfolio management challenging. This paper proposes a dynamic cryptocurrency portfolio strategy that integrates technical indicators and sentiment analysis to enhance investment decision-making. Market momentum is captured using the 14-day Relative Strength Index (RSI) and Simple Moving Average (SMA), while sentiment signals are extracted from news articles with VADER and further validated using the Google Gemini large language model. These signals are incorporated into expected return estimates and used in a constrained mean-variance optimization framework. Backtesting across multiple cryptocurrencies shows that the integrated approach outperforms traditional benchmarks, including momentum strategy, Bitcoin Long-Short strategy, and an equal-weighted portfolio, achieving stronger risk-adjusted returns and more consistent cumulative growth. Furthermore, comparing the sentiment-only and technical-only strategies shows that incorporating sentiment information alongside technical indicators can lead to more consistent performance gains. However, the strategies exhibit substantial drawdowns that coincide with known periods of market stress, indicating that additional risk-management components are required to improve stability.

cs.CE↗

arXiv2025

Sentiment-Aware Stock Price Prediction with Transformer and LLM-Generated Formulaic Alpha

Traditionally, traders and quantitative analysts address alpha decay by manually crafting formulaic alphas, mathematical expressions that identify patterns or signals in financial data, through domain expertise and trial-and-error. This process is often time-consuming and difficult to scale. With recent advances in large language models (LLMs), it is now possible to automate the generation of such alphas by leveraging the reasoning capabilities of LLMs. This paper introduces a novel framework that integrates a prompt-based LLM with a Transformer model for stock price prediction. The LLM first generates diverse and adaptive alphas using structured inputs such as historical stock features (Close, Open, High, Low, Volume), technical indicators, sentiment scores of both target and related companies. These alphas, instead of being used directly for trading, are treated as high-level features that capture complex dependencies within the financial data. To evaluate the effectiveness of these LLM-generated formulaic alphas, the alpha features are then fed into prediction models such as Transformer, LSTM, TCN, SVR, and Random Forest to forecast future stock prices. Experimental results demonstrate that the LLM-generated alphas significantly improve predictive accuracy. Moreover, the accompanying natural language reasoning provided by the LLM enhances the interpretability and transparency of the predictions, supporting more informed financial decision-making.

cs.CE↗

arXiv2025

Inference on Optimal Policy Values and Other Irregular Functionals via Softmax Smoothing

Constructing confidence intervals for the value of an (unknown) optimal treatment policy is a fundamental problem in causal inference. Insight into the optimal policy value can guide the development of reward-maximizing, individualized treatment regimes. However, because the functional that defines the optimal value is non-differentiable, standard semi-parametric approaches for performing inference fail to be directly applicable. Many existing works circumvent non-differentiability by making the unrealistic assumption of zero probability of treatment non-response, i.e. that every unit responds (either positively or negatively) to an assigned treatment. Further, works that don't circumvent this restriction rely on refitting nuisance models a number of times proportional to the sample size. In this paper, we construct and analyze a simple, softmax smoothing-based estimator for the value of an optimal treatment policy. Our estimator applies in both static and dynamic treatment regimes, only requires fitting a constant number of nuisance models, and is statistically efficient when there is zero probability of non-response to treatment. Also, while our estimator does not require making semi-parametric restrictions, it can exploit them when they exist. We further show how our softmax smoothing approach can be used to estimate general parameters that are specified as a maximum of scores involving nuisance components, and look at conditional Balke and Pearl bounds and $L^1$ calibration error as salient examples.

econ.EM↗

arXiv2025

Comparing Different Transformer Model Structures for Stock Prediction

This paper compares different Transformer model architectures for stock index prediction. While many studies have shown that Transformers perform well in stock price forecasting, few have explored how different structural designs impact performance. Most existing works treat the Transformer as a black box, overlooking how specific architectural choices may affect predictive accuracy. However, understanding these differences is critical for developing more effective forecasting models. This study aims to identify which Transformer variant is most suitable for stock forecasting. This study evaluates five Transformer structures: (1) encoder-only Transformer, (2) decoder-only Transformer, (3) Vanilla Transformer (encoder + decoder), (4) Vanilla Transformer without embedding layers, and (5) Vanilla Transformer with ProbSparse attention. Results show that Transformer-based models generally outperform traditional approaches. Transformer with decoder only structure outperforms all other models in all scenarios. Transformer with ProbSparse attention has the worst performance in almost all cases.

cs.CE↗

arXiv2024

A Unified MPC Strategy for a Tilt-rotor VTOL UAV Towards Seamless Mode Transitioning

Capabilities of long-range flight and vertical take-off and landing (VTOL) are essential for Urban Air Mobility (UAM). Tiltrotor VTOLs have the advantage of balancing control simplicity and system complexity due to their redundant control authority. Prior work on controlling these aircraft either requires separate controllers and switching modes for different vehicle configurations or performs the control allocation on separate actuator sets, which cannot fully use the potential of the redundancy of tiltrotor. This paper introduces a unified MPC-based control strategy for a customized tiltrotor VTOL Unmanned Aerial Vehicle (UAV), which does not require mode-switching and can perform the control allocation in a consistent way. The incorporation of four independently controllable rotors in VTOL design offers an extra level of redundancy, allowing the VTOL to accommodate actuator failures. The result shows that our approach outperforms PID controllers while maintaining unified control. It allows the VTOL to perform smooth acceleration/deceleration, and precise coordinated turns. In addition, the independently controlled tilts enable the vehicle to handle actuator failures, ensuring that the aircraft remains operational even in the event of a servo or motor malfunction.

eess.SY↗

arXiv2023

Inference on Optimal Dynamic Policies via Softmax Approximation

Estimating optimal dynamic policies from offline data is a fundamental problem in dynamic decision making. In the context of causal inference, the problem is known as estimating the optimal dynamic treatment regime. Even though there exists a plethora of methods for estimation, constructing confidence intervals for the value of the optimal regime and structural parameters associated with it is inherently harder, as it involves non-linear and non-differentiable functionals of unknown quantities that need to be estimated. Prior work resorted to sub-sample approaches that can deteriorate the quality of the estimate. We show that a simple soft-max approximation to the optimal treatment regime, for an appropriately fast growing temperature parameter, can achieve valid inference on the truly optimal regime. We illustrate our result for a two-period optimal dynamic regime, though our approach should directly extend to the finite horizon case. Our work combines techniques from semi-parametric inference and $g$-estimation, together with an appropriate triangular array central limit theorem, as well as a novel analysis of the asymptotic influence and asymptotic bias of softmax approximations.

econ.EM↗

arXiv2022

Debiased Machine Learning without Sample-Splitting for Stable Estimators

Estimation and inference on causal parameters is typically reduced to a generalized method of moments problem, which involves auxiliary functions that correspond to solutions to a regression or classification problem. Recent line of work on debiased machine learning shows how one can use generic machine learning estimators for these auxiliary problems, while maintaining asymptotic normality and root-$n$ consistency of the target parameter of interest, while only requiring mean-squared-error guarantees from the auxiliary estimation algorithms. The literature typically requires that these auxiliary problems are fitted on a separate sample or in a cross-fitting manner. We show that when these auxiliary estimation algorithms satisfy natural leave-one-out stability properties, then sample splitting is not required. This allows for sample re-use, which can be beneficial in moderately sized sample regimes. For instance, we show that the stability properties that we propose are satisfied for ensemble bagged estimators, built via sub-sampling without replacement, a popular technique in machine learning practice.

econ.EM↗

arXiv2019

Parking On A Random Rooted Plane Tree

In this paper, we investigate a parking process on a uniform random rooted plane tree with $n$ vertices. Every vertex of the tree has a parking space for a single car. Cars arrive at independent uniformly random vertices of the tree. If the parking space at a vertex is unoccupied when a car arrives there, it parks. If not, the car drives towards the root and parks in the first empty space it encounters (if there is one). We are interested in asymptotics of the probability of the event that all cars can park when $\lfloor αn \rfloor$ cars arrive, for $α> 0$. We observe that there is a phase transition at $α_c := \sqrt{2} -1$: if $α< α_c$ then the event has positive probability, whereas for $α> α_c$ it has probability 0. Analogous results have been proved by Lackner and Panholzer, Goldschmidt and Przykucki, and Jones for different underlying random tree models.

math.PR↗