Junyu Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Junyu Chen has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Zhenrui Chen
Columbia University in the City of New York, USA
3 shared publications - Kuan Lu
Cornell University, USA
2 shared publications - Menghao Huo
Santa Clara University, USA
2 shared publications - Huiyan Xing
Columbia University in the City of New York, USA
1 shared publication - Huiyan Xing
Fu Foundation School of Engineering and Applied Science, Columbia University in the City of New York, New York, New York, USA
1 shared publication - Miaobin Su
Columbia University in the City of New York, USA
1 shared publication
Latest Publications
- Open Access
- 16 July 2025
- Published by MDPI AG in Preprints.org
Multi-Model Approach for Stock Price Prediction and Trading Recommendations
The financial market has long been an important domain for the application of machine learning (ML). Since the 1980s, researchers have leveraged ML techniques to uncover patterns and trends in financial data. Yet, stock market prediction remains inherently challenging due to its volatility and non-linearity. This paper presents a comprehensive multi-model forecasting framework integrating classical statistical models and modern deep learning methods. We employ three baseline models—Linear Regression (LR), Autoregressive Integrated Moving Average (ARIMA), and Long Short-Term Memory (LSTM)—as well as two extended architectures: an ARIMA-LSTM ensemble and a Transformer-based model. Our experiments span five representative U.S. stocks (AAPL, KO, NVDA, PFE, TSLA), reflecting diverse volatility profiles. A unified preprocessing pipeline—including sliding window segmentation, normalization, and volatility-aware sampling—is applied across all models to ensure fair evaluation. Results show that while ARIMA performs best on volatile assets, LSTM benefits significantly from input normalization, and ensemble approaches offer enhanced robustness. This work contributes an extensible, model-agnostic framework for financial time series prediction, providing actionable insights for both academic research and practical trading applications.
- 16 July 2025
- Published by MDPI AG in Preprints.org
Open AccessMulti-Model Approach for Stock Price Prediction and Trading Recommendations
The financial market has long been an important domain for the application of machine learning (ML). Since the 1980s, researchers have leveraged ML techniques to uncover patterns and trends in financial data. Yet, stock market prediction remains inherently challenging due to its volatility and non-linearity. This paper presents a comprehensive multi-model forecasting framework integrating classical statistical models and modern deep learning methods. We employ three baseline models—Linear Regression (LR), Autoregressive Integrated Moving Average (ARIMA), and Long Short-Term Memory (LSTM)—as well as two extended architectures: an ARIMA-LSTM ensemble and a Transformer-based model. Our experiments span five representative U.S. stocks (AAPL, KO, NVDA, PFE, TSLA), reflecting diverse volatility profiles. A unified preprocessing pipeline—including sliding window segmentation, normalization, and volatility-aware sampling—is applied across all models to ensure fair evaluation. Results show that while ARIMA performs best on volatile assets, LSTM benefits significantly from input normalization, and ensemble approaches offer enhanced robustness. This work contributes an extensible, model-agnostic framework for financial time series prediction, providing actionable insights for both academic research and practical trading applications.
- Open Access
- 13 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 701-709
Multi-Model Approach for Stock Price Prediction and Trading Recommendations
- 13 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 701-709
Open AccessMulti-Model Approach for Stock Price Prediction and Trading Recommendations
- Open Access
- 20 January 2025
- Published by MDPI AG in Preprints.org
Real-time New York Traffic Heatmap Analysis and Visualization
We developed a web application demo, dynamically monitoring and visualizing the traffic condition in New York City, with a focus on Manhattan. We took advantage of two public API providers: Tomtom and Openweather, for traffic-related streaming data including real-time intersection speed, historical density, and weather details. We proposed a novel congestion model and updated the traffic heatmap every 15 minutes. We implemented our visualization by Google Maps APIs and similar to Google Map layout, we quantified the congestion into three degrees and plotted streets with colors. Our demo achieved similar performance with Google Maps, but with more reasonable and accurate results. Our work can also accept streaming data to achieve high-level concurrency.
- 20 January 2025
- Published by MDPI AG in Preprints.org
Open AccessReal-time New York Traffic Heatmap Analysis and Visualization
We developed a web application demo, dynamically monitoring and visualizing the traffic condition in New York City, with a focus on Manhattan. We took advantage of two public API providers: Tomtom and Openweather, for traffic-related streaming data including real-time intersection speed, historical density, and weather details. We proposed a novel congestion model and updated the traffic heatmap every 15 minutes. We implemented our visualization by Google Maps APIs and similar to Google Map layout, we quantified the congestion into three degrees and plotted streets with colors. Our demo achieved similar performance with Google Maps, but with more reasonable and accurate results. Our work can also accept streaming data to achieve high-level concurrency.