econbot

Financial Sentiment Analysis Model

This model analyzes financial news and predicts sentiment (Bullish, Bearish, or Neutral) for specific stock tickers.

Model Description

This model is fine-tuned on financial news and market data to provide sentiment analysis for stock tickers. It analyzes news headlines and descriptions to determine whether the sentiment is bullish (positive), bearish (negative), or neutral for a given stock.

Use Cases

• Financial market analysis
• Investment decision support
• Stock market sentiment tracking
• Automated financial news filtering

Usage

This model can be used through the Hugging Face API or integrated into a web application.

Example Input

You are a financial analyst in a leading hedge fund. 
Analyze the sentiment of the following financial news for the given stock ticker step by step.
Title: "Apple Inc. (AAPL) Faces Sluggish Growth as iPhone 16 Demand Disappoints, Analysts Warn"
Summary: "Overall, AAPL ranks 4th on our list of AI stocks that are on Wall Street's radar today. While we acknowledge the potential of AAPL as an investment, our conviction lies in the belief that some AI stocks hold greater promise for delivering higher returns and doing so within a shorter time frame."
Stock Ticker: AAPL

Example Output

{
  "ticker": "AAPL",
  "sentiment": "Bearish",
  "sentiment_reasoning": "The news suggests a negative outlook for Apple stock due to sluggish growth, disappointing iPhone 16 demand, and analysts' warnings. The summary also implies that other AI stocks may offer better investment opportunities than AAPL."
}

Training

This model was fine-tuned on a dataset of financial news and corresponding market movements. It uses a PEFT (Parameter-Efficient Fine-Tuning) approach with LoRA adapters to optimize performance while maintaining efficiency.

• Base Model: GPT-2
• Fine-tuning Method: LoRA
• Training Dataset: Financial news articles with labeled sentiment

Limitations

• The model analyzes text only and does not incorporate quantitative financial data
• Predictions are based on textual content and may not capture all market factors
• The model should be used as one of many tools for financial analysis, not as the sole decision-maker

Citation

If you use this model in your research or application, please cite:

@misc{econbot2025,
  author = {Jiale Charlotte, Zhilin Zhu},
  title = {Financial Sentiment Analysis Model},
  year = {2025},
  publisher = {HuggingFace},
  howpublished = {\url{https://huggingface.co/jialeCharlotte/econbot}}
}

License

This model is licensed under Apache 2.0.