PARMIDA-V.0

Professional stock market analyzer using LLM

how it works :

1-Getting fundamental analysis data from different APIs

2-Getting indicators (RSI, etc.), patterns (Fibonacci, etc.) and technical analysis data from different APIs

3-Getting the news and analyzing the sentiments of the news to identify the market in the future

4-We give all this information to LLM Chatgpt for analysis

---

Fundamental analysis factors:

Company financial data:

Revenue

Net Income

Earnings before interest and taxes (EBIT)

Earnings before interest, taxes, depreciation and amortization (EBITDA)

Profit Margin

dividend

Dividend Payout Ratio

Earnings Per Share - EPS

Price-to-earnings ratio (P/E Ratio)

P/B Ratio

return on equity (ROE)

return on assets (ROA)

Free Cash Flow

Debt to Equity Ratio (D/E)

Current Ratio

Quick Ratio

balance sheet

Income Statement

Cash Flow Statement

Economic data:

Interest Rate

inflation rate

Exchange Rate

GDP Growth Rate

Unemployment Rate

Consumer Price Index (CPI)

Producer Price Index (PPI)

Trade Balance

Changes in monetary and fiscal policies

how to run Fundamental-Analysis:

first download library :

pip install requirements.txt

enter the api key of .env file :

QUANDL_API =  "Your QUANDL_API"
OPENAI_API_KEY = "Your OPENAI_API_KEY"
FRED_API = "Your FRED_API"
ALPHA_VANTAGE_API = "Your ALPHA_VANTAGE_API" 

go to directory : Fundamental-Analysis and enter this command

streamlit run Main-Fundamental-Analysis.py.py

---

Technical analysis factors:

Trend indicators:

Simple Moving Average (SMA): Identifying long-term trends.

Exponential Moving Average (EMA): More sensitive to recent price changes.

MACD (Moving Average Convergence Divergence): Identify trend changes and strengths.

ADX (Average Directional Index): measures the strength of the trend.

Volatility indicators: RSI (Relative Strength Index): Identifying the oversold situation.

Stochastic Oscillator: identifying buy and sell points.

CCI (Commodity Channel Index): identifying the saturation points of buying and selling.

ATR (Average True Range): A measure of price volatility.

Bands and tapes:

Bollinger Bands: identify swings and potential reversal points.

Andrews Pitchfork: identification of support and resistance levels.

Graphic patterns::

Candlestick patterns: such as doji pattern, evening star, etc.

Price patterns: such as the head and shoulders pattern, the pattern of two peaks and two valleys.

Triangle patterns: ascending, descending and symmetrical triangles.

Volume indicators:

OBV (On-Balance Volume): Evaluation of volume flow.

Volume Price Trend (VPT): Combination of volume and price to identify trends.

Chaikin Money Flow (CMF): Assessing buying and selling pressure.

Support and resistance levels:

Horizontal support and resistance levels: using price points that have acted as support or resistance in the past.

Trend Lines: Lines drawn on a price chart that show support and resistance levels.

Fibonacci: Using Fibonacci levels to predict possible support and resistance points.

Other indicators: Parabolic SAR: detection of stop and reversal points.

Ichimoku Cloud: identifying support and resistance levels and determining trend direction.

Williams %R: Identification of oversold conditions.

how to run Technical-Analysis:

first download library :

pip install requirements.txt

enter the api key of .env file :

QUANDL_API =  "Your QUANDL_API"
OPENAI_API_KEY = "Your OPENAI_API_KEY"
FRED_API = "Your FRED_API"
ALPHA_VANTAGE_API = "Your ALPHA_VANTAGE_API" 

go to directory : Technical-Analysis and enter this command

streamlit run Main-Technical-Analysis.py

---

interface of project : Technical photo: how to save data :

---

#how to run :

in the .env file enter your api-key

QUANDL_API =  "Your QUANDL_API"
OPENAI_API_KEY = "Your OPENAI_API_KEY"
FRED_API = "Your FRED_API"
ALPHA_VANTAGE_API = "Your ALPHA_VANTAGE_API" 

---

**how to run MVP-TEST **:

first download library :

pip install requirements.txt

streamlit run MVP-TEST.py

**suggestion: **

If you want to use this source code for stock market analysis seriously, not for educational purposes:

I suggest you do this:

1-Change the prompt of Openai :

        def analyze_data_in_chunks(data, chunk_size=1000):
            try:
                instructions = f"Please analyze the following financial data for {stock_name}. Provide detailed analysis, identify key trends, patterns, and factors affecting the stock. Provide a forecast for future stock price and profitability:\n"
                responses = []
                for i in range(0, len(data), chunk_size):
                    chunk = data[i:i + chunk_size]
                    chunk_instructions = instructions + f"Data chunk {i // chunk_size + 1}:\n{chunk}\n\n"
                    completion = openai.chat.completions.create(
                        model="gpt-3.5-turbo",
                        messages=[
                            {"role": "system", "content": chunk_instructions},
                            {"role": "user", "content": "Provide a concise and precise analysis and forecast."}
                        ]
                    )
                    response = completion.choices[0].message.content
                    responses.append(response)
                return responses
            except Exception as e:
                st.error(f"Failed to analyze data with OpenAI: {e}")

2- set the time for get data.

This project is for educational purposes only and is not guaranteed to analyze stock markets

The address of the llm analysis sample folder: doucment\result of AAPL.txt

Developer E-mail : [email protected]

Thanks for you support:)))))