app.py

import streamlit as st
import pandas as pd
import yfinance as yf
from niftystocks import ns
from openai import AzureOpenAI
import os
# Create a Streamlit app
st.title("NIFTY Finance Analysis")

st.markdown(
    """
    <style>
    @import url('https://fonts.googleapis.com/css2?family=Quicksand:[email protected]&display=swap');
    body {
        font-family: 'Quicksand', sans-serif;
    }
    .st-emotion-cache-13ln4jf.ea3mdgi5 {
        position: -webkit-sticky;
        position: sticky;
        top: 10;
        z-index: 100;
        padding-top: 42px;
        padding-bottom: 10px;
        border-bottom: 1px solid #e6e6e6;
    }
    header.st-emotion-cache-12fmjuu.ezrtsby2 {
      background-color: #f0f2f6 !important;
    }
    .st-emotion-cache-12fmjuu.ezrtsby2{
        background:url("https://www.pngplay.com/wp-content/uploads/5/Lloyds-Banking-Group-Logo-Transparent-PNG.png") no-repeat;
        background-size: 250px 50px;
        background-position: center;
        padding-left: 50px;
        padding-top:10px;
        padding-bottom:10px;
      }
    .st-emotion-cache-1vt4y43.ef3psqc13{
    }
    </style>
    """,
    unsafe_allow_html=True,
)

# Function to fetch financial data
def get_financial_data(ticker):
    company = yf.Ticker(ticker)
    financial_data = {}

    # Price Ratios
    financial_data['price_to_earnings'] = company.info.get('trailingPE')
    financial_data['price_to_book'] = company.info.get('priceToBook')

    # Debt
    financial_data['debt_to_equity'] = company.info.get('debtToEquity')

    # Financial Statements
    financial_data['balance_sheet'] = company.balance_sheet
    financial_data['income_statement'] = company.financials
    financial_data['cashflow_statement'] = company.cashflow

    # Management Quality
    financial_data['management'] = company.info.get('management')

    # Financial Ratios
    financial_data['current_ratio'] = company.info.get('currentRatio')
    financial_data['dividend_yield'] = company.info.get('dividendYield')
    financial_data['return_on_equity'] = company.info.get('returnOnEquity')

    # Other metrics
    financial_data['market_cap'] = company.info.get('marketCap')
    financial_data['earnings_growth'] = company.info.get('earningsGrowth')
    financial_data['return_on_assets'] = company.info.get('returnOnAssets')
    financial_data['enterprise_value'] = company.info.get('enterpriseValue')
    financial_data['recommendation_mean'] = company.info.get('recommendationMean')
    financial_data['beta'] = company.info.get('beta')
    financial_data['book_value'] = company.info.get('bookValue')
    financial_data['free_cashflow'] = company.info.get('freeCashflow')
    financial_data['revenue_growth'] = company.info.get('revenueGrowth')
    financial_data['trailing_eps'] = company.info.get('trailingEps')
    financial_data['previous_close'] = company.info.get('previousClose')
    financial_data['average_volume'] = company.info.get('averageVolume')
    financial_data['dividend_yield'] = company.info.get('dividendYield')
    financial_data['trailing_pe'] = company.info.get('trailingPE')
    financial_data['business_model'] = company.info.get('longBusinessSummary')

    return financial_data

# Summarize and analyze the data
def summarize_company(ticker, data):
    summary = {
        "Company": ticker,
        "Market Cap": data['market_cap'],
        "Price to Earnings": data['price_to_earnings'],
        "Price to Book": data['price_to_book'],
        "Debt to Equity": data['debt_to_equity'],
        "Current Ratio": data['current_ratio'],
        "Dividend Yield": data['dividend_yield'],
        "Return on Equity": data['return_on_equity'],
        "Earnings Growth": data['earnings_growth'],
        "Return on Assets": data['return_on_assets'],
        "Enterprise Value": data['enterprise_value'],
        "Recommendation Mean": data['recommendation_mean'],
        "Beta": data['beta'],
        "Book Value": data['book_value'],
        "Free Cash Flow": data['free_cashflow'],
        "Revenue Growth": data['revenue_growth'],
        "Trailing EPS": data['trailing_eps'],
        "Previous Close": data['previous_close'],
        "Average Volume": data['average_volume'],
        "Trailing PE": data['trailing_pe'],
        "Business Model": data['business_model']
    }
    return summary

def summarize_text_technical(details_of_companies):
    client = AzureOpenAI()
    conversation = [
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": f"based on your financial knowledge to analyse the given finance data to give is the company is (good or better or bad) to investing choose the one based on given data only.give the reason why its good or better or bad.if based on data the company not good for investing boldly say is not good for investing.your decision is based on provided data only do not ask additional data to make a judgement.Output Format: Analysis: Reason:```{details_of_companies}```."}
    ]

    # Call OpenAI GPT-3.5-turbo
    chat_completion = client.chat.completions.create(
        model = "GPT-3",
        messages = conversation,
        max_tokens=600,
        temperature=0
    )

    response = chat_completion.choices[0].message.content
    return response

def summarize_text_fundamental(details_of_companies):
    client = AzureOpenAI()
    conversation = [
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": f"i want detailed Summary from given finance details.give a summary for each companies detailed and include all the information of a company. finally you analyse the company data to give it's [good or better or bad] for investment. content in backticks.Give Only Summary.Output Format:Summary:summary Analysis: good or better or bad for investment.```{details_of_companies}```."}
    ]

    # Call OpenAI GPT-3.5-turbo
    chat_completion = client.chat.completions.create(
        model = "GPT-3",
        messages = conversation,
        max_tokens=600,
        temperature=0
    )

    response = chat_completion.choices[0].message.content
    return response

# Function to fetch historical data
def fetch_stock_data(company, start_date, end_date):
    stock = yf.Ticker(company)
    df = stock.history(start=start_date, end=end_date)
    return df

# Function to analyze price and volume changes
def analyze_data(df):
    df['Open_to_High'] = df['High'] - df['Open']
    df['Open_to_Low'] = df['Open'] - df['Low']
    df['Open_to_Close'] = df['Close'] - df['Open']
    return df

# Function to add technical indicators
def add_technical_indicators(df):
    # Simple Moving Averages
    df['SMA_20'] = df['Close'].rolling(window=20).mean()
    df['SMA_50'] = df['Close'].rolling(window=50).mean()

    # Exponential Moving Averages
    df['EMA_20'] = df['Close'].ewm(span=20, adjust=False).mean()
    df['EMA_50'] = df['Close'].ewm(span=50, adjust=False).mean()

    # Relative Strength Index (RSI)
    delta = df['Close'].diff(1)
    gain = delta.where(delta > 0, 0)
    loss = -delta.where(delta < 0, 0)
    avg_gain = gain.rolling(window=14).mean()
    avg_loss = loss.rolling(window=14).mean()
    rs = avg_gain / avg_loss
    df['RSI'] = 100 - (100 / (1 + rs))

    # Moving Average Convergence Divergence (MACD)
    df['MACD'] = df['Close'].ewm(span=12, adjust=False).mean() - df['Close'].ewm(span=26, adjust=False).mean()
    df['MACD_Signal'] = df['MACD'].ewm(span=9, adjust=False).mean()

    return df

# Analyze and store data for all companies
def analyze_nifty_50(companies, start_date, end_date):
    result = {}
    for company in companies:
        df = fetch_stock_data(company+".NS", start_date, end_date)
        if df.empty:
            continue
        df_analyzed = analyze_data(df)
        df_analyzed = add_technical_indicators(df_analyzed)

        result[company] = {
            'open_to_high_avg': df_analyzed['Open_to_High'].mean(),
            'open_to_low_avg': df_analyzed['Open_to_Low'].mean(),
            'open_to_close_avg': df_analyzed['Open_to_Close'].mean(),
            'average_volume': df_analyzed['Volume'].mean(),
            'historical_average_volume': df['Volume'].mean(),
            'SMA_20': df_analyzed['SMA_20'].iloc[-1],
            'SMA_50': df_analyzed['SMA_50'].iloc[-1],
            'EMA_20': df_analyzed['EMA_20'].iloc[-1],
            'EMA_50': df_analyzed['EMA_50'].iloc[-1],
            'RSI': df_analyzed['RSI'].iloc[-1],
            'MACD': df_analyzed['MACD'].iloc[-1],
            'MACD_Signal': df_analyzed['MACD_Signal'].iloc[-1]
        }
    return result


tabs = ["Fundamental Analysis", "Technical Analysis"]
tab = st.tabs(tabs)

with tab[0]:
    # Fundamental Analysis
    st.header("Fundamental Analysis")
    st.info("If You Enter List of Companies Manually Not Mention '.NS' after Company Name")
    fundamental_text_input = st.text_input("Enter List of Company Names",placeholder="Enter Comma Seperated List of Companies")
    st.write("or")
    st.write("Get Automatically Nifty 50 Companies, Click 'Run Fundamental Analysis'")
    fundamental_button = st.button("Run Fundamental Analysis")
    if fundamental_button:
        # Fetch data for all Nifty 50 companies
        if fundamental_text_input:
            nifty_50_tickers = fundamental_text_input.split(",")
        else:
            nifty_50_tickers = ns.get_nifty50()
        nifty_50_data = {ticker: get_financial_data(ticker+'.NS') for ticker in nifty_50_tickers}
        details_of_companies = {}
        for ticker, data in nifty_50_data.items():
            details_of_companies[ticker] = summarize_company(ticker, data)

        fundamental_df = pd.DataFrame(details_of_companies)
        fundamental_df.drop('Company', inplace=True)
        fundamental_df = fundamental_df.transpose()
        fundamental_df.index.name = 'Company'
        all_company_summary = ""
        for ticker, data in details_of_companies.items():
            all_company_summary += f"\n\n----{ticker}----"+"\n\n"+summarize_text_fundamental(data)
            print(data)
            fundamental_df.loc[ticker, 'summary'] =  "\n\nCompany:"+ticker+"\n\n"+summarize_text_fundamental(data)
            print(ticker)
        summary_for_fundamental = ""
        for i in range(len(all_company_summary.split("\n\n"))):
          summary_for_fundamental += all_company_summary.split("\n\n")[i]+"  \n\n"
        # Display the results
        st.write(fundamental_df)
        st.download_button("Download CSV", fundamental_df.to_csv(), "nifty_50_fundamental_analysis.csv")
        st.write(summary_for_fundamental)

with tab[1]:
    # Technical Analysis
    st.header("Technical Analysis")
    st.info("If You Enter List of Companies Manually Not Mention '.NS' after Company Name")
    st.info("If the Result 'difficult to make decision' it is may not good for invest.")
    start_date = st.date_input("Start Date")
    end_date = st.date_input("End Date")
    technical_text_input = st.text_input("Enter List of Company Names ---",placeholder="Enter Comma Seperated List of Companies")
    st.write("or")
    st.write("Get Automatically Nifty 50 Companies, Click 'Run Technical Analysis'")
    technical_button = st.button("Run Technical Analysis")
    if technical_button:
        # Analyze the data
        if technical_text_input:
            nifty_50_companies = technical_text_input.split(",")
        else:
            nifty_50_companies = ns.get_nifty50()
        nifty_50_analysis = analyze_nifty_50(nifty_50_companies, start_date, end_date)
        technical_df = pd.DataFrame(nifty_50_analysis)
        technical_df = technical_df.transpose()
        technical_df.index.name = 'Company'
        all_company_summary = ""
        for ticker, data in nifty_50_analysis.items():
            print(data)
            all_company_summary += f"\n\n----{ticker}----"+"\n\n"+summarize_text_technical(data)
            technical_df.loc[ticker, 'summary'] =  "\n\nCompany:"+ticker+"\n\n"+summarize_text_technical(data)

        technical_df.to_csv("nifty_50_technical_analysis.csv")
        summary_for_technical = ""

        for i in range(len(all_company_summary.split("\n\n"))):
          summary_for_technical +=all_company_summary.split("\n\n")[i]+"  \n\n"

        # Display the results
        st.write(technical_df)
        st.download_button("Download CSV", technical_df.to_csv(), "nifty_50_technical_analysis.csv")

        st.write(summary_for_technical)