app.py

import streamlit as st
import backtrader as bt
from datetime import datetime
from strategy import CustomStrategy
from data_fetch import fetch_data
from plot_utils import plot_data


# Streamlit App
st.set_page_config(page_title="Backtesting with Backtrader", layout="wide")

# Dictionary of stocks and indices
stocks_dict = {
    "Reliance Industries": "RELIANCE.NS",
    "Tata Consultancy Services": "TCS.NS",
    "HDFC Bank": "HDFCBANK.NS",
    "Infosys": "INFY.NS",
    "Hindustan Unilever": "HINDUNILVR.NS",
    "ICICI Bank": "ICICIBANK.NS",
    "Bharti Airtel": "BHARTIARTL.NS",
    "State Bank of India": "SBIN.NS",
    "Kotak Mahindra Bank": "KOTAKBANK.NS",
    "Larsen & Toubro": "LT.NS",
    "Nifty 50": "^NSEI",
    "Bank Nifty": "^NSEBANK",
    "Sensex": "^BSESN"
}

st.title('📈 Backtesting with Backtrader')

# User selection
selected_company = st.sidebar.selectbox("Select a company or index:", list(stocks_dict.keys()))
ticker = stocks_dict[selected_company]

# Input for date range
start_date = st.sidebar.date_input("Start Date", value=datetime(2020, 1, 1))
end_date = st.sidebar.date_input("End Date", value=datetime.today())
if start_date > end_date:
    st.error("Start date cannot be after end date.")

# Strategy Parameters
st.sidebar.header("Select Strategy Parameters")
short_window = st.sidebar.slider("Short Moving Average Window", 1, 50, 10)
long_window = st.sidebar.slider("Long Moving Average Window", 1, 50, 30)
rsi_period = st.sidebar.slider("RSI Period", 1, 50, 14)
rsi_overbought = st.sidebar.slider("RSI Overbought Level", 50, 100, 70)
rsi_oversold = st.sidebar.slider("RSI Oversold Level", 0, 50, 30)
stop_loss_percent = st.sidebar.slider("Stop-Loss Percentage", 0.0, 10.0, 1.0)

# MACD parameters
st.sidebar.header("MACD Parameters")
macd_short = st.sidebar.slider("MACD Short Period", 1, 50, 12)
macd_long = st.sidebar.slider("MACD Long Period", 1, 50, 26)
macd_signal = st.sidebar.slider("MACD Signal Period", 1, 50, 9)

# Indicator selection
st.sidebar.header("Select Indicators")
use_sma = st.sidebar.checkbox("Use Short Moving Average", key='use_sma', value=True)
use_lma = st.sidebar.checkbox("Use Long Moving Average", key='use_lma', value=True)
use_rsi = st.sidebar.checkbox("Use RSI", key='use_rsi', value=True)
use_ema = st.sidebar.checkbox("Use Exponential Moving Average", key='use_ema', value=True)
use_macd = st.sidebar.checkbox("Use MACD", key='use_macd', value=True)
use_stop_loss = st.sidebar.checkbox("Enable Stop-Loss", key='use_stop_loss', value=True)


# Check if at least one indicator is selected
if not (use_sma or use_lma or use_rsi or use_ema or use_macd):
    st.sidebar.error("Please select at least one indicator to proceed.")
else:

    if st.button("Run Backtest"):
        # Download data using yfinance
        data = fetch_data(ticker, start=start_date, end=end_date)

        # Check if data was fetched successfully
        if data.empty:
            st.error("No data fetched. Please check the ticker and date range.")
        else:
            # Plot stock data
            plot_data(data, selected_company, 'Close')

            # Prepare data for Backtrader
            data_bt = bt.feeds.PandasData(dataname=data)

            # Initialize Cerebro
            cerebro = bt.Cerebro()
            cerebro.adddata(data_bt)

            # Add strategy with user-defined parameters
            cerebro.addstrategy(CustomStrategy, 
                                short_window=short_window, 
                                long_window=long_window, 
                                rsi_period=rsi_period, 
                                rsi_overbought=rsi_overbought, 
                                rsi_oversold=rsi_oversold,
                                use_stop_loss=st.session_state['use_stop_loss'], 
                                stop_loss_percent=stop_loss_percent, 
                                macd_short=macd_short, 
                                macd_long=macd_long, 
                                macd_signal=macd_signal)
            
            initial_value = 1000000.0  # Initial balance
            cerebro.broker.setcash(initial_value)

            # Run the backtest
            results = cerebro.run()

            # Access the logs from the last strategy instance
            strategy_instance = results[0]
            log_messages = strategy_instance.log_messages

            # Get final balance and calculate net profit/loss
            final_value = cerebro.broker.getvalue()
            net_profit_loss = final_value - initial_value  

            # Display results in a more structured way
            st.markdown(f"### Initial Balance: **${initial_value:.2f}**")
            st.markdown(f"### Final Balance: **${final_value:.2f}**")
            st.markdown(f"### Net Profit/Loss: **${net_profit_loss:.2f}**")
            st.markdown(f"### Total Buy Trades: **{strategy_instance.total_buy_trades}**")
            st.markdown(f"### Total Sell Trades: **{strategy_instance.total_sell_trades}**")
            
            # Calculate total trades
            total_trades = strategy_instance.total_buy_trades + strategy_instance.total_sell_trades

            # Safely calculate the win rate
            if total_trades > 0:
                win_rate = (strategy_instance.profitable_trades / total_trades) * 100
            else:
                win_rate = 0  # Avoid division by zero

            # Display the win rate in the Streamlit app
            st.markdown(f"### Win Rate: **{win_rate:.2f}%**")


            st.markdown(f'### Sharpe Ratio: {strategy_instance.sharpe_ratio:.2f}')  # Ensure sharpe_ratio is calculated correctly
            st.markdown(f'### Maximum Drawdown: {strategy_instance.max_drawdown:.2f}')  # Ensure max_drawdown is calculated correctly

            # Show logs in a scrollable div
            with st.expander("View Log Messages", expanded=False):
                for msg in log_messages:
                    st.text(msg)

            # Plot the results
            fig = cerebro.plot()[0][0]  
            st.pyplot(fig)