TRADO <> IIT Ropar Hackathon

Welcome to the Trado <> IIT Ropar Hackathon! This is the second part of a two part hackathon conducted by Trado, at IIT Ropar.

Build a backtesting engine to evaluate an options trading strategy on Indian indices.

Problem Statement

Create a backtesting engine that tests a defined option buying strategy using historical market data from a PostgreSQL/TimescaleDB database.

Strategy to Test

1. Entry: At the configured entry time (default: 9:25 AM):

   • Buy 1 lot of ATM CE
   • Buy 1 lot of ATM PE

2. Exit Criteria:

   • Individual Stop Loss: Exit a position if it loses 25% of its premium
   • Combined Target: Exit both positions if they give a total P&L of 25%
   • Combined Stop Loss: Exit both positions if they give a total P&L of -10%
   • End of Day: Square off at 3:15 PM

3. Adjustment Logic:

   • If both positions are exited before 2:00 PM, re-enter with fresh ATM options
   • Only one adjustment per day allowed
   • Adjustment positions held until market close without SL/Target

Your Task

1. Complete the boilerplate code by implementing:

   • Data retrieval from the database
   • Strategy logic and position management
   • P&L calculation and reporting

2. The code should:

   • Work with any of the supported indices (BANKNIFTY, NIFTY, FINNIFTY, MIDCPNIFTY)
   • Support configurable parameters (stop loss percentages, target percentages)
   • Process data for the last 3 trading days

Database Structure

The database already has historical data stored with this schema:

-- topics table maps instruments to IDs
CREATE TABLE topics (
    topic_id SERIAL PRIMARY KEY,
    topic_name TEXT NOT NULL UNIQUE,
    index_name TEXT,
    type TEXT,
    strike NUMERIC
);

-- ltp_data table contains time-series price data
CREATE TABLE ltp_data (
    id INTEGER NOT NULL,
    topic_id INTEGER REFERENCES topics(topic_id),
    ltp NUMERIC(10, 2) NOT NULL,
    received_at TIMESTAMPTZ NOT NULL,
    PRIMARY KEY (id, received_at)
);

Setup Instructions

1. Install dependencies:

   npm install

2. Create .env file from the sample:

   cp .env.sample .env

3. Run the application:

   npm start