Quant Strategy

Investment Philosophy

Take investment decisions based on data analysis, minimising emotional bias

Optimise strategies based on rigorous backtesting, using machine learning techniques

Diversify across factor models, sectors and instruments to mitigate risk

Monitor and adapt quant models based on evolving market conditions

Investment Process

Data acquisition and processing

Collect and curate high quality financial data from stock exchanges, data providers, and regulatory filings

Factor model development

Develop factor models tailored to the Indian market, incorporating domestic factors and their historical performance

Backtesting and optimisation

Backtest and optimise the models and portfolio construction methodology to ensure effectiveness

Portfolio construction

Construct a diversified portfolio based on the signals generated by the quantitative models, considering liquidity constraints

Reweighting and monitoring

Continuously monitor market movements and rebalance the portfolio strategically to capture evolving opportunities and manage risk

Techniques:

Multi-factor investing: Exploits a combination of factors like value, momentum, quality, volatility, and liquidity specific to the Indian market

Mean reversion strategies: Capitalises on temporary price deviations of Indian stocks from their historical averages using statistical arbitrage techniques

Statistical arbitrage: Identifies short-term price discrepancies between correlated assets and capitalise on them

Machine learning: Utilises advanced ML algorithms to uncover complex relationships within vast datasets and generate alpha

Tech Stack:

Quantitative programming languages: Utilises Python, R for data analysis, model development, and backtesting

Cloud computing: Utilises cloud-based platforms for efficient data storage, processing, and model training

Machine learning libraries: Leverages libraries like TensorFlow, PyTorch, and scikit-learn for building and deploying ML models

Rich data APIs: Incorporates quantitative, fundamental and technical data sources

Machine learning model development: Trains ML models on historical data to predict future asset prices and market movements

Risk Management

Liquidity filters

Ensure sufficient portfolio liquidity to meet redemption needs and exploit attractive trading opportunities

Position sizing ceiling

Volatility based position sizing strategies to limit potential losses on any individual trade. This will involve position sizing strategies and dynamic portfolio adjustments

Diversification

Construct well diversified portfolio across multiple stocks/instruments. No position is allowed to grow beyond risk caps

Portfolio level drawdown limit

Market data and rules ensure portfolio adjustments are made before drawdowns grow to uncontrollable levels

Strategy Facts

Securities

  • Multicap Indian stocks
  • Tactical allocation to US ETFs, commodity ETFs and debt instruments
  • Derivatives for hedging

Benchmark

S&P BSE 500 TRI

Partners

  • Custodian: Orbis Financial
  • Brokers: Zerodha, Emkay

Minimum investment

INR 50 lakhs (total across strategies)