Quant funds attract investments on the basis of technology, AI/ML. Here the past data is the main input. In India, this idea was promoted by quantitative analysts to choose portfolio management services (PMS). Some PMS portfolios have given returns as high as 30 per cent (risk-adjusted) every year over a decade.
Quant funds attract a lot of scepticism, but they have got a toe-hold now in the Indian market. Quants were initiated in the West in early 1970s. They were viewed favourably in India in the mid-2000s. Indian investors relied on fundamental or technical research or punting. In the global financial crisis of 2008, no one talked about quant investing.
Quantitative trading in a new version reappeared after technological advances and enhanced computing power. The investment strategies are assisted by AI and ML to provide superior returns. The funds using these techniques are 29 per cent in 2019. Customers too are adopting these strategies.
Such strategies eliminate the inherent human biases. Over a period of time, technology chooses a better portfolio than the human beings. Quant investment is based on data, mathematical models and quantitive analysis. Data is used to create a primary screener (filter) or an algorithm. The stocks selected meet the criteria. The filtered stocks are backtested. It creates the best possible portfolio.
There are so many permutations and combinations to structure a 10-stock Nifty portfolio. An ideal Nifty portfolio cannot be created by fund managers during their lifetime. Using technology, one can create the ideal portfolio with stock weightages.
Still people feel quant strategies are risky, and there could be heavy losses if the market crashes. It is not true. There is an in-built risk management feature in quant modelling.
Data forms the basis of all quant strategies. Data is analysed to create screeners. Rules and algorithms are the backbone. It all depends on the quality of data. Fund managers keep this secret. However, most of them use pointers such as price and volume (PV), PE ratios, cash flows, return on equity, stock technicals, dividends, share-holding patterns, DE ratios. There are macro factors such as traffic, credit card spends, mobility, weather etc.
Access to quality data has enhanced the effectiveness of quant strategies.
AI/ML have been recently introduced, that too marginally. Broader market trends can be assessed. These complement quant strategies.
Quants have outperformed benchmarked indices by about 90 per cent over a period of time. The ideal portfolio may have 25-30 stocks of which 60 per cent beat broader markets over a period of time, 20-30 per cent yielding market-level returns and just 10 per cent underperforming benchmarks. You generate significant alpha if you get 60-70 per cent of portfolio right.