Why usFeaturesTemplatesBlogGlossary

Low Volatility 30

Holds the 30 calmest large- and mid-caps by realised annualised volatility, sized inversely to that volatility so the quietest names carry the most capital. The construction NSE uses for its Nifty100 Low Volatility 30 index, on the exchange's own published volatility series.

Why this is expected to work

The low-volatility anomaly is the most awkward fact in finance: low-risk stocks have historically delivered HIGHER risk-adjusted returns than high-risk stocks, inverting what CAPM predicts. Baker, Bradley and Wurgler (2011) attribute it to a structural cause rather than a statistical fluke — most institutional mandates are benchmarked and leverage-constrained, so managers who want higher returns must buy higher-beta stocks instead of levering a low-beta portfolio. That crowds the high-volatility end and leaves the quiet end persistently underpriced. Blitz and van Vliet (2007) found the effect across the US, Europe and Japan. This is also the template most worth running to see what a lower drawdown actually feels like.

low-volatilitydefensivefactorrisk
Universe
NSE Stocks XNSE
Rebalance
Twice a year Inverse volatility

How the pipeline works

Filter

NIFTY200 constituent

Keeps rows where nifty200_member == 1

nifty200_member == 1
Filter

Liquid: traded > Rs 5 cr/day

Keeps rows where turnover > 50000000

turnover > 50000000
Filter

Has a volatility reading

Keeps rows where annualised_volatility > 0

annualised_volatility > 0
Rank

30 calmest names

Selects the bottom 30 by annualised_volatility

What this template teaches

  • Exchange-published volatility as a first-class field
  • InverseVolatilityWeighting
  • Bottom-N ranking (order_type=bottom)
  • Semi-annual rebalancing to keep turnover low
Read the full write-up

The Low Volatility Anomaly: Getting Paid to Take Less Risk

Finance theory says higher risk earns higher return. Low-volatility stocks have historically delivered better risk-adjusted returns than high-volatility ones. The explanation is structural, not statistical.