We explore whether more sensitive volatility estimates (less data) or more stable volatility estimates (more data) produce better risk parity results.