Two index prices that move together still need their spread to stay tied to count as a pair

Two index prices that move together still need their spread to stay tied to count as a pairUpper panel: Nifty 50: from 100 at 1 Jul 2025 to 91 at 15 Sep 2026, high 103 at 2 Jan 2026, low 87 at 30 Mar 2026. Nifty Bank: from 100 at 1 Jul 2025 to 97 at 15 Sep 2026, high 107 at 18 Feb 2026, low 87 at 30 Mar 2026. Lower panel: z-score of the spread, with 27 readings beyond 2 standard deviations.Both rebased to 100 at the startNifty 50Nifty Bank9095100105z score of the spread, in standard deviations01−12−2Jul 2025OctJan 2026AprJul
Nifty 50 and bank index closes, rebased to the first day, from 1 Jul 2025 to 15 Sep 2026, with the z-score of their spread below. An illustration of the method, not a recommended pair. Source: NSE end of day data, computed by RupeeCase. As of 15 Sep 2026.

A pair is two prices that have moved together for a long time. The trade watches their spread, which is one price less a fitted multiple of the other, and takes a position when the spread sits far from its average. It exits when the spread returns. The chart shows two index prices and the z-score of their spread, as an illustration of the method and not a recommended pair.

Correlation is only the start. The two index series above move together closely from day to day, yet the test that matters is cointegration: the spread has to stay tied over time, so a move away is pulled back. Two prices can be highly correlated and still drift apart for long stretches.

Half-life measures how long a gap usually takes to close, and it sets the lookback and a time limit. The z-score measures how far the spread is from its mean, in units of its own volatility. These are measures, not rules. A threshold for the z-score is a choice, and a pair can move well past it.

Indian rules make pairs harder. A retail account cannot easily hold a cash short overnight, so both legs are often stock futures. The lot sizes differ, which makes the hedge ratio lumpy, and an F&O ban period can block new positions in one leg. Costs fall on both legs, including STT on the futures sale at 0.05% from 1 Apr 2026.

NSE publishes prices and contract data, not a list of pairs. A list shared elsewhere is chosen from past data. If the same data picks the pairs and tests them, the result is biased, so a pair should be tested on prices it was not picked from.

Before you trade. Derivatives trading carries a high risk of loss. SEBI's study "Profitability of Individual Traders in the Equity Derivatives Segment (FY25 to FY26)" (August 2026) found that 87.7% of individual traders in equity F&O made net losses in FY26, with aggregate net losses of Rs 91,685 crore. Read the SEBI study