This is based on the idea that a linear combination of 2 non-stationary time series is a stationary time series. So , what ? Well there are some interesting applications. I will try to list a few

- Index based strategies : Lets say you are running an index fund and you want to minimize tracking error. How do you do it ? Take a basket of stocks, regress with the index and then decide the weights based on the parameters . The question then arises, how frequent should this exercise be done ? The basic concept behind such a method would be correlation based and by definition, correlation is a short term measure. It doesn't say a thing about the drift or long term relationship. When things change , the correlation based approach imposes the task of recomputing weights. However, a cointegration based would be better as you one is likely to rebalance the portfolio less number of times, meaning, if you know that the index and the set of stocks are cointegrated , then one is more certain that the fluctuations of index returns and fund returns would be mean reverting
- Commodities : It can be applied to commodities where cost of carry is similar to the commodities used for cointegration
- Spot and Futures trading : Spots and futures converge at maturity, and hence the basis must be mean reverting.
- Spread Options
- Term Structures
- Pairs trading
- Forex (a little difficult though according to the current state of research)

So, one sees there is a lot of stuff that can be achieved through cointegration based trading strategies. However the key lies in understanding the residuals. How to look at the residuals and take long short positions . Here in lies the scope for understanding better. Some folks use impulse response modeling to see to take such calls.

Possible applications in India

- Designing ETFs
- Vol Spread trading products
- Factor Models with trading rules.

and many more i guess.One just needs to think through these products and there can be a terrific market for the products in India considering that it has seen a great volatility in the recent times.