Testing for cointegration is performed by:

Correct option is C

​Correct Option: (C): Engel-Granger test​

Explanation:
1. Cointegration is a statistical concept used in time series analysis. It refers to a situation where two or more non-stationary series are linked by a long-run, stable equilibrium relationship. Even though individual time series may drift over time (non-stationary), their linear combination may be stationary, indicating cointegration.
2. To test for cointegration, one of the most widely used procedures is the Engel-Granger test, introduced by Clive Granger and Robert Engle (1987), who later won the Nobel Prize for this contribution.

Information Booster:
​Steps of Engel-Granger Test:
1. Test for stationarity of individual series using ADF (Augmented Dickey-Fuller) or PP (Phillips-Perron) test.
2. If series are found to be integrated of the same order (commonly I(1)), estimate the regression equation.
3. Save the residuals from the regression and test whether residuals are stationary using the ADF test.
4. If residuals are stationary, the variables are cointegrated.
Thus, Engel-Granger test is specifically designed for testing cointegration between variables. 

Additional Knowledge:
* Option (A) Chow test: This is used to detect structural breaks in time series data. It checks whether the coefficients in two linear regressions on different datasets are equal. It does not test for cointegration.
* Option (B) Phillips-Peron test: This is a unit root test used to determine the stationarity of a time series. It is an alternative to the Augmented Dickey-Fuller (ADF) test and does not test for cointegration.
* Option (D) Error-correction mechanism: This is a model used after cointegration has been established. It captures short-term deviations from the long-term equilibrium and corrects them. It is not a test for cointegration but a modeling technique once cointegration is known.