Why is the cointegration model only applicable to two models?

Because cointegration allows us to describe the equilibrium or stationary relationship between two or more sequences. For each sequence alone, it may be non-stationary, and the moments of these sequences, such as mean, variance or covariance, have a co-integration relationship with time, so an error correction model can be established. If the error correction model can be established with nonstationary variables, there must be a co-integration relationship between variables. The stationarity of time series variables The stationarity of time series means that the statistical laws of time series will not change with the passage of time, that is, the characteristics of random processes that produce time series will not change with the passage of time. intuitively

2. Definition of unit root test and cointegration test of time series variables 1 Simplex process Simplex process is a special non-stationary random process. In short, unitary process refers to a nonstationary stochastic process that can be stabilized by difference.

3. Error correction model from cointegration to ARDL(p, q) Generally speaking, OLS estimation can only be carried out when both series are stationary time series. For non-stationary series, direct regression will lead to pseudo-regression (such as ...