Time Series Analysis in R
David S. Matteson
Associate Professor at Cornell University
Observed time series:
Weak stationary: mean, variance, covariance constant over time.
$Y_1, Y_2$, ...is a weakly stationary process if:
Covariance of $ Y_t$ and $ Y_s$ is same (constant) for all $ \vert t - s \vert = h$, for all$ h$.
$$ Cov(Y_2, Y_5) = Cov(Y_7, Y_{10})$$
since each pair is separated by three units of time.
A stationary process can be modeled with fewer parameters.
For example, we do not need a different expectation for each $ Y_t$; rather they all have a common expectation, $ \mu$.
Many financial time series do not exhibit stationarity, however:
Inflation rates and changes in inflation rates:
Time Series Analysis in R