1: \begin{abstract}

2: We consider the sparse principal component analysis for high-dimensional stationary processes.

3: The standard principal component analysis performs poorly when the dimension of the process is large.

4: We establish the oracle inequalities for penalized principal component estimators for the processes including heavy-tailed time series.

5: The rate of convergence of the estimators is established.

6: %\ttr{

7: %The rate of convergence of the estimators is established even when the dimension allows to be larger than the sample size.}

8: %grows at the exponential rate of the sample size.

9: We also elucidate the theoretical rate for choosing the tuning parameter in penalized estimators.

10: The performance of the sparse principal component analysis is demonstrated by numerical simulations.

11: The utility of the sparse principal component analysis for time series data is exemplified by the application to average temperature data.

12: \end{abstract}

13: