Air pollution data recorded in the U.S. are multivariate time series of many cities. Statistics analysis was used and tried to analyze multivariate air pollution data and discovered that the time series methods fit to the air pollution data. The Singular Spectrum Analysis (SSA) method was developed as the new time series method since 1970s and they are still growing mathematically. SSA regards as a model-free approach because SSA decomposes an original time series to trend, seasonal, semi-seasonal, and white noises according to the singular value decomposition (SVD) [4]. The new decomposed series help us to understand the trend of the original time series and to extract seasonal or monthly components and white noises. In recent years, SSA did not only apply in the analysis of climatic and geophysical time series, but also in the analysis of social science and economic time series. With these diverse areas, air pollution time series in Los Angeles is applied to SSA. As a methodology, the theories of SSA will be discussed in the paper.