4 packages on CRAN
Contains some tools for testing, analyzing time series data and fitting popular time series models such as ARIMA, Moving Average and Holt Winters, etc. Most functions also provide nice and clear outputs like SAS does, such as identify, estimate and forecast, which are the same statements in PROC ARIMA in SAS.
Contains the tools to screen grouped variables, and select screened grouped variables afterwards. The main function grpss() can perform the grouped variables screening as well as selection for ultra-high dimensional data with group structure. The screening step is primarily used to reduce the dimensions of data so that the selection procedure can easily handle the moderate or low dimensions instead of ultra-high dimensions.
Uses refined moving average filter based on the optimal and data-driven moving average lag q or smoothing spline to estimate trend and seasonal components, as well as irregularity (residuals) for univariate time series or data.
contains several supplementary non-parametric statistics methods including quantile test, Cox-Stuart trend test, runs test, normal score test, kernel PDF and CDF estimation, kernel regression estimation and kernel Kolmogorov-Smirnov test.