This package implements and extends the algorithm 'Sequential Adaptive Nonlinear Modeling of Time Series (SLANTS)'. SLANTS is a new method for online modeling and prediction of nonlinear and nonparametric autoregressive time series. It uses splines to approximate a wide range of nonlinear functions and adaptive filtering to accommodate time varying data generating processes. It is built on a new online group LASSO algorithm proposed in the reference paper. It can be applied to high dimensional time series where the dimension is larger than the sample size.
First install the devtools package
install.packages("devtools")
library("devtools")
Then install this package
install_github('JieGroup/slants')
To see the available function to use, type
ls("package:slants")
This open source project is still ongoing. A preliminary user guide of this package can be found here
Q .Han, J. Ding, E. Airoldi, V. Tarokh, "SLANTS: Sequential adaptive nonlinear modeling of time series," IEEE Transactions on Signal Processing 65 (19), 4994-5005. pdf
X. Xian, J. Ding, "Physics-assisted online learning," preprint.
This research is funded by the Defense Advanced Research Projects Agency (DARPA) under grant number HR00111890040.