Superfast Likelihood Inference for Stationary Gaussian Time Series
Convert position autocorrelations to increment autocorrelations.
Convert autocorrelation of stationary increments to mean squared displ...
Cholesky multiplication with Toeplitz variance matrices.
Constructor and methods for Circulant matrix objects.
Density of a multivariate normal with Toeplitz variance matrix.
Mean square displacement of fractional Brownian motion.
Matern autocorrelation function.
Convert mean square displacement of positions to autocorrelation of in...
Multivariate normal with Circulant variance matrix.
Multivariate normal with Toeplitz variance matrix.
Power-exponential autocorrelation function.
Simulate a stationary Gaussian time series.
Defunct functions in SuperGauss.
Superfast inference for stationary Gaussian time series.
Toeplitz matrix multiplication.
Constructor and methods for Toeplitz matrix objects.
Likelihood evaluations for stationary Gaussian time series are typically obtained via the Durbin-Levinson algorithm, which scales as O(n^2) in the number of time series observations. This package provides a "superfast" O(n log^2 n) algorithm written in C++, crossing over with Durbin-Levinson around n = 300. Efficient implementations of the score and Hessian functions are also provided, leading to superfast versions of inference algorithms such as Newton-Raphson and Hamiltonian Monte Carlo. The C++ code provides a Toeplitz matrix class packaged as a header-only library, to simplify low-level usage in other packages and outside of R.