Maximum Entropy Bootstrap for Time Series

Maximum entropy density based dependent data bootstrap. An algorithm is provided to create a population of time series (ensemble) without assuming stationarity. The reference paper (Vinod, H.D., 2004 ) explains how the algorithm satisfies the ergodic theorem and the central limit theorem.

Multivariate Time Series Data Imputation

This is an EM algorithm based method for imputation of missing values in multivariate normal time series. The imputation algorithm accounts for both spatial and temporal correlation structures. Temporal patterns can be modeled using an ARIMA(p,d,q), optionally with seasonal components, a non-parametric cubic spline or generalized additive models with exogenous covariates. This algorithm is specially tailored for climate data with missing measurements from several monitors along a given region.

Methods for Temporal Disaggregation and Interpolation of Time Series

Temporal disaggregation methods are used to disaggregate and interpolate a low frequency time series to a higher frequency series, where either the sum, the mean, the first or the last value of the resulting high frequency series is consistent with the low frequency series. Temporal disaggregation can be performed with or without one or more high frequency indicator series. Contains the methods of Chow-Lin, Santos-Silva-Cardoso, Fernandez, Litterman, Denton and Denton-Cholette, summarized in Sax and Steiner (2013) . Supports most R time series classes.

Goodness-of-Fit Functions for Comparison of Simulated and Observed Hydrological Time Series

S3 functions implementing both statistical and graphical goodness-of-fit measures between observed and simulated values, mainly oriented to be used during the calibration, validation, and application of hydrological models. Missing values in observed and/or simulated values can be removed before computations. Comments / questions / collaboration of any kind are very welcomed.

Deep Learning Model for Time Series Forecasting

RNNs are preferred for sequential data like time series, speech, text, etc. but when dealing with long range dependencies, vanishing gradient problems account for their poor performance. LSTM and GRU are effective solutions which are nothing but RNN networks with the abilities of learning both short-term and long-term dependencies. Their structural makeup enables them to remember information for a long period without any difficulty. LSTM consists of one cell state and three gates, namely, forget gate, input gate and output gate whereas GRU comprises only two gates, namely, reset gate and update gate. This package consists of three different functions for the application of RNN, LSTM and GRU to any time series data for its forecasting. For method details see Jaiswal, R. et al. (2022). .

Time Series Management and Analysis for Hydrological Modelling

S3 functions for management, analysis, interpolation and plotting of time series used in hydrology and related environmental sciences. In particular, this package is highly oriented to hydrological modelling tasks. The focus of this package has been put in providing a collection of tools useful for the daily work of hydrologists (although an effort was made to optimise each function as much as possible, functionality has had priority over speed). Bugs / comments / questions / collaboration of any kind are very welcomed, and in particular, datasets that can be included in this package for academic purposes.

Dimension Reduction Methods for Multivariate Time Series

Estimates VAR and VARX models with Structured Penalties using the methods developed by Nicholson et al (2017) and Nicholson et al (2020) .

Rmetrics - Autoregressive Conditional Heteroskedastic Modelling

Analyze and model heteroskedastic behavior in financial time series.

Time Value of Money, Time Series Analysis and Computational Finance

Package for time value of money calculation, time series analysis and computational finance.

Superfast Likelihood Inference for Stationary Gaussian Time Series

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.