Predicting and Forecasting Time Series via GMDH-Type Neural Network Algorithms (GMDH)
Group method of data handling (GMDH) – type neural network algorithm is the heuristic self-organization method for modelling the complex systems. In this package, GMDH-type neural network algorithms are applied to predict and forecast a univariate time series.
Store and Organize Time Series in a Database (timeseriesdb)
An R package to store and organize time series and their multi-lingual meta information in a database. The timeseriesdb package suggests a simple yet powerful database structure to store a large amount of time series in a relational PostgreSQL database. The package provides an interface to the R user to create, update and delete time series.
Point density for geospatial data (pointdensityP)
pointdensity returns a density count and the temporal average for every point in the original list. The dataframe returned includes four columns: lat, lon, count, and date_avg. The “lat” column is the original latitude data; the “lon” column is the original longitude data; the “count” is the density count of the number of points within a radius of radius*grid_size (the neighborhood); and the date_avg column includes the average date of each point in the neighborhood.
Ordered Lasso and Time-lag Sparse Regression (orderedLasso)
Ordered lasso and time-lag sparse regression. Ordered Lasso fits a linear model and imposes an order constraint on the coefficients. It writes the coefficients as positive and negative parts, and requires positive parts and negative parts are non-increasing and positive. Time-Lag Lasso generalizes the ordered Lasso to a general data matrix with multiple predictors. For more details, see Suo, X.,Tibshirani, R., (2014) ‘An Ordered Lasso and Sparse Time-lagged Regression’.
Method for Clustering Partially Observed Data (kpodclustr)
The kpodclustr package implements the k-POD method for clustering partially observed data.
Estimation and Simulation for PDMPs (EstSimPDMP)
This package deals with the estimation of the jump rate for piecewise-deterministic Markov processes (PDMPs), from only one observation of the process within a long time. The main functions provide an estimate of this function. The state space may be discrete or continuous. The associated paper has been published in Scandinavian Journal of Statistics and is given in references. Other functions provide a method to simulate random variables from their (conditional) hazard rate, and then to simulate PDMPs.