Learn R Programming

⚠️There's a newer version (1.0.2) of this package.Take me there.

onlineforecast (version 1.0.0)

Forecast Modelling for Online Applications

Description

A framework for fitting adaptive forecasting models. Provides a way to use forecasts as input to models, e.g. weather forecasts for energy related forecasting. The models can be fitted recursively and can easily be setup for updating parameters when new data arrives. See the included vignettes, the website and the pre-print paper "onlineforecast: An R package for adaptive and recursive forecasting" .

Copy Link

Version

Install

install.packages('onlineforecast')

Monthly Downloads

398

Version

1.0.0

License

GPL-3

Maintainer

Peder Bacher

Last Published

September 30th, 2021

Functions in onlineforecast (1.0.0)

Compute base splines of a variable using the R function splines::bs, use in the transform stage.

Generation of a name for a cache file for the value of a function.

Checking the object for appropriate form.

Save a cache file (name generated with code_name()

Auto-Regressive (AR) input

check.data.list

Checking the data.list for appropriate form.

Convertion to POSIXlt

as.data.frame.data.list

Convert to data.frame

Observations and weather forecasts from a single-family building, weather station and Danish Meteorological Institute (DMI)

Convert to data.list class

Determine if two data.lists are identical

Convertion to POSIXct

Selects a period

Find complete cases in forecast matrices

Multiplication of list with y, elementwise

Lagging which returns a data.list

Lag by shifting

Generation of Fourrier series.

lapply_rbind_df

Helper which does lapply, rbind and then as.data.frame

Class for forecastmodels

Helper which does lapply and then rbind

Make a data.list

Depth of a list

Long format of prediction data.frame

Prediction with an lm forecast model.

Create ones for model input intercept

Functions for online forecasting

Class for forecastmodel inputs

Make a forecast matrix (as data.frame) from observations.

Simple wrapper for graphics.off()

Make an hour-of-day forecast matrix

Getting subelement from list.

First-order low-pass filtering

Wrapper for bspline with periodic=TRUE

Fit an onlineforecast model with lm

Optimize parameters for onlineforecast model fitted with LM

Lagging which returns a data.frame

Optimize parameters for onlineforecast model fitted with RLS

Fit an onlineforecast model with Recursive Least Squares (RLS).

Low pass filtering of a vector.

Time series plotting

print_to_message

Simple function for capturing from the print function and send it in a message().

Return the column names

Computes the RMSE score.

Plotting stairs with time point at end of interval.

Simple wrapper for paste0().

Calculating k-step recursive least squares estimates

print.forecastmodel

Print forecast model

Set a state value to be kept for next the transformation function is called.

Forward and backward model selection

Get the state value kept in last call.

Generate persistence forecasts

residuals.data.frame

Calculate the residuals given a forecast matrix and the observations.

Setting par() plotting parameters

Calculate the score for each horizon.

Resampling to equidistant time series

Helper which does lapply and then cbind

lapply_cbind_df

Helper which does lapply, cbind and then as.data.frame

Updates the model fits

Print summary of an onlineforecast model fitted with RLS

Make an forecast matrix with a periodic time signal.

Set parameters for plot_ts()

pairs.data.list

Generation of pairs plot for a data.list.

Prediction with an rls model.

subset.data.list

Take a subset of a data.list.

Function for generating the parameters for RLS regression