Univariate and multivariate time series forecasting with uncertainty quantification (including simulation).
This package is a Julia wrapper for R package ahead,
and it follows its API as closely as possible: see https://techtonique.r-universe.dev/ahead#reference for details.
julia> using Pkg; Pkg.add(path = "https://github.com/Techtonique/Ahead.jl")using Ahead
y = [1,2,3,4,5,6,7,8,9,10]
z = rand(8, 2)
u = rand(50)Univariate forecasting with dynrmf, Dynamic regression model
val1 = Ahead.dynrmf(y, h=6)
println(val1) Multivariate forecasting with ridge2f, Random Vector functional link network model with 2 regularization parameters (see https://www.mdpi.com/2227-9091/6/1/22)
val2 = Ahead.ridge2f(z, h=4)
println(val2) - Use TimeSeries.jl (https://juliastats.org/TimeSeries.jl/stable/)
docker build -t ahead .&&docker run ahead
- Install
RCall: https://juliainterop.github.io/RCall.jl/stable/installation/ - Getting started with
RCall: https://juliainterop.github.io/RCall.jl/latest/gettingstarted.html - Julia
RCall: https://www.geeksforgeeks.org/julia-rcall/
$ rm -rf ~/.julia/compiled # clearing cache
$ rm -rf ~/.julia/lib # clearing cache
julia> using Pkg; Pkg.rm("Ahead")
pkg> rm Ahead
julia> using Pkg; Pkg.add(path="https://github.com/Techtonique/Ahead.jl")
julia> using Pkg; Pkg.update("Ahead")
pkg> activate Ahead