rename the package (#65)

LAMPSPUC · Feb 4, 2020 · 7bb02f8 · 7bb02f8 · guilhermebodin · Feb 4, 2020
1 parent 0008ada
commit 7bb02f8
Show file tree

Hide file tree

Showing 21 changed files with 109 additions and 109 deletions.
diff --git a/Project.toml b/Project.toml
@@ -1,4 +1,4 @@
-name = "GAS"
+name = "ScoreDrivenModels"
 uuid = "4a87933e-d659-11e9-0e65-7f40dedd4a3a"
 authors = ["guilhermebodin <[email protected]>, raphaelsaavedra <[email protected]>"]
 version = "0.1.0"

diff --git a/README.md b/README.md
@@ -1,13 +1,13 @@
-# GAS.jl
+# ScoreDrivenModels.jl
 
 | **Build Status** | **Coverage** | **Documentation** |
 |:-----------------:|:-----------------:|:-----------------:|
-| [![Build Status][build-img]][build-url] | [![Codecov branch][codecov-img]][codecov-url] |[![](https://img.shields.io/badge/docs-latest-blue.svg)](https://lampspuc.github.io/GAS.jl/latest/)
+| [![Build Status][build-img]][build-url] | [![Codecov branch][codecov-img]][codecov-url] |[![](https://img.shields.io/badge/docs-latest-blue.svg)](https://lampspuc.github.io/ScoreDrivenModels.jl/latest/)
 
-[build-img]: https://travis-ci.org/LAMPSPUC/GAS.jl.svg?branch=master
-[build-url]: https://travis-ci.org/LAMPSPUC/GAS.jl
+[build-img]: https://travis-ci.org/LAMPSPUC/ScoreDrivenModels.jl.svg?branch=master
+[build-url]: https://travis-ci.org/LAMPSPUC/ScoreDrivenModels.jl
 
-[codecov-img]: https://codecov.io/gh/LAMPSPUC/GAS.jl/coverage.svg?branch=master
-[codecov-url]: https://codecov.io/gh/LAMPSPUC/GAS.jl?branch=master
+[codecov-img]: https://codecov.io/gh/LAMPSPUC/ScoreDrivenModels.jl/coverage.svg?branch=master
+[codecov-url]: https://codecov.io/gh/LAMPSPUC/ScoreDrivenModels.jl?branch=master
 
-GAS.jl is a Julia package for modeling, forecasting, and simulating time series with generalized autoregressive score models (GAS), also known as score-driven models. Implementations are based on the paper [Generalized Autoregressive Models with Applications](http://dx.doi.org/10.1002/jae.1279) by D. Creal, S. J. Koopman, and A. Lucas.
+ScoreDrivenModels.jl is a Julia package for modeling, forecasting, and simulating time series with generalized autoregressive score models (ScoreDrivenModels), also known as score-driven models. Implementations are based on the paper [Generalized Autoregressive Models with Applications](http://dx.doi.org/10.1002/jae.1279) by D. Creal, S. J. Koopman, and A. Lucas.
diff --git a/bench/benchmark_estimate.jl b/bench/benchmark_estimate.jl
@@ -1,4 +1,4 @@
-using GAS, Distributions, BenchmarkTools, Random, Test
+using ScoreDrivenModels, Distributions, BenchmarkTools, Random, Test
 
 include("test/utils.jl")
 
@@ -7,7 +7,7 @@ scaling = 0.0
 A = [0.2 0; 0 0.2]
 B = [0.2 0; 0 0.2]
 simulation = simulate_GAS_1_1(Beta, scaling, ω, A, B, 1)
-gas = GAS.Model(1, 1, Beta, scaling)
+gas = ScoreDrivenModels.Model(1, 1, Beta, scaling)
 gas.ω = ω
 gas.A[1] = A
 gas.B[1] = B
@@ -34,8 +34,8 @@ p = 1
 q = 1
 num_seeds = 3
 @benchmark begin
-    gas = GAS.Model($p, $q, $Beta, $scaling)
-    opt_method = GAS.LBFGS(gas, $num_seeds)
+    gas = ScoreDrivenModels.Model($p, $q, $Beta, $scaling)
+    opt_method = ScoreDrivenModels.LBFGS(gas, $num_seeds)
     fit!(gas, $simulation; verbose = $verbose, opt_method = opt_method)
 end
 # BenchmarkTools.Trial:
@@ -60,8 +60,8 @@ p = 1
 q = 1
 num_seeds = 3
 @benchmark begin
-    gas = GAS.Model($p, $q, $LogNormal, $scaling)
-    opt_method = GAS.LBFGS(gas, $num_seeds)
+    gas = ScoreDrivenModels.Model($p, $q, $LogNormal, $scaling)
+    opt_method = ScoreDrivenModels.LBFGS(gas, $num_seeds)
     fit!(gas, $simulation; verbose = $verbose, opt_method = opt_method)
 end
 # BenchmarkTools.Trial:
@@ -79,8 +79,8 @@ end
 
 scaling = 0.5
 @benchmark begin
-    gas = GAS.Model($p, $q, $LogNormal, $scaling)
-    opt_method = GAS.LBFGS(gas, $num_seeds)
+    gas = ScoreDrivenModels.Model($p, $q, $LogNormal, $scaling)
+    opt_method = ScoreDrivenModels.LBFGS(gas, $num_seeds)
     fit!(gas, $simulation; verbose = $verbose, opt_method = opt_method)
 end
 # BenchmarkTools.Trial:
@@ -97,8 +97,8 @@ end
 
 scaling = 1.0
 @benchmark begin
-    gas = GAS.Model($p, $q, $LogNormal, $scaling)
-    opt_method = GAS.LBFGS(gas, $num_seeds)
+    gas = ScoreDrivenModels.Model($p, $q, $LogNormal, $scaling)
+    opt_method = ScoreDrivenModels.LBFGS(gas, $num_seeds)
     fit!(gas, $simulation; verbose = $verbose, opt_method = opt_method)
 end
 # BenchmarkTools.Trial:

diff --git a/docs/make.jl b/docs/make.jl
@@ -1,11 +1,11 @@
-using Documenter, GAS
+using Documenter, ScoreDrivenModels
 
 makedocs(
-    modules = [GAS],
+    modules = [ScoreDrivenModels],
     doctest  = false,
     clean    = true,
     format   = Documenter.HTML(mathengine = Documenter.MathJax()),
-    sitename = "GAS.jl",
+    sitename = "ScoreDrivenModels.jl",
     authors  = "Guilherme Bodin and Raphael Saavedra",
     pages   = [
         "Home" => "index.md",
@@ -15,5 +15,5 @@ makedocs(
 )
 
 deploydocs(
-    repo = "github.com/LAMPSPUC/GAS.jl.git",
+    repo = "github.com/LAMPSPUC/ScoreDrivenModels.jl.git",
 )
diff --git a/docs/src/examples.md b/docs/src/examples.md
@@ -2,15 +2,15 @@
 
 ## Water inflow
 
-Let's model some monthly water inflow data from the Northeast of Brazil using a Lognormal GAS model. Since inflow is a highly seasonal phenomenon, we will utilize lags 1 and 12. The former aims to characterize the short-term evolution of the series, while the latter characterizes the seasonality. The full code is in the examples folder.
+Let's model some monthly water inflow data from the Northeast of Brazil using a Lognormal ScoreDrivenModels model. Since inflow is a highly seasonal phenomenon, we will utilize lags 1 and 12. The former aims to characterize the short-term evolution of the series, while the latter characterizes the seasonality. The full code is in the examples folder.
 
 ```julia
 # Convert data to vector
 y = Vector{Float64}(vec(inflow'))
 
-# Specify GAS model: here we use lag 1 for trend characterization and 
+# Specify ScoreDrivenModels model: here we use lag 1 for trend characterization and 
 # lag 12 for seasonality characterization
-gas = GAS.Model([1, 12], [1, 12], LogNormal, 0.0)
+gas = ScoreDrivenModels.Model([1, 12], [1, 12], LogNormal, 0.0)
 
 # Estimate the model via MLE
 fit!(gas, y)

diff --git a/docs/src/index.md b/docs/src/index.md
@@ -1,16 +1,16 @@
-# GAS.jl Documentation
+# ScoreDrivenModels.jl Documentation
 
-GAS.jl is a Julia package for modeling, forecasting and simulating time series with or generalized autoregressive score models (GAS) models, also known as dynamic conditional score models (DCS). Implementations are based on the paper [Generalized Autoregressive Models with Applications](http://dx.doi.org/10.1002/jae.1279) by D. Creal, S. J. Koopman and A. Lucas.
+ScoreDrivenModels.jl is a Julia package for modeling, forecasting and simulating time series with or generalized autoregressive score models (ScoreDrivenModels) models, also known as dynamic conditional score models (DCS). Implementations are based on the paper [Generalized Autoregressive Models with Applications](http://dx.doi.org/10.1002/jae.1279) by D. Creal, S. J. Koopman and A. Lucas.
 
 ## Installation
 
 This package is not yet registered so you can `Pkg.add` it as follows:
 ```julia
-pkg> add https://github.com/LAMPSPUC/GAS.jl.git
+pkg> add https://github.com/LAMPSPUC/ScoreDrivenModels.jl.git
 ```
 
 ## Contributing
 
-Contributions to this package are more than welcome, if you find a bug or have any suggestions for the documentation please post it on the [github issue tracker](https://github.com/LAMPSPUC/GAS.jl/issues).
+Contributions to this package are more than welcome, if you find a bug or have any suggestions for the documentation please post it on the [github issue tracker](https://github.com/LAMPSPUC/ScoreDrivenModels.jl/issues).
 
 When contributing please note that the package follows the [JuMP style guide](https://www.juliaopt.org/JuMP.jl/stable/style/).
diff --git a/docs/src/manual.md b/docs/src/manual.md
@@ -6,7 +6,7 @@
 
 ## Links
 
-Links are reparametrizations utilized to ensure certain parameter is within its original domain, i.e. in a distribution one would like to ensure that the time varying parameter ``f \in \mathbb{R}^+``. The way to do this is to model ``\tilde{f} = \ln{f}``. More generally one can stablish that ``\tilde{f} = h(f)``. We refer to this procedure as **linking**. When the parameter is linked the GAS recursion happens in the domain of ``\tilde{f}`` and then one can recover the orginal parameter by ``f = \left(h\right)^-1(\tilde f)``. We refer to this procedure as **unlinking**. The new GAS recursion becomes.
+Links are reparametrizations utilized to ensure certain parameter is within its original domain, i.e. in a distribution one would like to ensure that the time varying parameter ``f \in \mathbb{R}^+``. The way to do this is to model ``\tilde{f} = \ln{f}``. More generally one can stablish that ``\tilde{f} = h(f)``. We refer to this procedure as **linking**. When the parameter is linked the ScoreDrivenModels recursion happens in the domain of ``\tilde{f}`` and then one can recover the orginal parameter by ``f = \left(h\right)^-1(\tilde f)``. We refer to this procedure as **unlinking**. The new ScoreDrivenModels recursion becomes.
 
 ```math
 \begin{equation*}\left\{\begin{array}{ccl}
@@ -17,24 +17,24 @@ Links are reparametrizations utilized to ensure certain parameter is within its
 \end{equation*}
 ```
 
-Notice that the change in parametrization changes the dynamics of the model. The GAS(1,1) for a Normal distribution with inverse scaling ``d = 1`` is equivalent to the GARCH(1, 1) model, but only on the original parameter, if you work with a different parametrization the model is no longer equivalent.
+Notice that the change in parametrization changes the dynamics of the model. The ScoreDrivenModels(1,1) for a Normal distribution with inverse scaling ``d = 1`` is equivalent to the GARCH(1, 1) model, but only on the original parameter, if you work with a different parametrization the model is no longer equivalent.
 
 
 ### Types of links
 
 The abstract type `Link` subsumes any type of link that can be expressed.
 
 ```@docs
-GAS.IdentityLink
-GAS.LogLink
-GAS.LogitLink
+ScoreDrivenModels.IdentityLink
+ScoreDrivenModels.LogLink
+ScoreDrivenModels.LogitLink
 ```
 
 ### Link functions
 
 ```@docs
-GAS.link
-GAS.unlink
-GAS.jacobian_link
+ScoreDrivenModels.link
+ScoreDrivenModels.unlink
+ScoreDrivenModels.jacobian_link
 ```
 
diff --git a/examples/garch.jl b/examples/garch.jl
@@ -1,21 +1,21 @@
-using GAS, Statistics, DelimitedFiles
+using ScoreDrivenModels, Statistics, DelimitedFiles
 
-# The GARCH(1, 1) is equivaent to a GAS.Model(1, 1) Normal model with inverse scaling. The models are only equivalent under 
+# The GARCH(1, 1) is equivaent to a ScoreDrivenModels.Model(1, 1) Normal model with inverse scaling. The models are only equivalent under 
 # no reparametrizations, here this means that the you should fit the model overwriting the default link!, unlink! and
 # jacobian_link! methods for the Normal distribution to use only the IdentityLink.
 
 # Overwrite link interface to work with Identity links.
-function GAS.link!(param_tilde::Matrix{T}, ::Type{Normal}, param::Matrix{T}, t::Int) where T 
+function ScoreDrivenModels.link!(param_tilde::Matrix{T}, ::Type{Normal}, param::Matrix{T}, t::Int) where T 
     param_tilde[t, 1] = link(IdentityLink, param[t, 1])
     param_tilde[t, 2] = link(IdentityLink, param[t, 2])
     return
 end
-function GAS.unlink!(param::Matrix{T}, ::Type{Normal}, param_tilde::Matrix{T}, t::Int) where T 
+function ScoreDrivenModels.unlink!(param::Matrix{T}, ::Type{Normal}, param_tilde::Matrix{T}, t::Int) where T 
     param[t, 1] = unlink(IdentityLink, param_tilde[t, 1])
     param[t, 2] = unlink(IdentityLink, param_tilde[t, 2])
     return
 end
-function GAS.jacobian_link!(aux::AuxiliaryLinAlg{T}, ::Type{Normal}, param::Matrix{T}, t::Int) where T 
+function ScoreDrivenModels.jacobian_link!(aux::AuxiliaryLinAlg{T}, ::Type{Normal}, param::Matrix{T}, t::Int) where T 
     aux.jac[1] = jacobian_link(IdentityLink, param[t, 1])
     aux.jac[2] = jacobian_link(IdentityLink, param[t, 2])
     return
@@ -30,7 +30,7 @@ initial_params = [mean(y) var(y)]
 # There are 2 possibilities to estimate the model.
 # The first one is to understand that in the GARCH estimation process usually one 
 # defines that the parameters ω, α_0, α_1 and β_1 are constrained. The problem is 
-# that the equivalent GAS model relies that 
+# that the equivalent ScoreDrivenModels model relies that 
 # ω_1 = ω
 # ω_2 = α_0
 # A_1 = α_1
@@ -42,8 +42,8 @@ initial_params = [mean(y) var(y)]
 ub = [1.0; 1.0; 0.5; 1.0]
 lb = [-1.0; 0.0; 0.0; 0.5]
 
-# Define a GAS model where only \sigma^2 is varying over time
-gas = GAS.Model(1, 1, Normal, 1.0, time_varying_params = [2])
+# Define a ScoreDrivenModels model where only \sigma^2 is varying over time
+gas = ScoreDrivenModels.Model(1, 1, Normal, 1.0, time_varying_params = [2])
 
 # Give an initial_point in the interior of the bounds.
 initial_point = [0.0; 0.5; 0.25; 0.75]

diff --git a/examples/inflow_lognormal.jl b/examples/inflow_lognormal.jl
@@ -1,4 +1,4 @@
-using GAS, Plots
+using ScoreDrivenModels, Plots
 
 # Some data on monthly inflow from Northeastern Brazil
 inflow = [
@@ -49,8 +49,8 @@ y = Vector{Float64}(vec(inflow'))
 y_train = y[1:400]
 y_test = y[401:end]
 
-# Specify GAS model: here we use lag 1 for trend characterization and lag 12 for seasonality characterization
-gas = GAS.Model([1, 2, 11, 12], [1, 2, 11, 12], LogNormal, 0.0; time_varying_params = [1])
+# Specify ScoreDrivenModels model: here we use lag 1 for trend characterization and lag 12 for seasonality characterization
+gas = ScoreDrivenModels.Model([1, 2, 11, 12], [1, 2, 11, 12], LogNormal, 0.0; time_varying_params = [1])
 
 # Define initial_params with
 initial_params = dynamic_initial_params(y_train, gas)
@@ -66,7 +66,7 @@ plot(y_train, label = "In-sample inflow")
 plot!(y_fitted, label = "in-sample estimates")
 
 # Forecasts with 95% confidence interval
-forec = GAS.forecast(y_train, gas, 80; initial_params = initial_params, ci = [0.95])
+forec = ScoreDrivenModels.forecast(y_train, gas, 80; initial_params = initial_params, ci = [0.95])
 
 plot(y_test, label = "Out-of-sample inflow")
 plot!(forec, label = "Forecast", color = "Steel Blue")
diff --git a/src/GAS.jl → src/ScoreDrivenModels.jl b/src/GAS.jl → src/ScoreDrivenModels.jl
@@ -1,4 +1,4 @@
-module GAS
+module ScoreDrivenModels
 
 using Distributions, Optim, SpecialFunctions
 using LinearAlgebra, Printf

diff --git a/src/gas/initial_params.jl b/src/gas/initial_params.jl
@@ -54,9 +54,9 @@ function dynamic_initial_params(obs::Vector{T}, gas::Model{D, T}) where {D, T}
         idx = collect(i:biggest_lag:len)
         # Fit MLE in the observations
         dist = fit_mle(D, obs[idx])
-        # Adequate to the GAS standard
+        # Adequate to the ScoreDrivenModels standard
         # In Distributions Normal is \mu and \sigma
-        # In GAS Normal is \mu and \sigma^2
+        # In ScoreDrivenModels Normal is \mu and \sigma^2
         sdm_dist = update_dist(D, permutedims([params(dist)...]), 1)
 
         initial_params[i, :] = [params(sdm_dist)...]

diff --git a/src/initial_params.jl b/src/initial_params.jl
@@ -54,9 +54,9 @@ function dynamic_initial_params(obs::Vector{T}, gas::Model{D, T}) where {D, T}
         idx = collect(i:biggest_lag:len)
         # Fit MLE in the observations
         dist = fit_mle(D, obs[idx])
-        # Adequate to the GAS standard
+        # Adequate to the ScoreDrivenModels standard
         # In Distributions Normal is \mu and \sigma
-        # In GAS Normal is \mu and \sigma^2
+        # In ScoreDrivenModels Normal is \mu and \sigma^2
         sdm_dist = update_dist(D, permutedims([params(dist)...]), 1)
 
         initial_params[i, :] = [params(sdm_dist)...]

diff --git a/src/model.jl b/src/model.jl
@@ -59,7 +59,7 @@ function number_of_lags(gas::Model)
 end
 
 """
-The unknows parameters of a GAS model
+The unknows parameters of a ScoreDrivenModels model
 Only for internal use
     
 Every unknonws must define

diff --git a/src/simulate.jl b/src/simulate.jl
@@ -3,7 +3,7 @@ export simulate, forecast
 """
     simulate(serie::Vector{T}, gas::Model{D, T}, N::Int, S::Int, kwargs...) where {D, T}
 
-Generate scenarios for the future of a time series by updating the GAS recursion `N` times and taking 
+Generate scenarios for the future of a time series by updating the ScoreDrivenModels recursion `N` times and taking 
 a sample of the distribution until generate `S` scenarios.
 """
 function simulate(serie::Vector{T}, gas::Model{D, T}, N::Int, S::Int;
@@ -28,7 +28,7 @@ end
 """
     forecast(serie::Vector{T}, gas::Model{D, T}, N::Int; kwargs...) where {D, T}
 
-Forecast future values of a time series by updating the GAS recursion `N` times and 
+Forecast future values of a time series by updating the ScoreDrivenModels recursion `N` times and 
 taking the mean of the distribution at each time. You can pass the desired confidence interval 
 as a `Vector{T}`. The forecast will be the first column and the confidence intervals are the remaining
 columns.

diff --git a/test/runtests.jl b/test/runtests.jl
@@ -1,12 +1,12 @@
-using GAS, Distributions, LinearAlgebra
+using ScoreDrivenModels, Distributions, LinearAlgebra
 
 using Test, Random, HypothesisTests, DelimitedFiles
 
 include("utils.jl")
 include("test_links.jl")
 include("test_distributions.jl")
 
-# GAS
+# ScoreDrivenModels
 include("test_recursion.jl")
 include("test_initial_params.jl")
 include("test_diagnostics.jl")

diff --git a/test/test_distributions.jl b/test/test_distributions.jl
@@ -1,13 +1,13 @@
 @testset "Score" begin
-    for dist in GAS.DISTS
+    for dist in ScoreDrivenModels.DISTS
         @testset "$dist" begin
             test_score_mean(dist)
         end
     end
 end
 
 @testset "Log-likelihood" begin
-    for dist in GAS.DISTS
+    for dist in ScoreDrivenModels.DISTS
         @testset "$dist" begin
             test_loglik(dist)
         end