Post Estimation Commands

compute_irfs(
  exovars,
  gx,
  hx,
  XSS,
  ids;
  T = 1000,
  init_val = fill(0.01, length(exovars)),
  verbose = true
  distribution = false
  comp_ids = nothing
)

Computes impulse response functions (IRFs) for a given set of shocks to exogenous variables.

Arguments

• exovars::Vector{Int64}: Vector of positional indices of exogenous variables to which shocks are applied.
• gx::Matrix{Float64}: Control matrix (states to controls equations)
• hx::Matrix{Float64}: Transition matrix for states (state transition equations)
• XSS::Vector{Float64}: Steady state values of the model variables.
• ids: Indexes of the model variables.

Keyword Arguments

• T::Int64: Number of periods for which to compute IRFs.
• init_val::Vector{Float64}: Initial value of the shock to each exogenous variable, defaults to 0.01 for all shocks.
• verbose::Bool: Print progress to console.
• distribution::Bool: Compute distributional IRFs, defaults to false.
• comp_ids: Compression indices for the distribution, as created by prepare_linearization. Needed if distribution is true. Defaults to nothing.

Returns

• IRFs, IRFs_lvl: 3D array of (level) IRFs for each exogenous variable, with dimensions:

  • 1: States and controls
  • 2: Time periods
  • 3: Exogenous variables

• IRFs_order: Names of the exogenous variables and their indices in the IRFs array.

• IRFs_dist: Dictionary containing the distributional IRFs in levels, if requested. Dimensions are the same as for IRFs and IRFs_lvl, with the following keys:

  • "Wb": Marginal value function of liquid assets, nb x T.
  • "Wk": Marginal value function of illiquid assets, nk x T.
  • "distr_b": Distribution of liquid assets, nb x T.
  • "distr_k": Distribution of illiquid assets, nk x T.
  • "distr": Joint distribution of liquid and illiquid assets, nb x nk x T.

plot_irfs(
    shocks_to_plot,
    vars_to_plot,
    IRFs_to_plot,
    IRFs_order,
    ids;
    horizon = 40,
    factor = 100,
    legend_on_all = false,
    show_fig = true,
    save_fig = false,
    save_fig_indiv = false,
    path = "",
    suffix = "",
    yscale = "standard",
    style_options = (lw = 2, color = :auto, linestyle = :solid),
)

Plots impulse response functions (IRFs) for specified shocks and variables, given a set of IRFs as produced by compute_irfs.

Arguments

• shocks_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, where each tuple contains a shock variable (as a Symbol) and its corresponding label (String).
• vars_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, each containing a variable to plot (as a Symbol) and its corresponding label (String).
• IRFs_to_plot::Vector{Tuple{Array{Float64,3},String}}: A vector of tuples, where each tuple consists of a 3D array of IRFs (Array{Float64,3}) and a string representing the specification name.
• IRFs_order::Vector{Symbol}: A vector of symbols specifying the order of shocks in the IRF arrays.
• ids: A structure mapping variable symbols to their corresponding indices in the IRF arrays, must be identical for all IRFs versions.

Keyword Arguments

• horizon::Int64: The time horizon (number of periods) over which IRFs are plotted. Default is 40.
• factor::Int64: Scaling factor for the IRFs (default: 100).
• legend_on_all::Bool: If true, includes the legend on all subplots; otherwise, only the first subplot has a legend. Default is false.
• show_fig::Bool: If true, displays the plot. Default is true.
• save_fig::Bool: If true, saves the combined plot as a PDF. Default is false.
• save_fig_indiv::Bool: If true, saves individual plots for each variable directly inside a shock-specific folder. Default is false.
• path::String: The directory path where the generated plots should be saved. Default is an empty string (no saving).
• suffix::String: A suffix to append to the saved plot filenames. Default is an empty string.
• yscale::Union{String, Tuple{Number,Number}, Dict{Symbol,Tuple{Number,Number}}}: Y-axis scaling specification. When set to "common", computes a common y-axis limit across all subplots from the data; when provided as a tuple, uses it as (ymin, ymax); when provided as a dictionary, applies specified y-axis limits for each variable. Default is "standard", which applies default scaling.
• style_options::NamedTuple: A named tuple specifying stylistic options for the plots, including line width (lw), color (default: :auto), and linestyle (default: :solid). Default is (lw = 2, color = :auto, linestyle = :solid).

plot_irfs_cat(
    shock_categories,
    vars_to_plot,
    IRFs_to_plot,
    IRFs_order,
    ids;
    horizon = 40,
    factor = 100,
    show_fig = true,
    save_fig = false,
    save_fig_indiv = false,
    path = "",
    suffix = "",
    yscale = "standard",
    style_options = (lw = 2, color = :auto, linestyle = :solid)
)

Plots impulse response functions (IRFs) for specified shocks and variables, given IRFs as produced by compute_irfs, organized by shock categories.

Arguments

• shock_categories::Dict{Tuple{String,String},Vector{Symbol}}: A dictionary where each key represents a category of shocks (with a label and a string for saving), and each value is a vector of symbols representing the shocks in that category.
• vars_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, each containing a variable to plot (as a Symbol) and its corresponding label (String).
• IRFs_to_plot::Array{Float64,3}: A 3D array of IRFs.
• IRFs_order::Vector{Symbol}: A vector of symbols specifying the order of shocks in the IRF arrays.
• ids: A structure mapping variable symbols to their corresponding indices in the IRF arrays, which must be identical for all IRFs versions.

Keyword Arguments

• horizon::Int64: The time horizon (number of periods) over which IRFs are plotted. Default is 40.
• factor::Int64: Scaling factor for the IRFs (default: 100).
• show_fig::Bool: If true, displays the plot. Default is true.
• save_fig::Bool: If true, saves the combined plot as a PDF. Default is false.
• save_fig_indiv::Bool: If true, saves individual plots for each variable or shock panel directly inside a category-specific folder. Default is false.
• path::String: The directory path where the generated plots should be saved. Default is an empty string (no saving).
• suffix::String: A suffix to append to the saved plot filenames. Default is an empty string.
• yscale::Union{String, Tuple{Number,Number}, Dict{Symbol,Tuple{Number,Number}}}: Y-axis scaling specification. When set to "common", computes a common y-axis limit across all subplots from the data; when provided as a tuple, uses it as (ymin, ymax); when provided as a dictionary, applies specified y-axis limits for each variable. Default is "standard", which applies default scaling.
• style_options::NamedTuple: A named tuple specifying stylistic options for the plots, including line width (lw), color (default: :auto), and linestyle (default: :solid). Default is (lw = 2, color = :auto, linestyle = :solid).

plot_distributional_irfs(
    shocks_to_plot,
    vars_to_plot,
    IRFs_to_plot,
    IRFs_order;
    horizon = 40,
    factor = 100,
    legend = false,
    show_fig = true,
    save_fig = false,
    path = "",
    suffix = ""
)

Plots impulse response functions (IRFs) for specified shocks and distributional variables, given a set of IRFs as produced by compute_irfs.

Arguments

• shocks_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, where each tuple contains a shock variable (as a Symbol) and its corresponding label (String).

• vars_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, each containing a variable to plot (as a Symbol) and its corresponding label (String).

• IRFs_to_plot::Dict{String,Array{Float64}}: A dictionary of distributional IRFs, where the keys are variable names and the values are arrays of IRFs.

• IRFs_order::Vector{Symbol}: A vector of symbols specifying the order of shocks in the IRF arrays.

Keyword Arguments

• horizon::Int64: The time horizon (number of periods) over which IRFs are plotted. Default is 40.
• factor::Int64: Scaling factor for the IRFs (default: 100).
• legend_on_all::Bool: If true, includes the legend on all subplots. Otherwise, only
• show_fig::Bool: If true, displays the plot (default: true). the first subplot has a legend (default: false).
• save_fig::Bool: If true, saves the plot as a PDF. Default is false.
• path::String: The directory path where the generated plots should be saved. Default is an empty string (no saving).
• suffix::String: A suffix to append to the saved plot filenames. Default is an empty string.

compute_vardecomp(IRFs)

Computes the variance decomposition of the impulse response functions (IRFs) of a model.

Arguments

• IRFs::Array{Float64}: 3D array of IRFs for each exogenous variable, with dimensions, returned from compute_irfs:

  • 1: States and controls
  • 2: Time periods
  • 3: Exogenous variables

Returns

• VDs: 3D array of variance decompositions for each exogenous variable, with dimensions:

  • 1: States and controls
  • 2: Time periods
  • 3: Exogenous variables

• exovars_names: Names of the exogenous variables.

compute_vardecomp_bcfreq(exovars_full, stds, gx, hx; passband = (6, 32), ngrid = 512)

This function is designed to produce a variance decomposition at business cycle frequencies. It produces a variance decomposition of the linearized solution. It returns the variance decomposition at business cycle frequencies based on Uhlig (2001) and the unconditional variance.

Arguments

• exovars_full::Vector{Int64}: Vector of positional indices of all exogenous variables to which shocks; note: here, no subset of exogenous variables is allowed.
• stds::Vector{Float64}: Vector of standard deviations of the shocks.
• gx::Matrix{Float64}: Control matrix (states to controls equations)
• hx::Matrix{Float64}: Transition matrix for states (state transition equations)

Keyword Arguments

• passband::Tuple{Int64,Int64}: A tuple specifying the horizons associated with the business cycle. Default is (6, 32).
• ngrid::Int64: The number of grid points for the computation of the band pass filter. Default is 512.

Returns

• VD: Variance decomposition at business cycle frequencies.
• UnconditionalVar: Unconditional variance.

plot_vardecomp(
    vars_to_plot,
    VDs_to_plot,
    VDs_order,
    ids;
    shock_categories = Dict(),
    VD_horizons = [4, 16, 100],
    colorlist = [...],
    show_fig = true,
    save_fig = false,
    path = "",
    suffix = "",
    factor = 100.0,
)

Plots variance decompositions (VDs) for selected variables across multiple models and forecast horizons, given a set of VDs as produced by compute_vardecomp, potentially organized by shock categories.

Arguments

• vars_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, where each tuple consists of a variable to plot (as a Symbol) and its corresponding label (String).
• VDs_to_plot::Vector{Tuple{Array{Float64,3},String}}: A vector of tuples, where each tuple consists of a 3D array of VDs (Array{Float64,3}) and a string representing the specification name.
• VDs_order::Vector{Symbol}: A vector of symbols specifying the order of shocks in the VD arrays.
• ids: A structure mapping variable symbols to their corresponding indices in the VD arrays, must be identical for all VDs versions.

Keyword Arguments

• shock_categories::Dict{Tuple{String,String},Vector{Symbol}}: A dictionary where each key represents a category of shocks (with a label and a string for saving), and each value is a vector of symbols representing the shocks in that category.
• VD_horizons::Vector{Int64}: Forecast horizons to consider. Default is [4, 16, 100].
• colorlist::Vector: A list of colors used to distinguish different shocks in the stacked bar plots. Default includes predefined hex colors and symbols.
• show_fig::Bool: If true, displays the plot. Default is true.
• save_fig::Bool: If true, saves the plot as a PDF. Default is false.
• path::String: The directory path where the generated plots should be saved. Default is an empty string (no saving).
• suffix::String: A suffix to append to the saved plot filenames. Default is an empty string.
• factor::Float64 = 100.0: Scaling factor for variance decomposition values.

plot_vardecomp_bcfreq(
    vars_to_plot,
    VDbcs_to_plot,
    VDbcs_order,
    ids;
    shock_categories = Dict(),
    colorlist = [...],
    show_fig = true,
    save_fig = false,
    path = "",
    suffix = "",
    factor = 100.0,
)

Plots busincess cycle frequency variance decompositions (VDbcs) for selected variables across multiple models and forecast horizons, given a set of VDbcs as produced by compute_vardecomp_bcfreq, potentially organized by shock categories.

Arguments

• vars_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, where each tuple consists of a variable to plot (as a Symbol) and its corresponding label (String).
• VDbcs_to_plot::Vector{Tuple{Matrix{Float64},String}}: A vector of tuples, where each tuple consists of a matrix of VDbcs (Matrix{Float64}) and a string representing the specification name.
• VDbcs_order::Vector{Symbol}: A vector of symbols specifying the order of shocks in the VDbc arrays.
• ids: A structure mapping variable symbols to their corresponding indices in the VD arrays, must be identical for all VDbcs versions.

Keyword Arguments

• shock_categories::Dict{Tuple{String,String},Vector{Symbol}}: A dictionary where each key represents a category of shocks (with a label and a string for saving), and each value is a vector of symbols representing the shocks in that category.
• colorlist::Vector: A list of colors used to distinguish different shocks in the stacked bar plots. Default includes predefined hex colors and symbols.
• show_fig::Bool: If true, displays the plot. Default is true.
• save_fig::Bool: If true, saves the plot as a PDF. Default is false.
• path::String: The directory path where the generated plots should be saved. Default is an empty string (no saving).
• suffix::String: A suffix to append to the saved plot filenames. Default is an empty string.
• factor::Float64 = 100.0: Scaling factor for variance decomposition values.

compute_hist_decomp(exovars_full, gx, hx, smoother_output, ids)

Computes the historical decomposition of the model's variables in response to a set of shocks, including the effect of initial conditions and each shock.

Arguments

• exovars_full::Vector{Int64}: Vector of positional indices of all exogenous variables to which shocks; note: here, no subset of exogenous variables is allowed.
• gx::Matrix{Float64}: Control matrix (states to controls equations)
• hx::Matrix{Float64}: Transition matrix for states (state transition equations)
• smoother_output: The output from a smoother function.
• ids: Indexes of the model variables.

Returns

• ShockContr::Array{Float64}: A 3D array representing the decomposition of the effects of each shock and the initial condition on states and controls over time. The dimensions are:

  • 1: States and controls
  • 2: Time periods
  • 3: Shocks (including an additional index for the initial condition)

• ShockContr_order::Vector{Any}: A vector containing the names or identifiers of the shocks and the initial condition, in the order they appear in the ShockContr array.

plot_hist_decomp(
    vars_to_plot,
    HDs_to_plot,
    HDs_order,
    ids;
    shock_categories = Dict(),
    timeline = collect(1:size(HDs_to_plot, 2)),
    colorlist = [...],
    show_fig = true,
    save_fig = false,
    path = "",
    suffix = ""
)

Plots historical decompositions (HDs) for specified variables, showing the contribution of different shocks over time, given a set of HDs as produced by compute_hist_decomp, potentially organized by shock categories.

Arguments

• vars_to_plot::Vector{Tuple{Symbol,String}}: A vector of tuples, where each tuple consists of a variable to plot (as a Symbol) and its corresponding label (String).
• HDs_to_plot::Array{Float64,3}: A 3D array containing historical decompositions, where dimensions correspond to variables, time periods, and shocks.
• HDs_order::Vector{Symbol}: A vector of symbols specifying the order of shocks in the HDs array.
• ids: A structure mapping variable symbols to their corresponding indices in the HDs array, ensuring consistency across all plotted variables.

Keyword Arguments

• shock_categories::Dict{Tuple{String,String},Vector{Symbol}}: A dictionary where each key represents a category of shocks (with a label and a string for saving), and each value is a vector of symbols representing the shocks in that category.
• timeline::Vector: A vector specifying the time axis for the plots. Default is a sequence from 1 to the number of time periods in HDs_to_plot.
• colorlist::Vector: A list of colors used to distinguish different shocks in the stacked bar plots. Default includes predefined hex colors and symbols.
• show_fig::Bool: If true, displays the plot. Default is true.
• save_fig::Bool: If true, saves the plot as a PDF. Default is false.
• path::String: The directory path where the generated plots should be saved. Default is an empty string (no saving).
• suffix::String: A suffix to append to the saved plot filenames. Default is an empty string.