Skip to content

blazer (Model)

Analyze sequential block structure of steady equations

Syntax

[nameBlk, eqtnBlk, blkType, blazerObj] = blazer(model, ...)

Input arguments

model [ Model ]

Model object

Output arguments

nameBlk [ cell ]

Lists of variables that each individual block will be solved for; the nameBlk{i}.Level element is a string array with the names of the variables whose levels will be solved for in the i-th block; the nameBlk{i}.Change element is a string array with the names of the variables whose changes (differences or rates of growth) will be solved for in the i-th block.

eqtnBlk [ cell ]

List of equations in each block.

blkType [ solver.block.Type ]

Type of each block: SOLVE or ASSIGN.

blazerObj [ blazer.Blazer ]

Blazer object.

Options

Endogenize={ } [ cellstr | char | string | empty ]

List of parameters that will be endogenized in steady equations.

Exogenize={ } [ cellstr | char | empty | string ] 

List of transition or measurement variables that will be exogenized
in steady equations.

Kind='Steady' [ 'Current' | 'Stacked' | 'Steady' ]

The method of sequential block analysis that will be performed.

Description

Three ways the sequential block analysis can be performed:

  • 'Steady' Investigate steady-state equations, considering lags and leads to be the same entity as the respective current dated variable.

  • 'Current' Investigate the current dated variables in dynamic equations, taking lags and leads as given.

  • 'Stacked' Investigate a whole structure of time-stacked equations (not available yet).

Reordering Algorithm

The reordering algorithm first identifies equations with a single variable in each, and variables occurring in a single equation each, and then uses a combination of column and row approximate minimum degree permutations (colamd) followed by a Dulmage-Mendelsohn permutation (dmperm).

Output Returned from Blazer

The output arguments NameBlk and EqtnBlk are 1-by-N cell arrays, where N is the number of blocks, and each cell is a 1-by-Kn cell array of strings, where Kn is the number of variables and equations in block N.

Examples