# empiricalbvarm

Bayesian vector autoregression (VAR) model with samples from prior or posterior distribution

## Description

The Bayesian VAR model object `empiricalbvarm`

contains samples from the distributions of the coefficients Λ and innovations covariance matrix Σ of a VAR(*p*) model, which MATLAB^{®} uses to characterize the corresponding prior or posterior distributions.

For Bayesian VAR model objects that have an intractable posterior, the `estimate`

function returns an `empiricalbvarm`

object representing the empirical posterior distribution. However, if you have random draws from the prior or posterior distributions of the coefficients and innovations covariance matrix, you can create a Bayesian VAR model with an empirical prior directly by using `empiricalbvarm`

.

## Creation

### Syntax

### Description

creates a `Mdl`

= empiricalbvarm(`numseries`

,`numlags`

,'CoeffDraws',`CoeffDraws`

,'SigmaDraws',`SigmaDraws`

)`numseries`

-D Bayesian VAR(`numlags`

) model object `Mdl`

characterized by the random samples from the prior or posterior distributions of $$\lambda =\text{vec}\left(\Lambda \right)=\text{vec}\left({\left[\begin{array}{ccccccc}{\Phi}_{1}& {\Phi}_{2}& \cdots & {\Phi}_{p}& c& \delta & {\rm B}\end{array}\right]}^{\prime}\right)$$ and Σ, `CoeffDraws`

and `SigmaDraws`

, respectively.

`numseries`

=*m*, a positive integer specifying the number of response time series variables.`numlags`

=*p*, a nonnegative integer specifying the AR polynomial order (that is, number of`numseries`

-by-`numseries`

AR coefficient matrices in the VAR model).

sets writable properties (except `Mdl`

= empiricalbvarm(`numseries`

,`numlags`

,'`CoeffDraws`

',CoeffDraws,'`SigmaDraws`

',SigmaDraws,`Name,Value`

)`NumSeries`

and `P`

) using name-value pair arguments. Enclose each property name in quotes. For example, `empiricalbvarm(3,2,'CoeffDraws',CoeffDraws,'SigmaDraws',SigmaDraws,'SeriesNames',["UnemploymentRate" "CPI" "FEDFUNDS"])`

specifies the random samples from the distributions of λ and Σ and the names of the three response variables.

Because the posterior distributions of a semiconjugate prior model (`semiconjugatebvarm`

) are analytically intractable, `estimate`

returns an `empiricalbvarm`

object that characterizes the posteriors and contains the Gibbs sampler draws from the full conditionals.

### Input Arguments

`numseries`

— Number of time series *m*

`1`

(default) | positive integer

Number of time series *m*, specified as a positive integer. `numseries`

specifies the dimensionality of the multivariate response variable *y _{t}* and innovation

*ε*.

_{t}`numseries`

sets the `NumSeries`

property.

**Data Types: **`double`

`numlags`

— Number of lagged responses

nonnegative integer

Number of lagged responses in each equation of *y*_{t}, specified as a nonnegative integer. The resulting model is a VAR(`numlags`

) model; each lag has a `numseries`

-by-`numseries`

coefficient matrix.

`numlags`

sets the `P`

property.

**Data Types: **`double`

## Properties

You can set writable property values when you create the model object by using name-value pair argument syntax, or after you create the model object by using dot notation. For example, to create a 3-D Bayesian VAR(1) model from the coefficient and innovations covariance arrays of draws `CoeffDraws`

and `SigmaDraws`

, respectively, and then label the response variables, enter:

Mdl = empiricalbvarm(3,1,'CoeffDraws',CoeffDraws,'SigmaDraws',SigmaDraws); Mdl.SeriesNames = ["UnemploymentRate" "CPI" "FEDFUNDS"];

## Required Draws from Distribution

`CoeffDraws`

— Random sample from prior or posterior distribution of *λ*

numeric matrix

Random sample from the prior or posterior distribution of *λ*, specified as a `NumSeries*`

-by-`k`

`numdraws`

numeric matrix, where

(the number of coefficients in a response equation). * k* = NumSeries*P + IncludeIntercept + IncludeTrend + NumPredictors

`CoeffDraws`

represents the empirical distribution of *λ*based on a size

`numdraws`

sample.Columns correspond to successive draws from the distribution. `CoeffDraws(1:`

corresponds to all coefficients in the equation of response variable * k*,:)

`SeriesNames(1)`

, `CoeffDraws((``k`

+ 1):(2*`k`

),:)

corresponds to all coefficients in the equation of response variable `SeriesNames(2)`

, and so on. For a set of row indices corresponding to an equation:Elements

`1`

through`NumSeries`

correspond to the lag 1 AR coefficients of the response variables ordered by`SeriesNames`

.Elements

`NumSeries + 1`

through`2*NumSeries`

correspond to the lag 2 AR coefficients of the response variables ordered by`SeriesNames`

.In general, elements

`(`

through– 1)*NumSeries + 1`q`

correspond to the lag*NumSeries`q`

AR coefficients of the response variables ordered by`q`

`SeriesNames`

.If

`IncludeConstant`

is`true`

, element`NumSeries*P + 1`

is the model constant.If

`IncludeTrend`

is`true`

, element`NumSeries*P + 2`

is the linear time trend coefficient.If

`NumPredictors`

> 0, elements`NumSeries*P + 3`

through

constitute the vector of regression coefficients of the exogenous variables.`k`

This figure shows the row structure of `CoeffDraws`

for a 2-D VAR(3) model that contains a constant vector and four exogenous predictors:

$$[\stackrel{{y}_{1,t}}{\overbrace{\begin{array}{ccccccccccc}{\varphi}_{1,11}& {\varphi}_{1,12}& {\varphi}_{2,11}& {\varphi}_{2,12}& {\varphi}_{3,11}& {\varphi}_{3,12}& {c}_{1}& {\beta}_{11}& {\beta}_{12}& {\beta}_{13}& {\beta}_{14}\end{array}}}\text{\hspace{0.17em}}\text{\hspace{0.17em}}\stackrel{{y}_{2,t}}{\overbrace{\begin{array}{ccccccccccc}{\varphi}_{1,21}& {\varphi}_{1,22}& {\varphi}_{2,21}& {\varphi}_{2,22}& {\varphi}_{3,21}& {\varphi}_{3,22}& {c}_{2}& {\beta}_{21}& {\beta}_{22}& {\beta}_{23}& {\beta}_{24}\end{array}}}],$$

where

*ϕ*_{q,jk}is element (*j*,*k*) of the lag*q*AR coefficient matrix.*c*_{j}is the model constant in the equation of response variable*j*.*B*_{ju}is the regression coefficient of the exogenous variable*u*in the equation of response variable*j*.

`CoeffDraws`

and `SigmaDraws`

must be based on the same number of draws, and both must represent draws from either the prior or posterior distribution.

`numdraws`

should be reasonably large, for example, `1e6`

.

**Data Types: **`double`

`SigmaDraws`

— Random sample from prior or posterior distribution of Σ

array of positive definite numeric matrices

Random sample from the prior or posterior distribution of Σ, specified as a `NumSeries`

-by-`NumSeries`

-by-`numdraws`

array of positive definite numeric matrices. `SigmaDraws`

represents the empirical distribution of Σ based on a size `numdraws`

sample.

Rows and columns correspond to innovations in the equations of the response variables ordered by `SeriesNames`

. Columns correspond to successive draws from the distribution.

`CoeffDraws`

and `SigmaDraws`

must be based on the same number of draws, and both must represent draws from either the prior or posterior distribution.

`numdraws`

should be reasonably large, for example, `1e6`

.

**Data Types: **`double`

## Model Characteristics and Dimensionality

`Description`

— Model description

string scalar | character vector

Model description, specified as a string scalar or character vector. The default value describes the model dimensionality, for example `'2-Dimensional VAR(3) Model'`

.

**Example: **`"Model 1"`

**Data Types: **`string`

| `char`

`NumSeries`

— Number of time series *m*

positive integer

This property is read-only.

Number of time series *m*, specified as a positive integer. `NumSeries`

specifies the dimensionality of the multivariate response variable *y _{t}* and innovation

*ε*.

_{t}**Data Types: **`double`

`P`

— Multivariate autoregressive polynomial order

nonnegative integer

This property is read-only.

Multivariate autoregressive polynomial order, specified as a nonnegative integer. `P`

is the maximum lag that has a nonzero coefficient matrix.

`P`

specifies the number of presample observations required to initialize the model.

**Data Types: **`double`

`SeriesNames`

— Response series names

string vector | cell array of character vectors

Response series names, specified as a `NumSeries`

length string vector. The default is `['Y1' 'Y2' ... 'Y`

. * NumSeries*']

`empiricalbvarm`

stores `SeriesNames`

as a string vector.**Example: **`["UnemploymentRate" "CPI" "FEDFUNDS"]`

**Data Types: **`string`

`IncludeConstant`

— Flag for including model constant *c*

`true`

(default) | `false`

Flag for including a model constant *c*, specified as a value in this table.

Value | Description |
---|---|

`false` | Response equations do not include a model constant. |

`true` | All response equations contain a model constant. |

**Data Types: **`logical`

`IncludeTrend`

— Flag for including linear time trend term *δt*

`false`

(default) | `true`

Flag for including a linear time trend term *δt*, specified as a value in this table.

Value | Description |
---|---|

`false` | Response equations do not include a linear time trend term. |

`true` | All response equations contain a linear time trend term. |

**Data Types: **`logical`

`NumPredictors`

— Number of exogenous predictor variables in model regression component

`0`

(default) | nonnegative integer

Number of exogenous predictor variables in the model regression component, specified as a nonnegative integer. `empiricalbvarm`

includes all predictor variables symmetrically in each response equation.

## VAR Model Parameters Derived from Distribution Draws

`AR`

— Distribution mean of autoregressive coefficient matrices Φ_{1},…,Φ_{p}

cell vector of numeric matrices

This property is read-only.

Distribution mean of the autoregressive coefficient matrices Φ_{1},…,Φ_{p} associated with the lagged responses, specified as a `P`

-D cell vector of `NumSeries`

-by-`NumSeries`

numeric matrices.

`AR{`

is Φ* j*}

_{j}, the coefficient matrix of lag

*. Rows correspond to equations and columns correspond to lagged response variables;*

`j`

`SeriesNames`

determines the order of response variables and equations. Coefficient signs are those of the VAR model expressed in difference-equation notation.If `P`

= 0, `AR`

is an empty cell. Otherwise, `AR`

is the collection of AR coefficient means extracted from `Mu`

.

**Data Types: **`cell`

`Constant`

— Distribution mean of model constant *c*

numeric vector

This property is read-only.

Distribution mean of the model constant *c* (or intercept), specified as a `NumSeries`

-by-1 numeric vector. `Constant(`

is the constant in equation * j*)

*;*

`j`

`SeriesNames`

determines the order of equations.If `IncludeConstant`

= `false`

, `Constant`

is an empty array. Otherwise, `Constant`

is the model constant vector mean extracted from `Mu`

.

**Data Types: **`double`

`Trend`

— Distribution mean of linear time trend *δ*

numeric vector

This property is read-only.

Distribution mean of the linear time trend *δ*, specified as a `NumSeries`

-by-1 numeric vector. `Trend(`

is the linear time trend in equation * j*)

*;*

`j`

`SeriesNames`

determines the order of equations.If `IncludeTrend`

= `false`

(the default), `Trend`

is an empty array. Otherwise, `Trend`

is the linear time trend coefficient mean extracted from `Mu`

.

**Data Types: **`double`

`Beta`

— Distribution mean of regression coefficient matrix Β

numeric matrix

This property is read-only.

Distribution mean of the regression coefficient matrix B associated with the exogenous predictor variables, specified as a `NumSeries`

-by-`NumPredictors`

numeric matrix.

`Beta(`

contains the regression coefficients of each predictor in the equation of response variable * j*,:)

*j*

*y*.

_{j,t}`Beta(:,``k`

)

contains the regression coefficient in each equation of predictor *x*. By default, all predictor variables are in the regression component of all response equations. You can down-weight a predictor from an equation by specifying, for the corresponding coefficient, a prior mean of 0 in

_{k}`Mu`

and a small variance in `V`

.When you create a model, the predictor variables are hypothetical. You specify predictor data when you operate on the model (for example, when you estimate the posterior by using `estimate`

). Columns of the predictor data determine the order of the columns of `Beta`

.

**Data Types: **`double`

`Covariance`

— Distribution mean of innovations covariance matrix Σ

positive definite numeric matrix

This property is read-only.

Distribution mean of the innovations covariance matrix Σ of the `NumSeries`

innovations at each time *t* = 1,...,*T*, specified as a `NumSeries`

-by-`NumSeries`

positive definite numeric matrix. Rows and columns correspond to innovations in the equations of the response variables ordered by `SeriesNames`

.

**Data Types: **`double`

## Object Functions

`summarize` | Distribution summary statistics of Bayesian vector autoregression (VAR) model |

## Examples

### Create Empirical Model

Consider the 3-D VAR(4) model for the US inflation (`INFL`

), unemployment (`UNRATE`

), and federal funds (`FEDFUNDS`

) rates.

$$\left[\begin{array}{l}{\text{INFL}}_{t}\\ {\text{UNRATE}}_{t}\\ {\text{FEDFUNDS}}_{t}\end{array}\right]=c+\sum _{j=1}^{4}{\Phi}_{j}\left[\begin{array}{l}{\text{INFL}}_{t-j}\\ {\text{UNRATE}}_{t-j}\\ {\text{FEDFUNDS}}_{t-j}\end{array}\right]+\left[\begin{array}{c}{\epsilon}_{1,t}\\ {\epsilon}_{2,t}\\ {\epsilon}_{3,t}\end{array}\right].$$

For all $$t$$, $${\epsilon}_{t}$$ is a series of independent 3-D normal innovations with a mean of 0 and covariance $\Sigma $.

You can create an empirical Bayesian VAR model for the coefficients ${\left[{\Phi}_{1},...,{\Phi}_{4},\mathit{c}\right]}^{\prime}$ and innovations covariance matrix $\Sigma $ in two ways:

Indirectly create an

`empiricalbvarm`

model by estimating the posterior distribution of a semiconjugate prior model.Directly create an

`empiricalbvarm`

model by supplying draws from the prior or posterior distribution of the parameters.

**Indirect Creation**

Assume the following prior distributions:

$\mathrm{vec}\left({\left[{\Phi}_{1},...,{\Phi}_{4},\mathit{c}\right]}^{\prime}\right)|\Sigma \sim {{\rm N}}_{39}\left(\mu ,\mathit{V}\right)$, where $\mu $ is a 39-by-1 vector of means and $\mathit{V}$ is the 39-by-39 covariance matrix.

$$\Sigma \sim Inverse\phantom{\rule{0.16666666666666666em}{0ex}}Wishart(\Omega ,\nu )$$, where $\Omega $ is the 3-by-3 scale matrix and $\nu $ is the degrees of freedom.

Create a semiconjugate prior model for the 3-D VAR(4) model parameters.

numseries = 3; numlags = 4; PriorMdl = semiconjugatebvarm(numseries,numlags)

PriorMdl = semiconjugatebvarm with properties: Description: "3-Dimensional VAR(4) Model" NumSeries: 3 P: 4 SeriesNames: ["Y1" "Y2" "Y3"] IncludeConstant: 1 IncludeTrend: 0 NumPredictors: 0 Mu: [39x1 double] V: [39x39 double] Omega: [3x3 double] DoF: 13 AR: {[3x3 double] [3x3 double] [3x3 double] [3x3 double]} Constant: [3x1 double] Trend: [3x0 double] Beta: [3x0 double] Covariance: [3x3 double]

`PriorMdl`

is a `semiconjugatebvarm`

Bayesian VAR model object representing the prior distribution of the coefficients and innovations covariance of the 3-D VAR(4) model.

Load the US macroeconomic data set. Compute the inflation rate. Plot all response series.

load Data_USEconModel seriesnames = ["INFL" "UNRATE" "FEDFUNDS"]; DataTimeTable.INFL = 100*[NaN; price2ret(DataTimeTable.CPIAUCSL)]; figure plot(DataTimeTable.Time,DataTimeTable{:,seriesnames}) legend(seriesnames)

Stabilize the unemployment and federal funds rates by applying the first difference to each series.

```
DataTimeTable.DUNRATE = [NaN; diff(DataTimeTable.UNRATE)];
DataTimeTable.DFEDFUNDS = [NaN; diff(DataTimeTable.FEDFUNDS)];
seriesnames(2:3) = "D" + seriesnames(2:3);
```

Remove all missing values from the data.

rmDataTimeTable = rmmissing(DataTimeTable);

Estimate the posterior distribution by passing the prior model and entire data series to `estimate`

.

rng(1); % For reproducibility PosteriorMdl = estimate(PriorMdl,rmDataTimeTable{:,seriesnames},'Display','off')

PosteriorMdl = empiricalbvarm with properties: Description: "3-Dimensional VAR(4) Model" NumSeries: 3 P: 4 SeriesNames: ["Y1" "Y2" "Y3"] IncludeConstant: 1 IncludeTrend: 0 NumPredictors: 0 CoeffDraws: [39x10000 double] SigmaDraws: [3x3x10000 double] AR: {[3x3 double] [3x3 double] [3x3 double] [3x3 double]} Constant: [3x1 double] Trend: [3x0 double] Beta: [3x0 double] Covariance: [3x3 double]

`PosteriorMdl`

is an `empiricalbvarm`

model representing the empirical posterior distribution of the coefficients and innovations covariance matrix. `empiricalbvarm`

stores the draws from the posteriors of $\lambda \text{\hspace{0.17em}}$ and $\Sigma \text{\hspace{0.17em}}$ in the `CoeffDraws`

and `SigmaDraws`

properties, respectively.

**Direct Creation**

Draw a random sample of size 1000 from the prior distribution `PriorMdl`

.

```
numdraws = 1000;
[CoeffDraws,SigmaDraws] = simulate(PriorMdl,'NumDraws',numdraws);
size(CoeffDraws)
```

`ans = `*1×2*
39 1000

size(SigmaDraws)

`ans = `*1×3*
3 3 1000

Create a Bayesian VAR model characterizing the empirical prior distributions of the parameters.

PriorMdlEmp = empiricalbvarm(numseries,numlags,'CoeffDraws',CoeffDraws,... 'SigmaDraws',SigmaDraws)

PriorMdlEmp = empiricalbvarm with properties: Description: "3-Dimensional VAR(4) Model" NumSeries: 3 P: 4 SeriesNames: ["Y1" "Y2" "Y3"] IncludeConstant: 1 IncludeTrend: 0 NumPredictors: 0 CoeffDraws: [39x1000 double] SigmaDraws: [3x3x1000 double] AR: {[3x3 double] [3x3 double] [3x3 double] [3x3 double]} Constant: [3x1 double] Trend: [3x0 double] Beta: [3x0 double] Covariance: [3x3 double]

Display the prior covariance mean matrices of the four AR coefficients by setting each matrix in the cell to a variable.

AR1 = PriorMdlEmp.AR{1}

`AR1 = `*3×3*
-0.0198 0.0181 -0.0273
-0.0207 -0.0301 -0.0070
-0.0009 0.0638 0.0113

AR2 = PriorMdlEmp.AR{2}

`AR2 = `*3×3*
-0.0453 0.0371 0.0110
-0.0103 -0.0304 -0.0011
0.0277 -0.0253 0.0061

AR3 = PriorMdlEmp.AR{3}

`AR3 = `*3×3*
0.0368 -0.0059 0.0018
-0.0306 -0.0106 0.0179
-0.0314 -0.0276 0.0116

AR4 = PriorMdlEmp.AR{4}

`AR4 = `*3×3*
0.0159 0.0406 -0.0315
-0.0178 0.0415 -0.0024
0.0476 -0.0128 -0.0165

## More About

### Bayesian Vector Autoregression (VAR) Model

A *Bayesian VAR model* treats all coefficients and the innovations covariance matrix as random variables in the *m*-dimensional, stationary VARX(*p*) model. The model has one of the three forms described in this table.

Model | Equation |
---|---|

Reduced-form VAR(p) in difference-equation notation |
$${y}_{t}={\Phi}_{1}{y}_{t-1}+\mathrm{...}+{\Phi}_{p}{y}_{t-p}+c+\delta t+{\rm B}{x}_{t}+{\epsilon}_{t}.$$ |

Multivariate regression |
$${y}_{t}={Z}_{t}\lambda +{\epsilon}_{t}.$$ |

Matrix regression |
$${y}_{t}={\Lambda}^{\prime}{z}_{t}^{\prime}+{\epsilon}_{t}.$$ |

For each time *t* = 1,...,*T*:

*y*is the_{t}*m*-dimensional observed response vector, where*m*=`numseries`

.Φ

_{1},…,Φ_{p}are the*m*-by-*m*AR coefficient matrices of lags 1 through*p*, where*p*=`numlags`

.*c*is the*m*-by-1 vector of model constants if`IncludeConstant`

is`true`

.*δ*is the*m*-by-1 vector of linear time trend coefficients if`IncludeTrend`

is`true`

.Β is the

*m*-by-*r*matrix of regression coefficients of the*r*-by-1 vector of observed exogenous predictors*x*_{t}, where*r*=`NumPredictors`

. All predictor variables appear in each equation.$${z}_{t}=\left[\begin{array}{ccccccc}{y}_{t-1}^{\prime}& {y}_{t-2}^{\prime}& \cdots & {y}_{t-p}^{\prime}& 1& t& {x}_{t}^{\prime}\end{array}\right],$$ which is a 1-by-(

*mp*+*r*+ 2) vector, and*Z*_{t}is the*m*-by-*m*(*mp*+*r*+ 2) block diagonal matrix$$\left[\begin{array}{cccc}{z}_{t}& {0}_{z}& \cdots & {0}_{z}\\ {0}_{z}& {z}_{t}& \cdots & {0}_{z}\\ \vdots & \vdots & \ddots & \vdots \\ {0}_{z}& {0}_{z}& {0}_{z}& {z}_{t}\end{array}\right],$$

where 0

_{z}is a 1-by-(*mp*+*r*+ 2) vector of zeros.$$\Lambda ={\left[\begin{array}{ccccccc}{\Phi}_{1}& {\Phi}_{2}& \cdots & {\Phi}_{p}& c& \delta & {\rm B}\end{array}\right]}^{\prime}$$, which is an (

*mp*+*r*+ 2)-by-*m*random matrix of the coefficients, and the*m*(*mp*+*r*+ 2)-by-1 vector*λ*= vec(Λ).*ε*is an_{t}*m*-by-1 vector of random, serially uncorrelated, multivariate normal innovations with the zero vector for the mean and the*m*-by-*m*matrix Σ for the covariance. This assumption implies that the data likelihood is$$\ell \left(\Lambda ,\Sigma |y,x\right)={\displaystyle \prod _{t=1}^{T}f\left({y}_{t};\Lambda ,\Sigma ,{z}_{t}\right)},$$

where

*f*is the*m*-dimensional multivariate normal density with mean*z*_{t}Λ and covariance Σ, evaluated at*y*_{t}.

Before considering the data, you impose a *joint prior distribution* assumption on (Λ,Σ), which is governed by the distribution *π*(Λ,Σ). In a Bayesian analysis, the distribution of the parameters is updated with information about the parameters obtained from the data likelihood. The result is the *joint posterior distribution*
*π*(Λ,Σ|*Y*,*X*,*Y*_{0}), where:

*Y*is a*T*-by-*m*matrix containing the entire response series {*y*_{t}},*t*= 1,…,*T*.*X*is a*T*-by-*m*matrix containing the entire exogenous series {*x*_{t}},*t*= 1,…,*T*.*Y*_{0}is a*p*-by-*m*matrix of presample data used to initialize the VAR model for estimation.

## Version History

**Introduced in R2020a**

## Open Example

You have a modified version of this example. Do you want to open this example with your edits?

## MATLAB Command

You clicked a link that corresponds to this MATLAB command:

Run the command by entering it in the MATLAB Command Window. Web browsers do not support MATLAB commands.

# Select a Web Site

Choose a web site to get translated content where available and see local events and offers. Based on your location, we recommend that you select: .

You can also select a web site from the following list:

## How to Get Best Site Performance

Select the China site (in Chinese or English) for best site performance. Other MathWorks country sites are not optimized for visits from your location.

### Americas

- América Latina (Español)
- Canada (English)
- United States (English)

### Europe

- Belgium (English)
- Denmark (English)
- Deutschland (Deutsch)
- España (Español)
- Finland (English)
- France (Français)
- Ireland (English)
- Italia (Italiano)
- Luxembourg (English)

- Netherlands (English)
- Norway (English)
- Österreich (Deutsch)
- Portugal (English)
- Sweden (English)
- Switzerland
- United Kingdom (English)