VoseTimeMultiAR2 | Vose Software

VoseTimeMultiAR2

See also: Time series in ModelRisk, Vose Multivariate Time Series window

VoseTimeMultiAR2({Means}, {Phi1}, {Phi2}, {CovMatrix}, {Period 0 log returns}, {Period -1 log returns}, Log Return, {Initial Values}, Data_in_rows)

 

 

 

Array function that simulates from a Multivariate Auto-Regressive time series model of order 2.

  • {Means} - array of mean log returns per period for each variable.

  • {Phi1} - matrix of first period autoregressive parameters.

  • {Phi2} - matrix of second period autoregressive parameters.

  • {CovMatrix} - covariance matrix of log returns.

  • {Period 0 log returns} - log returns at period zero.

  • {Period -1 log returns} - log returns at period -1.

  • Log Return - an optional parameter. Function generates log returns if set to TRUE, or variable values if set to FALSE or omitted.

  • {Initial Values} - array of starting values (at time zero) for each variable.

  • Data_in_rows - optional parameter that specifies if the data is in rows (TRUE) or columns (FALSE, default).

Equations

where:

k - number of variables

- k x 1 random vector, if   is the value of the variable at time t, then is the log return defined as

- k x 1 vector of means

- k x k autoregressive coefficient matrix, i=1,2,

- k x 1 vector of uncorrelated random variables, which is defined as follows:

 

VoseFunctions for this time series

VoseTimeMultiAR2 - generates an array of random values from this time series.

VoseTimeMultiAR2Object - creates an object for this time series.

 

Navigation