NFCP_domains {NFCP}

N-Factor MLE search boundaries

Description

Generate boundaries for the domain of parameters of the N-factor model for parameter estimation.

Usage

NFCP_domains(
  parameters,
  kappa = NULL,
  lambda = NULL,
  sigma = NULL,
  mu = NULL,
  mu_rn = NULL,
  rho = NULL,
  season = NULL,
  ME = NULL,
  x_0 = NULL,
  E = NULL
)

Arguments

parameters

a vector of parameter names of an N-factor model. Function NFCP_parameters is recommended.

kappa

A vector of length two specifying the lower and upper bounds for the 'kappa' parameter

lambda

A vector of length two specifying the lower and upper bounds for the 'lambda' parameter

sigma

A vector of length two specifying the lower and upper bounds for the 'sigma' parameter

mu

A vector of length two specifying the lower and upper bounds for the 'mu' parameter

mu_rn

A vector of length two specifying the lower and upper bounds for the 'mu_rn' parameter

rho

A vector of length two specifying the lower and upper bounds for the 'rho' parameter

season

A vector of length two specifying the lower and upper bounds for the 'season' parameter

ME

A vector of length two specifying the lower and upper bounds for the 'ME' (i.e., measurement error) parameter

x_0

A vector of length two specifying the lower and upper bounds for the 'x_0' parameter

E

A vector of length two specifying the lower and upper bounds for the 'E' parameter

Details

The NFCP_domains function generates lower and upper bounds for the parameter estimation procedure in the format required of the 'Domains' argument of the 'genoud' function. NFCP_domains allows easy setting of custom boundaries for parameter estimation, whilst also providing default domains of parameters.

Value

A matrix of defaulted domains for the given unknown parameters. The first column corresponds to the lower bound of the allowable search space for the parameter, whilst the second column corresponds to the upper bound. These values were set to allow for the 'realistic' possible values of given parameters as well as restricting some parameters (such as variance and mean-reverting terms) from taking negative values. The format of the returned matrix matches that required by the Domains argument of the Genoud function from the package RGenoud.

References

Mebane, W. R., and J. S. Sekhon, (2011). Genetic Optimization Using Derivatives: The rgenoud Package for R. Journal of Statistical Software, 42(11), 1-26. URL http://www.jstatsoft.org/v42/i11/.

Schwartz, E. S., and J. E. Smith, (2000). Short-Term Variations and Long-Term Dynamics in Commodity Prices. Manage. Sci., 46, 893-911.

Examples

##Specify the Schwartz and Smith (2000) two-factor model
##with fixed measurement error:
parameters_2F <- NFCP_parameters(N_factors = 2,
                                        GBM = TRUE,
                                        initial_states = TRUE,
                                        N_ME = 1)

###Generate the default 'domains' argument of 'NFCP_MLE' function:
NFCP_MLE_bounds <- NFCP_domains(parameters_2F)