build_design_matrix {ARGOS}

Build Design Matrix

Description

This function first smooths the data and approximates the derivative before building the design matrix to include monomial and fourier terms.

Usage

build_design_matrix(
  x_t,
  dt = 1,
  sg_poly_order = 4,
  library_degree = 5,
  library_type = c("poly", "four", "poly_four")
)

Arguments

x_t

Matrix of observations.

dt

Time step (default is 1).

sg_poly_order

Polynomial order for Savitzky-Golay Filter.

library_degree

Degree of polynomial library (default is 5).

library_type

Type of library to use. Can be one of "poly", "four", or "poly_four".

Value

A list with two elements:

• sorted_theta - A matrix with sorted polynomial/trigonometric terms.

• monomial_orders - A vector indicating the order of each polynomial term.

• xdot_filtered - A matrix with derivative terms (dependent variable).

Examples

# Build a design matrix using the Duffing Oscillator as the state-space.
# Output provides matrix, and derivative matrix monomial orders
# (needed for running `argos`).
x_t <- duffing_oscillator(n=5000, dt = 0.01,
                          init_conditions = c(1, 0),
                          gamma_value = 0.1, kappa_value = 1,
                          epsilon_value = 5, snr = 49)
duffing_design_matrix <-
         build_design_matrix(x_t, dt = 0.01, sg_poly_order = 4,
                             library_degree = 5, library_type = "poly")
head(duffing_design_matrix$sorted_theta)