stvcm.fit: STARX: spatiotemporal autoregressive model with exogenous...

In funstatpackages/STARX: Spatiotemporal Autoregressive Model with Exogenous Variables

Description

This function fits a spatiotemporal varying coefficient model.

Usage

stvcm.fit(
  data,
  Lambda,
  V,
  Tri,
  d,
  r,
  time.knots,
  rho,
  time.bound,
  intercept.X = TRUE,
  intercept.Z = FALSE,
  return.se = FALSE
)

Arguments

data	A list of five containing the data to fit model. Y is the response variable. Z is exogenous variables with constant linear coefficients. X is exogenous variables with varying linear coefficients. location is the location of data points.
Lambda	A matrix of tuning parameters of the smoothness penalty.
V	The N by two matrix of verities of a triangulation, where N is the number of vertices. Each row is the coordinates for a vertex.
Tri	The triangulation matrix of dimension nTr by three, where nTr is the number of triangles in the triangulation. Each row is the indices of vertices in V.
d	The degree of piecewise polynomials – default is 2, and usually d is greater than one. -1 represents piecewise constant.
r	The smoothness parameter – default is 1, and 0 ≤ r < d.
time.knots	The vector of interior time.knots for univariate spline.
rho	The order of univariate spline.
time.bound	The vector of two. The boundary of univariate spline.
intercept.X	A logical number indicating whether to include varying intercept term in the model – default is TRUE.
intercept.Z	A logical number indicating whether to include constant intercept term in the model – default is FALSE.
return.se	A logical number indicating whether to calculate standard deviation of the linear estimators.

Value

n.Z	number of linear coefficients.
n.X	number of varying coefficient functions.
best.lambda	the selected smoothing tuning parameters.
theta.hat	Estimated coefficents.
gamma.hat	Estimated tensor spline coefficients.
beta.hat	Estimated tensor spline coefficients with matrix Q_2.
se.eta	Standard deviation of α, σ^2 and linear coefficients.
Y.hat	Estimated response variable.
mse	Estimated σ^2.
mle	MLE of fitted model.
MSE.Y	Mean squared error of response variable.

funstatpackages/STARX documentation built on Jan. 30, 2021, 11:47 p.m.