model function

Multivariate structural time series model definition

model(y, components, seas.period = NULL, cycle.period = NULL)

Arguments

• y: t x d data.frame (or matrix) of observations, where d is the number of time series in the multivariate model.
• components: Character vector specifying the components of the multivariate structural time series model. Possible values are c("trend", "slope", "seasonal", "cycle").
• seas.period: Length of the seasonal pattern, if present.
• cycle.period: Length of the cycle pattern, if present.

Returns

An object of class 'SSModel'.

Examples

# Example 1 : local level + seasonal
y <- cbind(seq(0.5,200,by=0.5)*0.1 + rnorm(400),
           seq(100.25,200,by=0.25)*0.05 + rnorm(400),
           rnorm(400, 5,1))
model.1 <- model(y, components = c("trend","seasonal"), seas.period = 7)

# Example 2: local level  + cycle
t <- seq(from = 0,to = 4*pi, length.out=300)
y <- cbind(3*sin(2*t)+rnorm(300), 2*cos(2*t) + rnorm(300))
model.2 <- model(y, components = c("trend", "cycle"), cycle.period = 75)