hl.factor.number: Factor number estimator of Hallin and Liška (2007)

In Dom-Owens-UoB/fnets: Factor-Adjusted Network Estimation and Forecasting for High-Dimensional Time Series

Factor number estimator of Hallin and Liška (2007)

Description

Estimates the number of factors by minimising an information criterion over sub-samples of the data. Currently the three information criteria proposed in Hallin and Liška (2007) (ic.op = 1, 2 or 3) and their variations with logarithm taken on the cost (ic.op = 4, 5 or 6) are implemented, with ic.op = 5 recommended as a default choice based on numerical experiments.

Usage

hl.factor.number(
  x,
  q.max = NULL,
  mm = NULL,
  center = TRUE,
  p.seq = NULL,
  n.seq = NULL
)

Arguments

x	input time series matrix, with each row representing a variable
q.max	maximum number of factors; if q.max = NULL, a default value is selected as min(50, floor(sqrt(min(dim(x)[2] - 1, dim(x)[1]))))
mm	a positive integer specifying the kernel bandwidth for dynamic PCA; by default, it is set to floor(4 *(dim(x)[2]/log(dim(x)[2]))^(1/3)))
center	whether to de-mean the input x row-wise
p.seq	user-supplied sequence of dimensionality; if p.seq = NULL, a default sequence is generated as recommended by Hallin & Liška (2007)
n.seq	user-supplied sequence of sample size; if n.seq = NULL, a default sequence is generated as recommended by Hallin & Liška (2007)

Details

See Hallin and Liška (2007) for further details.

Value

a list containing

q.hat

a vector containing minimisers of the six information criteria

References

Hallin, M. & Liška, R. (2007) Determining the number of factors in the general dynamic factor model. Journal of the American Statistical Association, 102(478), 603–617.

Dom-Owens-UoB/fnets documentation built on Nov. 22, 2024, 7:09 a.m.