sptd {TSdisaggregation}

Function to do sparse temporal disaggregation from Mosley et al. (2021).

Description

Used in disaggregation.R to find estimates given the optimal rho parameter.

Usage

sptd(Y, X, rho, aggMat, aggRatio, adaptive = FALSE)

Arguments

Y

The low-frequency response series (n_l x 1 matrix).

X

The high-frequency indicator series (n x p matrix).

rho

The AR(1) residual parameter (strictly between -1 and 1).

aggMat

Aggregation matrix according to 'first', 'sum', 'average', 'last' (default is 'sum').

aggRatio

Aggregation ratio e.g. 4 for annual-to-quarterly, 3 for quarterly-to-monthly (default is 4).

adaptive

TRUE to use adaptive lasso penalty. FALSE for lasso penalty. Default is FALSE.

Value

y Estimated high-frequency response series (n x 1 matrix).

betaHat Estimated coefficient vector (p x 1 matrix).

u_l Estimated aggregate residual series (n_l x 1 matrix).

References

Mosley L, Eckley I, Gibberd A (2021). “Sparse Temporal Disaggregation.” arXiv preprint arXiv:2108.05783.