strucScaling

Takes the MinT implementation of the 'hts'<https://cran.r-project.org/package=hts> package and adapts it to allow degenerate hierarchical structures. Instead of the "nodes" argument, this function takes an S matrix which is more versatile in the structures it allows. For a demo, see Steinmeister and Pauly (2024)<doi:10.15488/17729>. The MinT algorithm is based on Wickramasuriya et al. (2019)<doi:10.1080/01621459.2018.1448825>.

Louis Steinmeister

htsDegenerateR

Degenerate Hierarchical Time Series Reconciliation

Markus Pauly

strucScaling function

<dl><dt>fcasts</dt>
<dd>forecasts to be reconciled</dd>
<dt>Smat</dt>
<dd>structure matrix representing the hierarchical structure of the hts</dd>
<dt>weights</dt>
<dd>use the default for structural scaling and a vector of the residual variances for variance scaling</dd></dl>

Arguments

Structural Scaling reconciliation — strucScaling

<dl>

<dt>fcasts</dt>
<dd>forecasts to be reconciled</dd>


<dt>Smat</dt>
<dd>structure matrix representing the hierarchical structure of the hts</dd>


<dt>weights</dt>
<dd>use the default for structural scaling and a vector of the residual variances for variance scaling</dd>

</dl>

strucScaling: Structural Scaling reconciliation

Description

Usage

Value

Arguments