MakeDENsample: Convenience function for converting log-quantile functions to...

In functionaldata/tDENS: Functional Data Analysis for Density Functions by Transformation to a Hilbert Space

Description

See 'lqd2dens' and 'RegulariseByAlpha' for more details. This function will automatically calculate the:sout support points that the 99.5-th percentile (calculated using all available samples) is not positive. It will then normalise the new densities to integraate to one, regularise by 'alpha' and calculate the LQ projection.

Usage

MakeDENsample(qmatrix, alpha = 0, dSupUsedInLQ, dSup, useAlpha = FALSE)

Arguments

qmatrix	Matrix holding the log-quantiles values on [0,1] and assumed to lay strictly on dSupUsedInLQ
alpha	Scalar to deregularise the supports with (default=0)
dSupUsedInLQ	Support (grid) for Density domain that was used when calculating the LQ projections (maybe different from the original)
dSup	Original support (grid) for Density domain of the data
useAlpha	Logical indicator to regularise the support so the smallest value of each density is zero

Value

list with the 'DEN' projected data, and 'dSup' that was originally used for the data.

References

Functional Data Analysis for Density Functions by Transformation to a Hilbert space, Alexander Petersen and Hans-Georg Mueller, 2015

functionaldata/tDENS documentation built on May 16, 2019, 3:38 p.m.