dimcalc_var: Calculate the optimal number of Eigenvalues to be retained.

In mpia: Meaningful Purposive Interaction Analysis

Description

This method calculates the optimal number of rooted Eigenvalues (of the singular value decomposition) to be kept in the MPIA factor reduction in such a way that 80% of the total amount of stretch is achieved. The total amount of stretch expands the Eigenvectors to match the mapping defined by the original ('raw') document-term matrix.

Usage

	dimcalc_var()

Details

This method provides a mathematically grounded way to predict the optimal number of dimensions to keep in the truncation of the truncated singular value decomposition.

Value

Returns a function accepting the rooted Eigenvalues as parameter s.

Author(s)

Fridolin Wild <[e-mail]>

References

Fridolin Wild (2016): Learning Analytics in R with SNA, LSA, and MPIA, Springer:Berlin.

See Also

dimcalc_raw()

Examples

dimcalc_var()(c(5,4,3,2,1))

mpia documentation built on May 2, 2019, 4:18 p.m.