calculate_standard_error {mfp2}

Helper function to compute standard error of a partial predictor

Description

To be used in predict.mfp2().

Usage

calculate_standard_error(model, X, xref = NULL)

Arguments

model

fitted mfp2 object.

X

transformed input matrix with variables of interest for partial predictor.

xref

transformed reference value for variable of interest. Default is NULL, in which case this function computes standard errors without reference values.

Details

See pages 91-92 and following in the book by Royston and Sauerbrei 2008 for the formulas and mathematical details.

Value

Standard error.

References

Royston, P. and Sauerbrei, W., 2008. Multivariable Model - Building: A Pragmatic Approach to Regression Anaylsis based on Fractional Polynomials for Modelling Continuous Variables. John Wiley & Sons.