slope

object


Since an I-prior model does not deal with the betas (slopes or coefficients
of an ordinary linear model), this function calculates the slopes of the
regression curves based on the fitted values. While it is meant for Canonical
RKHS functions, it still works for FBM RKHS, in that the slopes returned
would be the "average" slope, i.e. the slope of the fitted straight line
through the fitted I-prior values. Currently, only models with one
explanatory variable are supported.


In a linear regression setting, priors can be assigned to the regression function using a vector space framework, and the posterior estimate of the regression function obtained. I-priors are a class of such priors based on the principle of maximum entropy. While the main interest of I-prior modelling is prediction, inference is also possible, e.g. via log-likelihood ratio tests. This package supports both formula and non-formula based input. Estimation is mainly done via an EM algorithm, but there is flexibility in using any R optimiser to maximise the log-likelihood computed by the kernel loader function.

slope: Recover the betas (slopes) of the regression curves

Description

Usage

Arguments

Value

Examples