wmse: wMSE

In MatthewGrigsby/growthmetrics: Tools for applying statistic modelling and comparison techniques developed by researchers at HBGDKi and the Johns Hopkins University Center for Global NCD Research and Training

Description

Function for weighting MSE, nMSE, or aMSE by the specified variable for weights.

Usage

wmse(observed = "observed", predicted = "pred", id.var = "id",
  weight.var = "weights", type = "nmse", data)

Arguments

observed	observed growth values (e.g. height or weight)
predicted	predicted values from models fitted to observed data
id.var	variable that identifies individual subjects
weight.var	Variable used to weight MSE or nMSE. This should be a vector of values that will be used to divide subject-specific MSE estimates by. An example could be using subject-specific growth trajectories (i.e. weighting individuals with slowest growth).
type	Type of MSE estimate used as denominator. Default is nMSE but can be se to standard MSE.

Value

data.frame with id and subject-specific weighted MSE or nMSE estimates

References

Grigsby MR, Di J, Leroux A, Zipunnikov V, Xiao L, Crainiceanu C, Checkley W. Novel metrics for growth model selection. Emerging themes in epidemiology. 2018 Feb;15(1):4.

MatthewGrigsby/growthmetrics documentation built on May 25, 2019, 8:29 p.m.