MuMIn (version 0.13.3)

par.avg: Parameter averaging

Description

Averages single parameter based on provided weights

Usage

par.avg(x, se, npar, weight, alpha = 0.05)

Arguments

x

vector of parameters

se

vector of standard errors

npar

vector giving numbers of estimated parameters

weight

vector of weights

alpha

significance level for calculatinq confidence intervals

Value

• par.avg returns a vector with named elements:
• Coefficient
  model coefficients
• Variance
  unconditional variance of coefficients
• Unconditional SE
• Lower CI, Upper CI
  relative variable importances

encoding

utf-8

References

Burnham, K. P. and Anderson, D. R (2002) Model selection and multimodel inference: a practical information-theoretic approach. 2nd ed.

See Also

model.avg for averaging models.