| confint {hmmr} | R Documentation |
Confidence intervals Visser et al (2000)
Description
Values from Table 1 of Visser et al (2000) with confidence intervals obtained using four different methods; 1) likelihood profiles, 2) bootstrap with 500 samples, 3) bootstrap with 1000 samples, 4) finite differences approximation of the hessian.
Usage
data(confint)Format
A data.frame with 9 variables for each parameter of a 2-state hidden Markov model fitted
to data simplehmm:
- par
parameter name, see Visser et al (2000) for details.
- true
true value of the parameter.
- mle
maximum likelihood estimate of the parameter.
- prof
the size of the confidence interval resulting from the profile likelihood method.
- left
the left hand boundary of the (95%) confidence interval using the likelihood profile method.
- right
the right hand boundary of the (95%) confidence interval using the likelihood profile method.
- b500
the size of the confidence interval resulting from 500 bootstrap samples.
- b1000
the size of the confidence interval resulting from 1000 bootstrap samples.
- fdh
the size of the confidence interval based on the finite differences approximation of the hessian.
Source
Ingmar Visser, Maartje E. J. Raijmakers, and Peter C. M. Molenaar (2000). Confidence intervals for hidden Markov model parameters. British journal of mathematical and statistical psychology, 53, p. 317-327.
Examples
data(confint)