summary.nlreg.profile function

Summary Method for Objects of Class `nlreg.profile'

Returns a summary list for objects of class nlreg.profile.

## S3 method for class 'nlreg.profile'
summary(object, alpha = 0.05, twoside = TRUE, digits = NULL, ...)

Arguments

• object: a nlreg.profile object, that is, the result of a call to profile.nlreg.
• alpha: a vector of levels for confidence intervals; the default is $1-\code{alpha}=0.95$.
• twoside: a logical value. If TRUE, two-sided confidence intervals are returned. The default is TRUE.
• digits: the number of significant digits to be printed.
• ...: absorbs any additional argument.

Details

This function is a method for the generic function summary for objects of class nlreg.profile. It can be invoked by calling summary

or directly summary.nlreg.profile for an object of the appropriate class.

Returns

A list is returned with the following components: - CI: a matrix with $k*dim(\code{alpha})$ rows and 2 columns, where $k$ equals 2 or 4 depending on whether hoa = TRUE in the call that generated object. This matrix contains the upper and lower confidence bounds for the considered test statistics and for the confidence levels specified through alpha.

• inf.sk, np.sk, inf.fr, np.fr: the information and nuisance parameters aspects, that is, the two terms into which the higher order adjustment leading to the $r*$ statistic can be decomposed. The two versions refer to respectively Skovgaard's (1996) proposal and Fraser, Reid and Wu's (1999) solution. Only if hoa = TRUE in the function call that generated the nlreg.profile object argument object.

• mle: a numerical vector giving the MLE of the parameter of interest and its standard error.

• offset: character string giving the name of the interest parameter.

• twoside: a logical value indicating whether two-sided or one-sided confidence intervals were calculated.

• points: the number of output points at which the considered statistics were calculated exactly.

• n: the approximate number of points used in the spline interpolation of the considered statistics.

• hoa: a logical value indicating whether higher order solutions were calculated.

• digits: the number of significant digits to be printed.

• call: an image of the call that produced the object, but with all arguments named.

• ...: absorbs additional arguments.

References

Fraser, D.A.S., Reid, N. and Wu, J. (1999). A simple general formula for tail probabilities for frequentist and Bayesian inference. Biometrika, 86 , 249--264.

Skovgaard, I. (1996) An explicit large-deviation approximation to one-parameter tests. Bernoulli, 2 , 145--165.

See Also

nlreg.profile.object, profile.nlreg, summary

Examples

data(metsulfuron)
metsulfuron.nl <- 
    nlreg( formula = log(area) ~ log( b1+(b2-b1) / (1+(dose/b4)^b3) ), 
           weights = ~ ( 1+dose^exp(g) )^2, data = metsulfuron, 
           start = c(b1 = 138, b2 = 2470, b3 = 2, b4 = 0.07, g = log(0.3)),
           hoa = TRUE )
##
metsulfuron.prof <- profile( metsulfuron.nl, offset = g, trace = TRUE )
summary( metsulfuron.prof, alpha = c(0.9, 0.95) )