Main Content

confint

Confidence intervals for fit coefficients ofcfitorsfitobject

所有的页面崩溃

Syntax

ci = confint(fitresult)
ci = confint(fitresult,level)

Description

ci= confint(fitresult)returns 95% confidence boundscion the coefficients associated with thecfitorsfitobjectfitresult。fitresultmust be an output from thefitfunction to contain the necessary information forci。ciis a 2-by-narray wheren = numcoeffs(fitresult)。第一行的cicontains the lower bound for each coefficient; the bottom row contains the upper bound.

example

ci= confint(fitresult,level)returns confidence bounds at the confidence level specified bylevel。levelmust be between0and1。The default value oflevelis0.95。

Examples

collapse all

Load the data and call thefitfunction to obtain thefitresultinformation.

loadcensusfitresult = fit(cdate,pop,'poly2')
fitresult = Linear model Poly2: fitresult(x) = p1*x^2 + p2*x + p3 Coefficients (with 95% confidence bounds): p1 = 0.006541 (0.006124, 0.006958) p2 = -23.51 (-25.09, -21.93) p3 = 2.113e+04 (1.964e+04, 2.262e+04)

To obtain the confidence intervals, call theconfintfunction onfitresult。

ci = confint(fitresult,0.95)
ci = 0.0061242 -25.086 19641 0.0069581 -21.934 22618

fitandconfintdisplay the confidence bounds in slightly different formats.

Input Arguments

collapse all

Fit information for confidence bounds calculation, specified as acfitorsfitobject.fitresultmust be an output from thefitfunction.

Confidence level, specified as a scalar. This argument must be between0and1

Data Types:single|double

Output Arguments

collapse all

Confidence bounds on the coefficients associated with thecfitorsfitobjectfitresult, returned as a matrix.

Tips

To calculate confidence bounds,confintusesR-1(the inverseRfactor fromQRdecomposition of the Jacobian), the degrees of freedom for error, and the root mean squared error. This information is automatically returned by thefitfunction and contained withinfitresult。

If coefficients are bounded and one or more of the estimates are at their bounds, those estimates are regarded as fixed and do not have confidence bounds.

Note that you cannot calculate confidence bounds ifcategory(fitresult)is'spline'or'interpolant'。

See Also

|

Topics

Introduced before R2006a

Curve Fitting Toolbox Documentation

万博1manbetx

Machine Learning Challenges: Choosing the Best Classification Model and Avoiding Overfitting

Machine Learning Challenges: Choosing the Best Classification Model and Avoiding Overfitting

Download white paper