|
|
|
| R2012a Documentation → Statistics Toolbox | |
Learn more about Statistics Toolbox |
|
| Contents | Index |
Return parameter confidence intervals of ProbDistUnivParam object
CI = paramci(PD)
CI = paramci(PD, Alpha)
CI = paramci(PD) returns CI, a 2-by-N array containing 95% confidence intervals for the parameters of the ProbDistUnivParam object PD. N is the number of parameters in the distribution. When you create PD by specifying parameters (such as using the ProbDistUnivParam constructor or using the fitdist function and specifying a 'binomial' or 'generalized pareto' distribution) rather than by fitting to data, the confidence intervals have a width of 0 because the parameters are viewed as estimates of an unknown parameter.
CI = paramci(PD, Alpha) returns 100*(1 - Alpha)% confidence intervals. Default Alpha is 0.05, which specifies 95% confidence intervals.
Note If you create PD with a distribution that does not support confidence intervals, then CI contains NaN values. |
| PD | An object of the class ProbDistUnivParam. |
| Alpha | A value between 0 and 1 that specifies a confidence interval. Default is 0.05, which specifies 95% confidence intervals. |
| CI | A 2-by-N array containing 100*(1 - Alpha)% confidence intervals for the parameters of the ProbDistUnivParam object PD. N is the number of parameters in the distribution. |
| © 1984-2012- The MathWorks, Inc. - Site Help - Patents - Trademarks - Privacy Policy - Preventing Piracy - RSS |