In classical statistics, a confidence interval is a statistical estamate which should capture the true value of the population a specified percentage of the time if you repeat the experiment over and over again. According to the central limit theorem, if you have a large enough sample size the point estimate usually follows a normal distribution with mean given by the true value of the parameter and the standard deviation given by the standard error of the estimate.

This web page demostrates this by constructing a specified number of confidence intervals with the requested confidence level, mean and standard error.

The following information is required.

Population Mean:  
Standard error:  
Confidence level:  
Number of Simulations:  

The following parameters are optional, but can be used to customize the appearance. It is okay to leave everything below the line alone, or you can play with the parameters to see what they do.

Line Type for target line:
Line Color for confidence intervals:

To gain a better understanding of how to write new demonstrations for this web site, you can view the source of this page. Then you can view the source of the IntervalEstimation.R script that is run as a callback.

Page developed by Russell Almond.