# Confidence Interval for the mean

Share:

### Description

Computes a confidence interval for the mean of the variable (parameter or feature of the process), and prints the data, a histogram with a density line, the result of the Shapiro-Wilks normality test and a quantile-quantile plot.

### Usage

 1 2 3 4 ss.ci(x, sigma2 = NA, alpha = 0.05, data = NA, xname = "x", approx.z = FALSE, main = "Confidence Interval for the Mean", digits = 3, sub = "", ss.col = c("#666666", "#BBBBBB", "#CCCCCC", "#DDDDDD", "#EEEEEE")) 

### Arguments

 x A numeric vector with the variable data sigma2 The population variance, if known alpha The eqn\alpha error used to compute the 100*(1-\α)\% confidence interval data The data frame containing the vector xname The name of the variable to be shown in the graph approx.z If TRUE it uses z statistic instead of t when sigma is unknown and sample size is greater than 30. The default is FALSE, change only if you want to compare with results obtained with the old-fashioned method mentioned in some books. main The main title for the graph digits Significant digits for output sub The subtitle for the graph (recommended: six sigma project name) ss.col A vector with colors

### Details

When the population variance is known, or the size is greater than 30, it uses z statistic. Otherwise, it is uses t statistic.
If the sample size is lower than 30, a warning is displayed so as to verify normality.

### Value

The confidence Interval.
A graph with the figures, the Shapiro-Wilks test, and a histogram.

### Note

Thanks to the kind comments and suggestions from the anonymous reviewer of a tentative article.

EL Cano

### References

Cano, Emilio L., Moguerza, Javier M. and Redchuk, Andres. 2012. Six Sigma with R. Statistical Engineering for Process Improvement, Use R!, vol. 36. Springer, New York. http://www.springer.com/statistics/book/978-1-4614-3651-5.

ss.data.rr
 1 2 3 ss.ci(len, data=ss.data.strings, alpha = 0.05, sub = "Guitar Strings Test | String Length", xname = "Length")