# confidence interval for an unknown difference of two means. # (It implements part of R's t.test().) # Usage: baby. Welch(x, y, mu=0, conf.level=.95) # Parameters: # X: a numeric vector of data values (must have at least two values) # y: a numeric vector of data values (must have at least two values) # mu: the hypothesised difference of true means # conf.level: confidence level of the interval (must be in (0, 1)) # Details: the test is for the null hypothesis that the # difference of true means is mu against the alternative # that the true mean is not mu. # Value: a list containing these components (in this order): # $statistic: the t statistic # $parameter: degrees of freedom for the t statistic (not always an integer) # $p.value: probability of a t statistic more extreme than the one computed # $conf.int: a confidence interval for the true mean # $estimate: the estimated (sample) difference in means (Note: this corresponds to the documentation in ?t.test, but not to the behavior of t.test().) # $null.value: the specifited hypothesized value of the difference in means (mu) # (You may not use the real t.test() or related functions.)