Next: toclip() Up: MacAnova Help File Previous: time_series   Contents


tint(x,Coverage), x a REAL vector or matrix, 0 < Coverage < 1 a REAL

Keywords: probabilities, descriptive statistics, confidence intervals
tint(x,Coverage) computes a (two sided) t confidence interval with
coverage rate Coverage for the population mean of the data in REAL
vector x.  The result is vector(lowerLimit, upperLimit).

Coverage must be a REAL scalar between 0 and 1.

When x is a matrix with M columns, tint(x, Coverage) returns a 2 by M
matrix, with lower and upper limits in rows 1 and 2, respectively.

When there are missing values, an informative message is printed.

  Cmd> alpha <- .05; tint(x,1-alpha) # computes 95% conf. interval.

See also tval(), t2val(), and t2int().

Gary Oehlert 2003-01-15