TwoMeanComparison: Comparison of two sample means
In andresfpc/AMUN: Analisis Multivariado Universidad Nacional de Colombia (AMUN)

Description Usage Arguments Value Author(s) References Examples

Many times two observations coming from a multidimensional variate and have been obtained under different conditions, a multinormal distribution is assumed with the same although unknown covariance matrix and it is desire to test the hypothesis of same mean vectors. The procedure uses the T square of Hotelling and presents a confidence interval Morrison (2005).

1	TwoMeanComparison(xbar, ybar, nx, ny, S, alpha = 0.05)

`xbar`	a vector of means of the variable x
`ybar`	a vector of means of the variable y
`nx`	sample size of x
`ny`	sample size of y
`S`	a positive define matrix of the variance and covariance
`alpha`	the significance level (0.05 by default)

`FCalc`	the value of the Computed F statistic
`FTheo`	the value of the Theoric F statistic
`IC`	a confidence interval

Jesus Gonzalez <jmgonzalezf@unal.edu.co>, Andres Palacios <anfpalacioscl@unal.edu.co>, Campo Elias Pardo <cepardot@unal.edu.co>

Morrison, D. F. (2005), Multivariate statistical methods, Series in Probability and Statistics, 4 edn, McGraw-Hill, New York.

xbar <- c(12.57, 9.57, 11.49, 7.97)
ybar <- c(8.75, 5.33, 8.50, 4.75)

nx <- 37
ny <- 12

S <- matrix(c(11.2624, 9.4060, 7.1550, 3.3791,
              9.4060, 13.5265, 7.3784, 2.5014,
              7.1550, 7.3784, 11.5796, 2.6167,
              3.3791, 2.5014, 2.6167, 5.8133),
            nrow = 4, byrow = TRUE)

TwoMeanComparison(xbar, ybar, nx, ny, S)