Scilab Home page | Wiki | Bug tracker | Forge | Mailing list archives | ATOMS | File exchange
Scilab 5.4.0
Change language to: English - Français - Português - 日本語 -

See the recommended documentation of this function

Scilab help >> Statistics > Hypothesis Testing > ftuneq

# ftuneq

Fisher ratio for samples of unequal size.

### Calling Sequence

```f=ftuneq(sample1[,sample2[,sample3]...]])
[f,p]=ftuneq(sample1[,sample2[,sample3]...]])```

### Arguments

sample1, sample2, sample3,...

real or complex matrix of any type

### Description

This function computes the F ratio for samples of unequal size.

"The most efficient design is to make all samples the same size n. However when this is nor feasible, it still is possible to modify the ANOVA calculations." Note that the definition of xbarbar is no longer mean(xbar), but rather a weighted average with weights ni. Additionnally it gives (in p) the p-value of the computed Fisher ratio.

Given a number a of samples each of them composed of n_i (i from 1 to a) observations this fonction computes in f the Fisher ratio (it is the ratio between nr times the variance of the means of samples and the mean of the variances of each sample).

`f=ftest(samples)` computes the Fisher ratio of the nc samples whose values are in the columns of the matrix `samples`. Each one of these samples is composed of nr values. (The Fisher ratio is the ratio between nr times the variance of the means of samples and the mean of variances of each sample)

`[f,p]=ftest(samples)` gives in p the p-value of the computed Fisher ratio f.

### References

Wonacott, T.H. & Wonacott, R.J.; Introductory Statistics, J.Wiley & Sons, 1990.

### Examples

```samples=[46 55 54;53 54 50; 49 58 51;50 61 51;46 52 49]
[f,p]=ftest(samples)```