# mardia: The Mardia's test of normality In dprep: Data Pre-Processing and Visualization Functions for Classification

## Description

Performs the Mardia's test to check for multivariate normality

## Usage

 `1` ```mardia(data) ```

## Arguments

 `data` The dataset containing the features for which multivariate normality is going to be tested. The last column contains the class. In case of unsupervised data add a dummy colummn of ones. In case of regression data, transform the response column in a column of ones

## Value

Returns the p-values for the corresponding third and fourth moments of the multivariate normal distribution.

Edgar Acuna

## References

Mardia, K.V. (1985). "Mardia's Test of Multinormality," in S. Kotz and N.L. Johnson, eds., Encyclopedia of Statistical Sciences, vol. 5 (NY: Wiley), pp. 217-221.

`vvalen`

## Examples

 ```1 2 3``` ```#-----Mardia test for supervised data----- data(iris) mardia(iris) ```

### Example output

```Warning messages:
1: In rgl.init(initValue, onlyNULL) : RGL: unable to open X11 display
2: 'rgl_init' failed, running with rgl.useNULL = TRUE
call: fun(...)
[1] 3
Mardia's test for class 1
mard1= 24.15508
pvalue for m3= 0.2356838
mard2= 0.7587116
p-value for m4= 0.4480251
There is statistical evidence for normality
Mardia's test for class 2
mard1= 23.70393
pvalue for m3= 0.2555643
mard2= -1.034219
p-value for m4= 0.3010336
There is statistical evidence for normality
Mardia's test for class 3
mard1= 24.72568
pvalue for m3= 0.2121282
mard2= -0.3384283
p-value for m4= 0.7350404
There is statistical evidence for normality
```

dprep documentation built on May 29, 2017, 11:01 a.m.