Anavex Life Sciences Corp (AVXL)

[dayneyus]

I suspect MWU test is in the SAP so the FDA is on board and they used the Mann Whitney-U-test in Slide #6 to correlate mRNA with positive ADSC_ADL scores. MWU is accepetable for nonparametric data an small sample size.

https://www.sciencedirect.com/sdfe/pdf/download/eid/3-s2.0-B9780123694928500145/first-

page-pdf9.1 Why Nonparametric Tests?
The methods studied in the previous chapter were mostly concerned with data from a normal distribution. In many situations the data may consist of a number of ordered categories such as a subjective rating of the amount of pain relief (none, a little, a lot, total) a patient perceives after receiving a treatment. In other cases the data may simply be the presence or absence of a condition. In such cases the investigator may be unwill- ing to use a numerical scale but still wants to test a hypothesis related to the effect of a treatment or to the effects of two different treatments. The sign test discussed in this chapter can be used for situations with two outcomes. Other methods in this chapter are used with ordered data or with numerical data that do not follow the normal distribution.
The methods for testing the mean and proportion in the previous chapter are based on normality assumptions. If there is obvious nonnormality in the data, distribution-free methods can be used. In some cases we may suspect that the data do not follow the normal distribution, but we cannot determine the lack of normality for sure because the sample size is too small. Distribution-free methods can be used then and are also often used for small samples when the central limit theorem may not apply.

https://www.cuemath.com/data/non-parametric-test/

Reasons to Use Non-Parametric Tests

It is important to access when to apply parametric and non-parametric tests in order to arrive at the correct statistical inference. The reasons to use a non-parametric test are given below:

When the distribution is skewed, a non-parametric test is used. For skewed distributions, the mean is not the best measure of central tendency, hence, parametric tests cannot be used.
If the size of the data is too small then validating the distribution of the data becomes difficult. Thus, in such cases, a non-parametric test is used to analyze the data.
If the data is nominal or ordinal, a non-parametric test is used. This is because a parametric test can only be used for continuous data.[color=red][/color]