Replies to post #98154 on Cytodyn Inc (CYDY)

[iddrisw]

Hi iddrisw and thank you for your posts predicting statistics.

I wanted to ask you a few questions if you don’t mind:

1) I take that you are using Montecarlo simulations to generate the random samples, correct? If so, what type of assumption, if any, do you need to make about the data distribution? (e.g., normal)
it is a random number generator in excel. using RANDBETWEEN(6,11) for the treatment arm and RANDBETWEEN(7,12) to generate slightly worse scores for placebo. calculate the average and sd for each pop. Use F9 to cycle through generator until the difference between the averages is greater than 1. copy paste the data into online statistical calculator eg https://www.socscistatistics.com/tests/mannwhitney/default2.aspx
2) Are the parametric simulations based on the the assignation of the 0-12 scale? What p values are you obtaining with the average difference of 1 between treated and placebo arms? Did you check p values assuming larger differences like 1.5 or 2? pvalues approx. 0.001
3j Are the non-parametric simulations solely based on rankings? What p values are you obtaining with the average difference of 1 between treated and placebo arms? Did you check p values assuming larger differences like 1.5 or 2?

the difference of 1 is the worse case scenario or threshold and the pvalue is close to 0.05 or worse in some instances. , if the difference is 1.2, the pvalue is 0.02'ish, if the difference is 1.8'ish the p value is approx 0.001. the results were very consistent for each simulation run meeting the condition and consistent across the different test statistics.

Thank you and I truly appreciate that you are sharing this good work.