Legume,
It is bad to bias positively for a stock, I agree, however it is also bad to be negatively Biased.
Your assumption about a random test of 50% is incorrectly placed. So is your remark about not being a world first, I'll explain why.
Firstly, statistically, picking 38/57 out of a total pool of 576 is above a P99 of statistical certainty that this has not happened by accident (i.e. Hypothesis = true). If that was the whole scenario I'd agree with some of the potential people are spouting here, but as you have correctly noted, they have not told you how many false positives were given out.
Were only 38 people identified total? or were 80 people identified and 38 had DKD?
I agree with this point, but if it does come true that only 38 were picked, with no false positives (or 1-2 false positives only), then this technique is statistically validated for this sample size.
Secondly, the similar test that they mention is for a different disease and utilises different anti-body markers. You shouldn't take things out of context to support your own argument, at the start of the announcement it clearly states:
"The test, called PromarkerD, is the world’s first proteomics-derived predictive (prognostic) test for the diagnosis of DKD, and represents a global breakthrough in the diagnosis and treatment of the disease. There is currently no available test for predicting the onset of DKD"
Thus the conclusion of this statement, mixed with the similar procedure statement (for FDA approval) would lead to the deduction that there is another disease which is currently predicted by a similar method on different antibodies.
I don't know what PIQ is worth and I don't hold currently, I just think there is some value to this announcement. I also think there is key points left off, and the cynic in me does think that they wouldn't leave those out accidently.. but it has happened before.
Expand
(20min delay)