I have to give credit where it is due, this did not come from me I am simply reposting the info. This comes from here: http://www.mystockbuddy.com/forum/bl...fda_plays.html

Quote Originally Posted by w00tw00t View Post
As some of you may know I have been busy in the past few weeks collecting data about past FDA approval dates. I have managed to create a list of 62 dates going back to December 2007 and I’m going to extend this list in the near future. I’m going to perform as small analysis even if the list is somewhat small. The results are temporary and will be adjusted in the near future.

Description:
%2MBDtoDate: % return if you buy the stock 2months before approval date and sell it the day before the approval date.

%1MBDtoDate: % return if you buy the stock 1month before approval date and sell it the day before the approval date.

%2WBDtoDate: % return if you buy the stock 2weeks before approval date and sell it the day before the approval date.

%2MBDto1MBD: % return if you buy the stock 2months before approval date and sell it 1month before the approval date.

%1MBDto2WBD: % return if you buy the stock 1month before approval date and sell it 2weeks before the approval date.

I decided to split the list into small groups using the the pps 2 months before the approval date as a criterion(or is it criterium?? :p)

The approval date is the date based upon the FDA guidelines that is mentioned everywhere.

Here are the results:


1)The first group I created is the one containing pps from 0-4.99 2 months before the approval date.

%2MBDtoDate 0,72 statistically significant (5% confidence interval)
%1MBDtoDate 0,52 statistically significant (5% confidence interval)
%2WBDtoDate 0,22 statistically significant (10% confidence interval)
%2MBDto1MBD 0,15 statistically significant (10% confidence interval)
%1MBDto2WBD 0,22 statistically significant (5% confidence interval)

Conclusion: The sample size for this test was smaller then allowed (20 (<30)) so I’m not going to see this as a definitive conclusion but we can see that holding a share for 2months before the approval date and selling it at the date before the approval will earn you an average of 72%. Doing so for shares bought 1month before the approval will give you an average return of 52% and 22% for shares bought 2 weeks before approval. Even if you buy shares 2months before approval and sell them 1month before approval, they could give you a return of 15%
(I have let the 10% confidence interval slip through this time because of the smaller sample. I normally only work with 5% confidence intervals or lower)


2) The second group I created contains pps from 5-14.99 2 months before the approval date.

%2MBDtoDate 0/06
%1MBDtoDate 0/05
%2WBDtoDate 0,01
%2MBDto1MBD 0,01
%1MBDto2WBD 0, 04

Conclusion: none of the found results are statistically significant. Basically this tells us that shares with pps in this range don’t move by much on average before a possible FDA approval. The market forces have more influence on them than the speculators playing on FDA approvals.


3) the third groups contains pps from 15 and higher 2 months before the approval date

Conclusion: The found results are even worse than the ones shown in the previous group so I wont even bother to post them.



As my list gets bigger I can use smaller ranges for groups which will give me even better results. I’m also indentifying all the possible factors that can influence the pps in those FDA play stocks and by how much those affect them. This information could be used to create time series regressions in order to predict possible future movements. This concludes my small analysis. I hope this was informative.