Boards
News
Market Data
Markets
Discover
Discover
AI Subscribe Login/Register account icon
S&P 500
6,845.03
(
0.41%
)
US TECH 100
24,558.80
(
0.47%
)
Dow Jones
47,917.69
(
0.00%
)
Brent Oil
96.88
(
-2.98%
)
Bitcoin
72,277.52
(
2.12%
)
News Focus
News Focus
Post# of 257397
Next 10
Public Reply Private Reply New Post
Keep Last Read
Replies (1) Keep Next 10 Report Keyboard Shortcuts
Replies (1) Next 10 Prev Next

iwfal

Send PM Follow Ignore
Followers 77
Posts 4790
Boards Moderated 0
Alias Born 09/06/2003

iwfal

Re: TastyTheElf post# 138551

Sunday, 03/11/2012 12:51:50 PM

Sunday, March 11, 2012 12:51:50 PM

Post# of 257397
ONTY

What I don't understand is how you derived "final alpha" vs the nominal alpha of .05



There are "sample size" formulas for determining the size a no-interim trial needs to be to in order to have the desired power. I.e.

Number of Events=F(Alpha, presumed HR, enrollment ratio, desired power).

I just inverted the formula to find alpha given the number of events.

Public Reply Private Reply New Post
Keep Last Read
Replies (1) Keep Last Read Next 10 Report Keyboard Shortcuts
Replies (1) Next 10 Prev Next

Discover What Traders Are Watching

Explore small cap ideas before they hit the headlines.

Join Today