No that's incorrect, the first interim Bayesian analysis makes a predictive distribution on the outcome of the entire trial, using only the first 30%. Each interim analysis after becomes more accurate with less chance of error. As I described, it uses the data set of the first 30% to calculate a distribution of possible outcomes for future new, unobserved data points. This is basic Bayesian theory.
The analysis is not predicting the survival or death of known individual patients who have not completed 30 days of monitoring. It is estimating the probability of a success by simulating the known data over future possible scenarios. Hopefully this makes sense.
This is a good reference: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4247348/
And also this video on Bayes Theorem by Veritasium:
No that's incorrect, the first interim Bayesian analysis makes a...
-
- There are more pages in this discussion • 297 more messages in this thread...
You’re viewing a single post only. To view the entire thread just sign in or Join Now (FREE)
Add MSB (ASX) to my watchlist
 (20min delay)
|
|||||
|
Last
95.0¢ |
Change
0.025(2.70%) |
Mkt cap ! $1.084B | |||
| Open | High | Low | Value | Volume |
| 94.0¢ | 96.8¢ | 93.5¢ | $4.207M | 4.417M |
Buyers (Bids)
| No. | Vol. | Price($) |
|---|---|---|
| 3 | 42268 | 94.0¢ |
Sellers (Offers)
| Price($) | Vol. | No. |
|---|---|---|
| 95.0¢ | 48002 | 3 |
View Market Depth
| Last trade - 16.10pm 09/08/2024 (20 minute delay) ? |
| MSB (ASX) Chart |