Thanks for making this video, but gosh this really is challenging for me to understand, especially as someone who, and I think is surrounded by, people who misinterpret confidence intervals as you describe in the video. I am a pharmacy resident, and often read studies that report 95% confidence intervals. Would you please check if my thinking is correct on this subject? A 95% confidence interval makes the statement that, given the design of the study, if this study were conducted a hypothetical 100 times, the confidence intervals from the 95 of the studies will include the population parameter. The problem with the 95% confidence interval statement, is that it says nothing about this actual single study's confidence interval. There is no verifying that the confidence interval of this one particular study actually contains the population the parameter. Whereas, a 95% credible interval makes the statement that, given the data collected, there is a 95% percent chance that the population parameter is in that interval. Which, unfortunately, is how I believe most (if not all) people I have encountered interpret the the confidence interval. Is my summation about confidecce interval and credible interval about right? Again, I am not a statistician, but a clinician who is trying to make sure my interpretation of meta-analyses and trial data results is correct. Thanks very much!
You are right confidence intervals can't say a whole lot about the actual parameter relative to this confidence interval, we can only make long run frequency statements about a lot of hypothetical intervals. While for confidence credible says the true parameter would lie between these credible intervals 95%. Confidence intervals bound are random and the parameter is fixed , while credible interval is fixed and the parameters are random between this fixed bound
6:38, the analogy of ring and post for the confidence interval is excellent!
High quality material. Please keep up the good work.
This was a very well explained video . Best one on credible interval found so far
Thank you very much for this video, you explain so well :-)
excellent video! thank you!
Thanks for making this video, but gosh this really is challenging for me to understand, especially as someone who, and I think is surrounded by, people who misinterpret confidence intervals as you describe in the video. I am a pharmacy resident, and often read studies that report 95% confidence intervals. Would you please check if my thinking is correct on this subject?
A 95% confidence interval makes the statement that, given the design of the study, if this study were conducted a hypothetical 100 times, the confidence intervals from the 95 of the studies will include the population parameter.
The problem with the 95% confidence interval statement, is that it says nothing about this actual single study's confidence interval. There is no verifying that the confidence interval of this one particular study actually contains the population the parameter.
Whereas, a 95% credible interval makes the statement that, given the data collected, there is a 95% percent chance that the population parameter is in that interval. Which, unfortunately, is how I believe most (if not all) people I have encountered interpret the the confidence interval.
Is my summation about confidecce interval and credible interval about right? Again, I am not a statistician, but a clinician who is trying to make sure my interpretation of meta-analyses and trial data results is correct.
Thanks very much!
You are right confidence intervals can't say a whole lot about the actual parameter relative to this confidence interval, we can only make long run frequency statements about a lot of hypothetical intervals. While for confidence credible says the true parameter would lie between these credible intervals 95%. Confidence intervals bound are random and the parameter is fixed , while credible interval is fixed and the parameters are random between this fixed bound
Great explanation!
A huge thank you!!!
This is very explanatory. Thank you
good speaker. very impressive!!
Thanks. Very well done.
Good Explanation. Thanks.
Amazing
very clear
Very nice (Y)