Confidence interval...

25 Views Asked by At

State whether or not the following is true. Explain why: Someone says that a computed result (0.06,0.07) of a 95% confidence interval of a parameter, β, implies that P[0.06 < β < 0.07] = 0.95.

It seems to me that the above statement is true. However, my instinct somehow tells me that it is not.

Any ideas?

1

There are 1 best solutions below

0
On

Let me attempt to describe what is going on with confidence intervals. I'll probably muddle it at a few points.

In the frequentist model, $\beta$ is not a random variable and thus it does not make sense to talk about the probability of $\beta$ being within the two values given in a confidence interval. The probability statement in the frequentist model is about the construction of the intervals, the probability that the interval constructed in a certain manner will contain the actual parameter.

In the Bayesian model, $\beta$ is a random variable and you can talk about the probability of the parameter.

It is more nuanced than this, I recommend reading any good discussion on CIs to see what is really going on.