issues-in-hypo

# issues-in-hypo - Miscellaneous Issues in Hypothesis Testing...

This preview shows pages 1–4. Sign up to view the full content.

Miscellaneous Issues in Hypothesis Testing -. Overall Error Rate -. P(H|D) vs P(D|H) -. Oakes’ Study -. Scale Transformation -. Meta Analysis -. Alternative Approaches to Hypothesis Testing Overall Error Rate in Hypothesis Testing 1. The procedure of hypothesis testing is set up so that the type I error rate is controlled to be equal to " in the long run over repeated hypothetical, imaginary samples: P(Reject H o * H o is true) = " . Obs 1: "Over repeated sample" means that H o is assumed to be true in all such samples. Obs 2: On average, out of 100 samples drawn from the H o population, there will be approximately " (100) cases in which H o is incorrectly rejected.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2. In reality, however, we do not know whether H o is true or not. Taking into account this uncertainty about Ho, the overall error rate of making incorrect decisions, is then given by P(incorrect decision) = P(accept H 1 * H o is true) P(H o is true) + P(retain H o * H 1 is true) p(H 1 is true) = " P(H o ) + \$ (1 - P(H o )) Therefore, to obtain the overall error rate, additional information (i.e., p(H o )) is needed. P(H|D) vs P(D|H) The probability of data D given H o , P(D * H o ), is NOT the same as the probability of H o given the data, P(H o * D). As an illustrated example, consider
D = "male adult" H o = "US Senate member" For these events, note that P(D * H o ) = 0.98 but P(H o * D) = 0.000001!

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

### Page1 / 11

issues-in-hypo - Miscellaneous Issues in Hypothesis Testing...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online