Abstract:
|
Arguments and criticisms against the p-value are numerous, yet the use of the p-value in statistical inference has not slowed appreciably. Ideally, a statistical testing problem would be summarized through a measure of belief and/or a measure of evidence. Typically, a p-value does not provide such an interpretation. However, there are many situations in which the p-value does provide an insightful measure. In this talk, we look at some competing statistical philosophies to arrive at different motivations for the use of a p-value.
|