Expected value of sample information

From Wikipedia, the free encyclopedia

In decision theory, the expected value of sample information (EVSI) is the price that one would be willing to pay in order to gain access to a sample from the distribution about which the prediction has to be made. For example, if one needs to predict whether a coin is fair, a valuable sample might be the result (number of heads and number of tails) of a number of flips that were done with the coin.

The expected value of sample information is defined to be the difference between the expected value given the sample information and the expected monetary value (without that information):

$\mbox{EVSI} = \mbox{EV}|\mbox{SI} - \mbox{EMV} \,$

[edit] See also

Expected value of perfect information

[edit] References

This article does not cite any references or sources. (July 2007)
Please help improve this article by adding citations to reliable sources. Unverifiable material may be challenged and removed.