A nonmanipulable test
Olszewski, Wojciech ; Sandroni, Alvaro
Ann. Statist., Tome 37 (2009) no. 1, p. 1013-1039 / Harvested from Project Euclid
Text on Project Euclid
PDF
Alt PDF
A test is said to control for type I error if it is unlikely to reject the data-generating process. However, if it is possible to produce stochastic processes at random such that, for all possible future realizations of the data, the selected process is unlikely to be rejected, then the test is said to be manipulable. So, a manipulable test has essentially no capacity to reject a strategic expert. ¶ Many tests proposed in the existing literature, including calibration tests, control for type I error but are manipulable. We construct a test that controls for type I error and is nonmanipulable.
Publié le : 2009-04-15
Classification:  Test manipulation,  calibration,  model diagnostics,  62A01,  91A40 
@article{1236693158,
     author = {Olszewski, Wojciech and Sandroni, Alvaro},
     title = {A nonmanipulable test},
     journal = {Ann. Statist.},
     volume = {37},
     number = {1},
     year = {2009},
     pages = { 1013-1039},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1236693158}
}

Olszewski, Wojciech; Sandroni, Alvaro. A nonmanipulable test. Ann. Statist., Tome 37 (2009) no. 1, pp.  1013-1039. http://gdmltest.u-ga.fr/item/1236693158/