Test Design Under Falsification - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year :

Test Design Under Falsification

(1, 2) , (3)
1
2
3

Abstract

We characterize a receiver-optimal test when manipulations are possible in the form of type falsification. Optimal design exploits the following manipulator trade-off: while falsification may lead to better grades, it devalues their meaning. We show that optimal tests can be derived among falsification-proof ones. Our optimal test has a single ‘failing’ grade, and a continuum of ‘passing’ grades. It makes the manipulator indifferent across all moderate levels of falsification. Good types never fail, but bad types may pass. An optimal test delivers at least half of the full-information value to the receiver. A three-grade optimal test also performs well.
Fichier principal
Vignette du fichier
2018-perez-and-skreta-test-design-under-falsification.pdf (872.92 Ko) Télécharger le fichier
Origin : Publisher files allowed on an open archive

Dates and versions

hal-03393136 , version 1 (21-10-2021)

Licence

Attribution - NoDerivatives - CC BY 4.0

Identifiers

Cite

Eduardo Perez, Vasiliki Skreta. Test Design Under Falsification. 2018. ⟨hal-03393136⟩
13 View
11 Download

Share

Gmail Facebook Twitter LinkedIn More