TY - JOUR
T1 - Nonstandard Errors
AU - Nonstandard Errors
AU - Menkveld, Albert J.
AU - Dreber, Anna
AU - Holzmeister, Felix
AU - Huber, Juergen
AU - Johannesson, Magnus
AU - Kirchler, Michael
AU - NEUSÜß, Sebastian
AU - Razen, Michael
AU - Weitzel, Utz
AU - Abad‐Díaz, David
AU - Abudy, Menachem (Meni)
AU - Adrian, Tobias
AU - Ait‐Sahalia, Yacine
AU - Akmansoy, Olivier
AU - Alcock, Jamie T.
AU - Alexeev, Vitali
AU - Aloosh, Arash
AU - Amato, Livia
AU - Amaya, Diego
AU - Angel, James J.
AU - Avetikian, Alejandro T.
AU - Bach, Amadeus
AU - Baidoo, Edwin
AU - Bakalli, Gaetan
AU - Bao, Li
AU - Barbon, Andrea
AU - Bashchenko, Oksana
AU - Bindra, Parampreet C.
AU - Bjønnes, Geir H.
AU - Black, Jeffrey R.
AU - Black, Bernard S.
AU - Bogoev, Dimitar
AU - Correa, Santiago Bohorquez
AU - Bondarenko, Oleg
AU - Bos, Charles S.
AU - Bosch‐Rosa, Ciril
AU - Bouri, Elie
AU - Brownlees, Christian
AU - Calamia, Anna
AU - Cao, Viet Nga
AU - Capelle‐Blancard, Gunther
AU - Romero, Laura M. Capera
AU - Caporin, Massimiliano
AU - Carrion, Allen
AU - Caskurlu, Tolga
AU - Chakrabarty, Bidisha
AU - Chen, Jian
AU - Chernov, Mikhail
AU - Moore, David
AU - Tran, Hai
PY - 2024/4/1
Y1 - 2024/4/1
N2 - ABSTRACT In statistics, samples are drawn from a population in a data‐generating process (DGP). Standard errors measure the uncertainty in estimates of population parameters. In science, evidence is generated to test hypotheses in an evidence‐generating process (EGP). We claim that EGP variation across researchers adds uncertainty—nonstandard errors (NSEs). We study NSEs by letting 164 teams test the same hypotheses on the same data. NSEs turn out to be sizable, but smaller for more reproducible or higher rated research. Adding peer‐review stages reduces NSEs. We further find that this type of uncertainty is underestimated by participants.
AB - ABSTRACT In statistics, samples are drawn from a population in a data‐generating process (DGP). Standard errors measure the uncertainty in estimates of population parameters. In science, evidence is generated to test hypotheses in an evidence‐generating process (EGP). We claim that EGP variation across researchers adds uncertainty—nonstandard errors (NSEs). We study NSEs by letting 164 teams test the same hypotheses on the same data. NSEs turn out to be sizable, but smaller for more reproducible or higher rated research. Adding peer‐review stages reduces NSEs. We further find that this type of uncertainty is underestimated by participants.
KW - Email
KW - department: Finance
KW - email [email protected] and [email protected]
KW - Pure
U2 - 10.1111/jofi.13337
DO - 10.1111/jofi.13337
M3 - Article
SN - 0022-1082
SP - jofi.13337
JO - Journal of Finance
JF - Journal of Finance
ER -