Obtaining reliable Likelihood Ratio tests from simulated likelihood functions
Research output: Working paper › Research
Standard
Obtaining reliable Likelihood Ratio tests from simulated likelihood functions. / Andersen, Laura Mørch.
Frederiksberg : Department of Food and Resource Economics, University of Copenhagen, 2013. p. 1-27.Research output: Working paper › Research
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - UNPB
T1 - Obtaining reliable Likelihood Ratio tests from simulated likelihood functions
AU - Andersen, Laura Mørch
PY - 2013
Y1 - 2013
N2 - It is standard practice by researchers and the default option in many statistical programs to base test statistics for mixed models on simulations using asymmetric draws (e.g. Halton draws). This paper shows that when the estimated likelihood functions depend on standard deviations of mixed parameters this practice is very likely to cause misleading test results for the number of draws usually used today. The paper shows that increasing the number of draws is a very inefficient solution strategy requiring very large numbers of draws to ensure against misleading test statistics. The paper shows that using one dimensionally antithetic draws does not solve the problem but that the problem can be solved completely by using fully antithetic draws. The paper also shows that even when fully antithetic draws are used, models testing away mixing dimensions must replicate the relevant dimensions of the quasirandom draws in the simulation of the restricted likelihood. Again this is not standard in research or statistical programs. The paper therefore recommends using fully antithetic draws replicating the relevant dimensions of the quasi-random draws in the simulation of the restricted likelihood and that this should become the default option in statistical programs.
AB - It is standard practice by researchers and the default option in many statistical programs to base test statistics for mixed models on simulations using asymmetric draws (e.g. Halton draws). This paper shows that when the estimated likelihood functions depend on standard deviations of mixed parameters this practice is very likely to cause misleading test results for the number of draws usually used today. The paper shows that increasing the number of draws is a very inefficient solution strategy requiring very large numbers of draws to ensure against misleading test statistics. The paper shows that using one dimensionally antithetic draws does not solve the problem but that the problem can be solved completely by using fully antithetic draws. The paper also shows that even when fully antithetic draws are used, models testing away mixing dimensions must replicate the relevant dimensions of the quasirandom draws in the simulation of the restricted likelihood. Again this is not standard in research or statistical programs. The paper therefore recommends using fully antithetic draws replicating the relevant dimensions of the quasi-random draws in the simulation of the restricted likelihood and that this should become the default option in statistical programs.
M3 - Working paper
T3 - IFRO Working Paper
SP - 1
EP - 27
BT - Obtaining reliable Likelihood Ratio tests from simulated likelihood functions
PB - Department of Food and Resource Economics, University of Copenhagen
CY - Frederiksberg
ER -
ID: 46953693