Please use this identifier to cite or link to this item:
Abberger, Klaus
Year of Publication: 
Series/Report no.: 
CoFE Discussion Paper No. 02/09
University of Konstanz, Center of Finance and Econometrics (CoFE), Konstanz
To estimate cell probabilities for ordered sparse contingency tables several smooth- ing techniques have been investigated. It has been recognized that nonparametric smoothing methods provide estimators of cell probabilities that have better performance than the pure frequency estimators. With the help of simulation examples it is shown in this paper that these smoothing techniques may help to get test which are more powerful than Chi-Squared test with raw data. But the distribution of the Chi-Squared statistics after smoothing is unknown. This distribution can also be estimated by simulation methods.
nonparametric estimation
local polynomial smoothers
local likelihood
sparse contingency tables
Chi-Squared test
independence test
Persistent Identifier of the first edition: 
Document Type: 
Working Paper

Files in This Item:
171.59 kB

Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.