A nonlinear generalization of the country-product-dummy method

The present paper shows that product-specific regional price dispersion usually causes the Country-Product-Dummy (CPD) method to be biased. In cases where it is not, this index number method is still inefficient and inference is invalid. In view of this, a nonlinear generalization of the CPD method has been developed. This NLCPD method can be employed at all levels of aggregation and allows for inference. A comprehensive simulation reveals that the NLCPD method's root mean squared error is smaller than that of the CPD method, even in cases where the latter is unbiased. Finally, this paper applies the NLCPD method to regional price information derived from Germany's consumer price index micro data. Price levels of the 401 German districts are computed.