Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/161931 
Year of Publication: 
2017
Series/Report no.: 
Discussion Papers No. 228
Publisher: 
Georg-August-Universität Göttingen, Courant Research Centre - Poverty, Equity and Growth (CRC-PEG), Göttingen
Abstract: 
We present a new approach to treatment assignment in (field) experiments for the case of one or multiple treatment groups. This approach - which we call minimizing MSE approach - uses sample characteristics to obtain balanced treatment groups. Compared to other methods, the min MSE procedure is attrition tolerant, more flexible and very fast, it can conveniently be implemented and balances different moments of the distribution of the treatment groups. Additionally, it has a clear theoretical foundation, which bases on the idea by Kasy (2016), but involves randomness, works without any parameter to be specified by the researcher and is extended to multiple treatments. The information used for treatment assignment can be multivariate and continuous and consist of arbitrary many variables. In this paper, we theoretically derive the underlying selection criteria which we then apply to various simulated treatment effect scenarios and datasets, comparing it to established approaches. Our proposed method performs superior or comparable to competing approaches such as matching in most measures of balance commonly used. We provide software to apply the min MSE approach as an ado-package for Stata.
Document Type: 
Working Paper

Files in This Item:
File
Size
706.98 kB





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