Proxy variable

A proxy variable, or simply proxy for short, is a variable that measures a property that is usually not, not objectively , not reliable , not valid or not accessible with reasonable effort for direct measurement .

It should be noted that although the proxy variable can be determined with very high quality and accuracy under certain circumstances, this does not say anything about the thus achieved accuracy of the determination of the variable actually to be measured.

Examples of proxy variables

• Income for professional success
• Income for disposable income
• Blood alcohol concentration or breath alcohol concentration for the degree of alcohol-related impairment
• Isotope ratios in ice cores as climate proxies

literature

• Helge Toutenburg, Götz Trenkler : Proxy variables and mean square error dominance in linear regression . In: Regensburg contributions to statistics and econometrics . 28, 1991, p. 14.
• Helge Toutenburg, Götz Trenkler: Proxy variables and mean square error dominance in linear regression . In: Journal of Quantitative Economics . 8, 1992, pp. 433-442.
• Peter Stahlecker, Götz Trenkler: Some further results on the use of proxy variables in prediction . In: The MIT Press (Ed.): The Review of Economics and Statistics . 75, 1993, pp. 707-711.
• Götz Trenkler, Peter Stahlecker : Dropping variables versus use of proxy variables in linear regression . In: NORTH-HOLLAND (Ed.): Journal of Statistical Planning and Inference . 50, No. 1, 1996, pp. 65-75. doi : 10.1016 / 0378-3758 (95) 00045-3 .