Proxy (statistics)
{{Short description|Variable used in place of another variable}}
In statistics, a proxy or proxy variable is a variable that is not in itself directly relevant, but that serves in place of an unobservable or immeasurable variable.Upton, G., Cook, I. (2002) Oxford Dictionary of Statistics. OUP {{ISBN|978-0-19-954145-4}} In order for a variable to be a good proxy, it must have a close correlation, not necessarily linear, with the variable of interest. This correlation might be either positive or negative.
Proxy variable must relate to an unobserved variable, must correlate with disturbance, and must not correlate with regressors once the disturbance is controlled for.
Examples
In social sciences, proxy measurements are often required to stand in for variables that cannot be directly measured. This process of standing in is also known as operationalization. Per-capita gross domestic product (GDP) is often used as a proxy for measures of standard of living or quality of life. Montgomery et al. examine several proxies used, and point out limitations with each, stating "In poor countries, no single empirical measure can be expected to display all of the facets of the concept of income. Our judgment is that consumption per adult is the best measure among those collected in cross-sectional surveys."Mark R. Montgomery, Michele Gragnolati, Kathleen Burke, and Edmundo Paredes, [http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.230.7240&rep=rep1&type=pdf Measuring Living Standards with Proxy Variables], Demography, Vol. 37 No. 2, pp. 155-174 (2000). (retrieved 9 Nov. 2015)
Frost lists several examples of proxy variables:Jim Frost, [http://blog.minitab.com/blog/adventures-in-statistics/proxy-variables-the-good-twin-of-confounding-variables Proxy Variables: The Good Twin of Confounding Variables], 22 September 2011 (retrieved 9 Nov. 2015)
| class="wikitable"
|+ !Proxy variable !Unobserved variable |
| Tree ring width |
| GDP per capita |
| Body mass index |
| Years of education
| rowspan="2" |Intelligence |
| Grade Point Average |
| Height growth |
See also
References
- {{cite journal | last = Toutenburg | first = Helge |author2=Götz Trenkler | title = Proxy variables and mean square error dominance in linear regression | journal = Journal of Quantitative Economics | volume = 8 | pages = 433–442 | date = 1992}}
- {{cite journal | last = Stahlecker | first = Peter |author2=Götz Trenkler | title = Some further results on the use of proxy variables in prediction | journal = The Review of Economics and Statistics | volume = 75 | pages = 707–711 | publisher = The MIT Press | date = 1993 | doi = 10.2307/2110026 | issue = 4 | jstor = 2110026}}
- {{cite journal | last = Trenkler | first = Götz |author2=Peter Stahlecker | title = Dropping variables versus use of proxy variables in linear regression | journal = Journal of Statistical Planning and Inference | volume = 50 | issue = 1 | pages = 65–75 | publisher = NORTH-HOLLAND | date = 1996 | doi = 10.1016/0378-3758(95)00045-3 }}
{{statistics-stub}}
{{econometrics-stub}}