Normalization
Normalization
The issue of normalization arises when the nature of an economic model is unaffected by a vector of structural parameters (coefficients) that can be arbitrarily scaled. This scaling is formally defined as normalization. A primary example is the simultaneousequation framework of money supply and money demand:
In equations (1) and (2), M_{t} stands for money stock, R_{t} for the nominal interest rate, y_{t} for real output or real income, and P_{t} for the price level. All variables are expressed in log value. The moneysupply and moneydemand shocks, and are random variables with zero mean and are independent of each other in probability distribution.
It has been long recognized that the parameter αs and βs in equations (1) and (2) can be normalized (scaled) in any way, with no consequence on economic interpretations of this equation. The conventional rule is to normalize the supply equation (1) as
With and to normalize the demand equation (2) as
With and
The coefficient ᾱ _{M } is often interpreted as an interest elasticity of the money supply, β̄ _{R } as an elasticity of demand for money, β̄ _{y } as a moneydemand elasticity with respect to changes in output or income, and β̄ _{p }as an elasticity of demand for money with respect to changes in the price level. The following analysis for this example applies to any supplydemand framework in which M_{t} is replaced by the quantity of a commodity under consideration, R_{t} replaced by the price of this commodity, and y_{t} and P_{t} replaced by any variables that affect supply but not demand.
In the 1970s econometricians began to recognize that how the supply or demand equation is normalized affects the estimator of the supply or demand elasticity (ᾱ _{M }or β̄ _{R }) when the twostage least squares (2SLS) approach is employed. The quality of this estimator is sensitive to the strength of instruments used in the 2SLS estimation, which in turn depends on whether the price variable or the quantity variable is normalized to be on the lefthand side of the supply or demand equation, as in (3) or (4). There are other methods that one can use to estimate the supply and demand equations. One dominating alternative is the fullinformation maximum likelihood (ML) approach. This approach used to be computationally infeasible for many practical problems. As computing technology improves over time, the ML approach has become more feasible to implement. One advantage of the ML approach over the 2SLS approach is that the economic meaning of the ML estimates will not be affected by normalization.
Not until the 1990s, however, did it become known that normalization matters to smallsample statistical inference about the ML estimates. Likelihoodbased smallsample inferences are affected because normalization governs the likelihood shape around the ML estimates. A poor normalization can lead to multimodal distribution, disjoint confidence intervals, and very misleading characterizations of the true statistical uncertainty.
Related to this discovery, in the Bayesian econometric literature there have been theoretical results showing that normalization can lead to illbehaved posterior distributions when a flat or symmetric prior is used. The empirical and policy significance of these results has been largely unexplored until very recently. Daniel Waggoner and Tao Zha (2003) and James Hamilton, Waggoner, and Zha (2007) show that normalization can alter economic interpretations of dynamic responses of the variables M_{t} and R_{t} to a supply or demand shock or in the above example. They use this and other examples to demonstrate that inadequate normalization may confound statistical and economic interpretations.
There are a variety of economic applications in which normalization plays an important role in likelihoodbased statistical inferences. Unfortunately, there is no mechanical way to implement the best normalization across different models. As a practical guide, therefore, it is essential to report the smallsample distributions of parameters of interest rather than the mean and standard deviation only. Bimodal and widespread distributions are the first clue that the chosen normalization may be inadequate. Carefully chosen normalization should follow the principle of preserving the likelihood shape around the ML estimate. A successful implementation of this principle for normalization is likely to maintain coherent economic interpretations when statistical uncertainty is summarized.
SEE ALSO Bayesian Econometrics; Demand; Econometrics; Matrix Algebra; Maximum Likelihood Regression; Regression Analysis; Simultaneous Equation Bias
BIBLIOGRAPHY
Hamilton, James D., Daniel F. Waggoner, and Tao Zha. 2007. Normalization in Econometrics. Econometric Reviews 26 (2–4): 221–252.
Waggoner, Daniel F., and Tao Zha. 2003. Likelihood Preserving Normalization in Multiple Equation Models. Journal of Econometrics 114: 329–347.
Tao Zha
Cite this article
Pick a style below, and copy the text for your bibliography.

MLA

Chicago

APA
"Normalization." International Encyclopedia of the Social Sciences. . Encyclopedia.com. 10 Dec. 2017 <http://www.encyclopedia.com>.
"Normalization." International Encyclopedia of the Social Sciences. . Encyclopedia.com. (December 10, 2017). http://www.encyclopedia.com/socialsciences/appliedandsocialsciencesmagazines/normalization
"Normalization." International Encyclopedia of the Social Sciences. . Retrieved December 10, 2017 from Encyclopedia.com: http://www.encyclopedia.com/socialsciences/appliedandsocialsciencesmagazines/normalization
normalization
normalization The normal distribution is one of the key distributions providing the basis for probability statistics, so that when a particular distribution is not normal, some transformation of the data may be attempted so as to achieve a normal distribution—for example charting the logarithms of values instead of the values themselves. This is known as normalizing the distribution.
Cite this article
Pick a style below, and copy the text for your bibliography.

MLA

Chicago

APA
"normalization." A Dictionary of Sociology. . Encyclopedia.com. 10 Dec. 2017 <http://www.encyclopedia.com>.
"normalization." A Dictionary of Sociology. . Encyclopedia.com. (December 10, 2017). http://www.encyclopedia.com/socialsciences/dictionariesthesaurusespicturesandpressreleases/normalization
"normalization." A Dictionary of Sociology. . Retrieved December 10, 2017 from Encyclopedia.com: http://www.encyclopedia.com/socialsciences/dictionariesthesaurusespicturesandpressreleases/normalization
normalization
normalization (normălyzayshŏn) n. (in psychiatry) the process of making the living conditions of people with learning disabilities as similar as possible to those of people who are not handicapped. This includes moves to living outside institutions and encouragement to cope with work, pay, social life, sexuality, and civil rights.
Cite this article
Pick a style below, and copy the text for your bibliography.

MLA

Chicago

APA
"normalization." A Dictionary of Nursing. . Encyclopedia.com. 10 Dec. 2017 <http://www.encyclopedia.com>.
"normalization." A Dictionary of Nursing. . Encyclopedia.com. (December 10, 2017). http://www.encyclopedia.com/caregiving/dictionariesthesaurusespicturesandpressreleases/normalization
"normalization." A Dictionary of Nursing. . Retrieved December 10, 2017 from Encyclopedia.com: http://www.encyclopedia.com/caregiving/dictionariesthesaurusespicturesandpressreleases/normalization