Set identification

Last updated March 23, 2024

In statistics and econometrics, set identification (or partial identification) extends the concept of identifiability (or "point identification") in statistical models to environments where the model and the distribution of observable variables are not sufficient to determine a unique value for the model parameters, but instead constrain the parameters to lie in a strict subset of the parameter space. Statistical models that are set (or partially) identified arise in a variety of settings in economics, including game theory and the Rubin causal model. Unlike approaches that deliver point-identification of the model parameters, methods from the literature on partial identification are used to obtain set estimates that are valid under weaker modelling assumptions.^[1]

History

Early works containing the main ideas of set identification included Frisch (1934) and Marschak & Andrews (1944). However, the methods were significantly developed and promoted by Charles Manski, beginning with Manski (1989) and Manski (1990).

Partial identification continues to be a major theme in research in econometrics. Powell (2017) named partial identification as an example of theoretical progress in the econometrics literature, and Bonhomme & Shaikh (2017) list partial identification as “one of the most prominent recent themes in econometrics.”

Definition

Let $U\in {\mathcal {U}}\subseteq \mathbb {R} ^{d_{u}}$ denote a vector of latent variables, let $Z\in {\mathcal {Z}}\subseteq \mathbb {R} ^{d_{z}}$ denote a vector of observed (possibly endogenous) explanatory variables, and let ${\textstyle Y\in {\mathcal {Y}}\subseteq \mathbb {R} ^{d_{y}}}$ denote a vector of observed endogenous outcome variables. A structure is a pair $s=(h,{\mathcal {P}}_{U\mid Z})$ , where ${\mathcal {P}}_{U\mid Z}$ represents a collection of conditional distributions, and $h$ is a structural function such that $h(y,z,u)=0$ for all realizations $(y,z,u)$ of the random vectors $(Y,Z,U)$ . A model is a collection of admissible (i.e. possible) structures $s$ .^[2]^[3]

Let ${\mathcal {P}}_{Y\mid Z}(s)$ denote the collection of conditional distributions of $Y\mid Z$ consistent with the structure $s$ . The admissible structures $s$ and $s'$ are said to be observationally equivalent if ${\mathcal {P}}_{Y\mid Z}(s)={\mathcal {P}}_{Y\mid Z}(s')$ .^[2]^[3] Let $s^{\star }$ denotes the true (i.e. data-generating) structure. The model is said to be point-identified if for every $s\neq s'$ we have ${\mathcal {P}}_{Y\mid Z}(s)\neq {\mathcal {P}}_{Y\mid Z}(s^{\star })$ . More generally, the model is said to be set (or partially) identified if there exists at least one admissible $s\neq s^{\star }$ such that ${\mathcal {P}}_{Y\mid Z}(s)\neq {\mathcal {P}}_{Y\mid Z}(s^{\star })$ . The identified set of structures is the collection of admissible structures that are observationally equivalent to $s^{\star }$ .^[4]

In most cases the definition can be substantially simplified. In particular, when $U$ is independent of $Z$ and has a known (up to some finite-dimensional parameter) distribution, and when $h$ is known up to some finite-dimensional vector of parameters, each structure $s$ can be characterized by a finite-dimensional parameter vector $\theta \in \Theta \subset \mathbb {R} ^{d_{\theta }}$ . If $\theta _{0}$ denotes the true (i.e. data-generating) vector of parameters, then the identified set, often denoted as $\Theta _{I}\subset \Theta$ , is the set of parameter values that are observationally equivalent to $\theta _{0}$ .^[4]

Example: missing data

This example is due to Tamer (2010). Suppose there are two binary random variables, $Y$ and $Z$ . The econometrician is interested in $\mathrm {P} (Y=1)$ . There is a missing data problem, however: $Y$ can only be observed if $Z=1$ .

By the law of total probability,

\mathrm {P} (Y=1)=\mathrm {P} (Y=1\mid Z=1)\mathrm {P} (Z=1)+\mathrm {P} (Y=1\mid Z=0)\mathrm {P} (Z=0).

The only unknown object is $\mathrm {P} (Y=1\mid Z=0)$ , which is constrained to lie between 0 and 1. Therefore, the identified set is

\Theta _{I}=\{p\in [0,1]:p=\mathrm {P} (Y=1\mid Z=1)\mathrm {P} (Z=1)+q\mathrm {P} (Z=0),{\text{ for some }}q\in [0,1]\}.

Given the missing data constraint, the econometrician can only say that $\mathrm {P} (Y=1)\in \Theta _{I}$ . This makes use of all available information.

Statistical inference

Set estimation cannot rely on the usual tools for statistical inference developed for point estimation. A literature in statistics and econometrics studies methods for statistical inference in the context of set-identified models, focusing on constructing confidence intervals or confidence regions with appropriate properties. For example, a method developed by Chernozhukov, Hong & Tamer (2007) constructs confidence regions that cover the identified set with a given probability.

Notes

↑ Tamer 2010.
1 2 "Generalized Instrumental Variable Models - The Econometric Society". www.econometricsociety.org. doi:10.3982/ecta12223 . Retrieved 2024-01-05.
1 2 Matzkin, Rosa L. (2013-08-02). "Nonparametric Identification in Structural Economic Models". Annual Review of Economics. 5 (1): 457–486. doi:10.1146/annurev-economics-082912-110231. ISSN 1941-1383.
1 2 Lewbel 2019.

Related Research Articles

In particle physics, the electroweak interaction or electroweak force is the unified description of two of the four known fundamental interactions of nature: electromagnetism (electromagnetic interaction) and the weak interaction. Although these two forces appear very different at everyday low energies, the theory models them as two different aspects of the same force. Above the unification energy, on the order of 246 GeV, they would merge into a single force. Thus, if the temperature is high enough – approximately 10¹⁵ K – then the electromagnetic force and weak force merge into a combined electroweak force. During the quark epoch (shortly after the Big Bang), the electroweak force split into the electromagnetic and weak force. It is thought that the required temperature of 10¹⁵ K has not been seen widely throughout the universe since before the quark epoch, and currently the highest human-made temperature in thermal equilibrium is around 5.5x10¹² K (from the Large Hadron Collider).

The likelihood function is the joint probability mass of observed data viewed as a function of the parameters of a statistical model. Intuitively, the likelihood function $is the probability of observing data assuming is the actual parameter.$

In statistics, maximum likelihood estimation (MLE) is a method of estimating the parameters of an assumed probability distribution, given some observed data. This is achieved by maximizing a likelihood function so that, under the assumed statistical model, the observed data is most probable. The point in the parameter space that maximizes the likelihood function is called the maximum likelihood estimate. The logic of maximum likelihood is both intuitive and flexible, and as such the method has become a dominant means of statistical inference.

In statistics, the logistic model is a statistical model that models the log-odds of an event as a linear combination of one or more independent variables. In regression analysis, logistic regression is estimating the parameters of a logistic model. Formally, in binary logistic regression there is a single binary dependent variable, coded by an indicator variable, where the two values are labeled "0" and "1", while the independent variables can each be a binary variable or a continuous variable. The corresponding probability of the value labeled "1" can vary between 0 and 1, hence the labeling; the function that converts log-odds to probability is the logistic function, hence the name. The unit of measurement for the log-odds scale is called a logit, from logistic unit, hence the alternative names. See § Background and § Definition for formal mathematics, and § Example for a worked example.

Simultaneous equations models are a type of statistical model in which the dependent variables are functions of other dependent variables, rather than just independent variables. This means some of the explanatory variables are jointly determined with the dependent variable, which in economics usually is the consequence of some underlying equilibrium mechanism. Take the typical supply and demand model: whilst typically one would determine the quantity supplied and demanded to be a function of the price set by the market, it is also possible for the reverse to be true, where producers observe the quantity that consumers demand and then set the price.

In statistics, the score is the gradient of the log-likelihood function with respect to the parameter vector. Evaluated at a particular point of the parameter vector, the score indicates the steepness of the log-likelihood function and thereby the sensitivity to infinitesimal changes to the parameter values. If the log-likelihood function is continuous over the parameter space, the score will vanish at a local maximum or minimum; this fact is used in maximum likelihood estimation to find the parameter values that maximize the likelihood function.

In mathematical statistics, the Fisher information is a way of measuring the amount of information that an observable random variable X carries about an unknown parameter θ of a distribution that models X. Formally, it is the variance of the score, or the expected value of the observed information.

In mathematical statistics, the Kullback–Leibler (KL) divergence, denoted $, is a type of statistical distance: a measure of how one probability distribution P is different from a second, reference probability distribution Q . A simple interpretation of the KL divergence of P from Q is the expected excess surprise from using Q as a model when the actual distribution is P . While it is a measure of how different two distributions are, and in some sense is thus a "distance", it is not actually a metric, which is the most familiar and formal type of distance. In particular, it is not symmetric in the two distributions, and does not satisfy the triangle inequality. Instead, in terms of information geometry, it is a type of divergence, a generalization of squared distance, and for certain classes of distributions, it satisfies a generalized Pythagorean theorem.$

In statistics, econometrics, epidemiology and related disciplines, the method of instrumental variables (IV) is used to estimate causal relationships when controlled experiments are not feasible or when a treatment is not successfully delivered to every unit in a randomized experiment. Intuitively, IVs are used when an explanatory variable of interest is correlated with the error term (endogenous), in which case ordinary least squares and ANOVA give biased results. A valid instrument induces changes in the explanatory variable but has no independent effect on the dependent variable and is not correlated with the error term, allowing a researcher to uncover the causal effect of the explanatory variable on the dependent variable.

In statistics, the score test assesses constraints on statistical parameters based on the gradient of the likelihood function—known as the score—evaluated at the hypothesized parameter value under the null hypothesis. Intuitively, if the restricted estimator is near the maximum of the likelihood function, the score should not differ from zero by more than sampling error. While the finite sample distributions of score tests are generally unknown, they have an asymptotic χ²-distribution under the null hypothesis as first proved by C. R. Rao in 1948, a fact that can be used to determine statistical significance.

In statistics, ordinary least squares (OLS) is a type of linear least squares method for choosing the unknown parameters in a linear regression model by the principle of least squares: minimizing the sum of the squares of the differences between the observed dependent variable in the input dataset and the output of the (linear) function of the independent variable.

Vector autoregression (VAR) is a statistical model used to capture the relationship between multiple quantities as they change over time. VAR is a type of stochastic process model. VAR models generalize the single-variable (univariate) autoregressive model by allowing for multivariate time series. VAR models are often used in economics and the natural sciences.

In statistics, the Bayesian information criterion (BIC) or Schwarz information criterion is a criterion for model selection among a finite set of models; models with lower BIC are generally preferred. It is based, in part, on the likelihood function and it is closely related to the Akaike information criterion (AIC).

In statistics, Poisson regression is a generalized linear model form of regression analysis used to model count data and contingency tables. Poisson regression assumes the response variable Y has a Poisson distribution, and assumes the logarithm of its expected value can be modeled by a linear combination of unknown parameters. A Poisson regression model is sometimes known as a log-linear model, especially when used to model contingency tables.

In statistics, M-estimators are a broad class of extremum estimators for which the objective function is a sample average. Both non-linear least squares and maximum likelihood estimation are special cases of M-estimators. The definition of M-estimators was motivated by robust statistics, which contributed new types of M-estimators. However, M-estimators are not inherently robust, as is clear from the fact that they include maximum likelihood estimators, which are in general not robust. The statistical procedure of evaluating an M-estimator on a data set is called M-estimation.

In statistics, identifiability is a property which a model must satisfy for precise inference to be possible. A model is identifiable if it is theoretically possible to learn the true values of this model's underlying parameters after obtaining an infinite number of observations from it. Mathematically, this is equivalent to saying that different values of the parameters must generate different probability distributions of the observable variables. Usually the model is identifiable only under certain technical restrictions, in which case the set of these requirements is called the identification conditions.

In statistics, errors-in-variables models or measurement error models are regression models that account for measurement errors in the independent variables. In contrast, standard regression models assume that those regressors have been measured exactly, or observed without error; as such, those models account only for errors in the dependent variables, or responses.

In statistics, the class of vector generalized linear models (VGLMs) was proposed to enlarge the scope of models catered for by generalized linear models (GLMs). In particular, VGLMs allow for response variables outside the classical exponential family and for more than one parameter. Each parameter can be transformed by a link function. The VGLM framework is also large enough to naturally accommodate multiple responses; these are several independent responses each coming from a particular statistical distribution with possibly different parameter values.

In statistics and econometrics, optimal instruments are a technique for improving the efficiency of estimators in conditional moment models, a class of semiparametric models that generate conditional expectation functions. To estimate parameters of a conditional moment model, the statistician can derive an expectation function and use the generalized method of moments (GMM). However, there are infinitely many moment conditions that can be generated from a single model; optimal instruments provide the most efficient moment conditions.

Tau functions are an important ingredient in the modern mathematical theory of integrable systems, and have numerous applications in a variety of other domains. They were originally introduced by Ryogo Hirota in his direct method approach to soliton equations, based on expressing them in an equivalent bilinear form.

References

Bonhomme, Stephane; Shaikh, Azeem (2017). "Keeping the econ in econometrics:(micro-) econometrics in the journal of political economy". The Journal of Political Economy. 125 (6): 1846–1853. doi:10.1086/694620.
Chernozhukov, Victor; Hong, Han; Tamer, Elie (2007). "Estimation and Confidence Regions for Parameter Sets in Econometric Models". Econometrica. The Econometric Society. 75 (5): 1243–1284. doi:10.1111/j.1468-0262.2007.00794.x. hdl: 1721.1/63545 . ISSN 0012-9682.
Frisch, Ragnar (1934). Statistical Confluence Analysis by means of Complete Regression Systems. University Institute of Economics, Oslo.
Manski, Charles (1989). "Anatomy of the Selection Problem". The Journal of Human Resources. 24 (3): 343–360. doi:10.2307/145818.
Manski, Charles (1990). "Nonparametric Bounds on Treatment Effects". The American Economic Review. 80 (2): 319–323. JSTOR 2006592.
Marschak, Jacob; Andrews, Williams (1944). "Random Simultaneous Equations and the Theory of Production". Econometrica. The Econometric Society. 12 (3/4): 143–205. doi:10.2307/1905432.
Powell, James (2017). "Identification and Asymptotic Approximations: Three Examples of Progress in Econometric Theory". Journal of Economic Perspectives. 31 (2): 107–124. doi:10.1257/jep.31.2.107.
Lewbel, Arthur (2019-12-01). "The Identification Zoo: Meanings of Identification in Econometrics". Journal of Economic Literature . American Economic Association. 57 (4): 835–903. doi:10.1257/jel.20181361. ISSN 0022-0515. S2CID 125792293.
Tamer, Elie (2010). "Partial Identification in Econometrics". Annual Review of Economics . 2 (1): 167–195. doi:10.1146/annurev.economics.050708.143401.