Tobit model

From Wikipedia, the free encyclopedia

Economics
GDP per capita by country (World Bank, 2011)
General classifications
Microeconomics Macroeconomics History of economic thought Methodology Heterodox approaches
Technical methods
Econometrics Experimental Mathematical National accounting
Fields and subfields
Agricultural Behavioral Business Computational Cultural Demographic Development Ecological Economic systems Education Environmental Evolutionary Expeditionary Game theory Geography Growth Health History Industrial organization Information International Labour Law Managerial Monetary and Financial economics Natural resource Personnel Public and Welfare economics Regional Rural Urban Welfare
Lists
Categories Economists Index Journals Outline Publications
Business and economics portal
v t e

The Tobit model is a statistical model proposed by James Tobin (1958) to describe the relationship between a non-negative dependent variable $y_{i}$ and an independent variable (or vector) $x_{i}$ . The term Tobit was derived from Tobin's name by truncating and adding -it by analogy with the probit model.^[1]

The model supposes that there is a latent (i.e. unobservable) variable $y_{i}^{*}$ . This variable linearly depends on $x_{i}$ via a parameter (vector) $\beta$ which determines the relationship between the independent variable (or vector) $x_{i}$ and the latent variable $y_{i}^{*}$ (just as in a linear model). In addition, there is a normally distributed error term $u_{i}$ to capture random influences on this relationship. The observable variable $y_{i}$ is defined to be equal to the latent variable whenever the latent variable is above zero and zero otherwise.

$y_{i}={\begin{cases}y_{i}^{*}&{\textrm {if}}\;y_{i}^{*}>0\\0&{\textrm {if}}\;y_{i}^{*}\leq 0\end{cases}}$

where $y_{i}^{*}$ is a latent variable:

$y_{i}^{*}=\beta x_{i}+u_{i},u_{i}\sim N(0,\sigma ^{2})\,$

Consistency

If the relationship parameter $\beta$ is estimated by regressing the observed $y_{i}$ on $x_{i}$ , the resulting ordinary least squares regression estimator is inconsistent. It will yield a downwards-biased estimate of the slope coefficient and an upward-biased estimate of the intercept. Takeshi Amemiya (1973) has proven that the maximum likelihood estimator suggested by Tobin for this model is consistent.

Interpretation

The $\beta$ coefficient should not be interpreted as the effect of $x_{i}$ on $y_{i}$ , as one would with a linear regression model; this is a common error. Instead, it should be interpreted as the combination of (1) the change in $y_{i}$ of those above the limit, weighted by the probability of being above the limit; and (2) the change in the probability of being above the limit, weighted by the expected value of $y_{i}$ if above.^[2]

Variations of the Tobit model

Variations of the Tobit model can be produced by changing where and when censoring occurs. Amemiya (1985, p. 384) classifies these variations into five categories (Tobit type I - Tobit type V), where Tobit type I stands for the first model described above. Schnedler (2005) provides a general formula to obtain consistent likelihood estimators for these and other variations of the Tobit model.

Type I

The Tobit model is a special case of a censored regression model, because the latent variable $y_{i}^{*}$ cannot always be observed while the independent variable $x_{i}$ is observable. A common variation of the Tobit model is censoring at a value $y_{L}$ different from zero:

$y_{i}={\begin{cases}y_{i}^{*}&{\textrm {if}}\;y_{i}^{*}>y_{L}\\y_{L}&{\textrm {if}}\;y_{i}^{*}\leq y_{L}.\end{cases}}$

Another example is censoring of values above $y_{U}$ .

$y_{i}={\begin{cases}y_{i}^{*}&{\textrm {if}}\;y_{i}^{*}<y_{U}\\y_{U}&{\textrm {if}}\;y_{i}^{*}\geq y_{U}.\end{cases}}$

Yet another model results when $y_{i}$ is censored from above and below at the same time.

$y_{i}={\begin{cases}y_{i}^{*}&{\textrm {if}}\;y_{L}<y_{i}^{*}<y_{U}\\y_{L}&{\textrm {if}}\;y_{i}^{*}\leq y_{L}\\y_{U}&{\textrm {if}}\;y_{i}^{*}\geq y_{U}.\end{cases}}$

The rest of the models will be presented as being bounded from below at 0, though this can be generalized as we have done for Type I.

Type II

Type II Tobit models introduce a second latent variable.

$y_{{2i}}={\begin{cases}y_{{2i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

Heckman (1987) falls into the Type II Tobit. In Type I Tobit, the latent variable absorb both the process of participation and 'outcome' of interest. Type II Tobit allows the process of participation/selection and the process of 'outcome' to be independent, conditional on x.

Type III

Type III introduces a second observed dependent variable.

$y_{{1i}}={\begin{cases}y_{{1i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

$y_{{2i}}={\begin{cases}y_{{2i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

The Heckman model falls into this type.

Type IV

Type IV introduces a third observed dependent variable and a third latent variable.

$y_{{1i}}={\begin{cases}y_{{1i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

$y_{{2i}}={\begin{cases}y_{{2i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

$y_{{3i}}={\begin{cases}y_{{3i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

Type V

Similar to Type II, in Type V we only observe the sign of $y_{{1i}}^{*}$ .

$y_{{2i}}={\begin{cases}y_{{2i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

$y_{{3i}}={\begin{cases}y_{{3i}}^{*}&{\textrm {if}}\;y_{{1i}}^{*}>0\\0&{\textrm {if}}\;y_{{1i}}^{*}\leq 0.\end{cases}}$

The likelihood function

Below are the likelihood and log likelihood functions for a type I Tobit. This is a Tobit that is censored from below at $y_{L}$ when the latent variable $y_{j}^{*}\leq y_{L}$ . In writing out the likelihood function, we first define an indicator function $I(y_{j})$ where:

$I(y_{j})={\begin{cases}0&{\textrm {if}}\;y_{j}=y_{L}\\1&{\textrm {if}}\;y_{j}\neq y_{L}.\end{cases}}$

Next, we mean $\Phi$ to be the standard normal cumulative distribution function and $\phi$ to be the standard normal probability density function. For a data set with N observations the likelihood function for a type I Tobit is

$\prod _{{j=1}}^{N}\left({\frac {1}{\sigma }}\phi \left({\frac {Y_{j}-X_{j}\beta }{\sigma }}\right)\right)^{{I\left(y_{j}\right)}}\left(1-\Phi \left({\frac {X_{j}\beta -y_{L}}{\sigma }}\right)\right)^{{1-I\left(y_{j}\right)}}$

References

↑ International Encyclopedia of the Social Sciences (2008)
↑ McDonald, John F.; Moffit, Robert A. (1980), "The Uses of Tobit Analysis", The Review of Economics and Statistics (The MIT Press) 62 (2): 318–321

External links

Guided tour on Tobit models

v t e Economics

Macroeconomics	Adaptive expectations Aggregate demand Balance of payments Business cycle Capacity utilization Capital flight Central bank Consumer confidence Currency Demand shock DSGE Economic growth Economic indicator Effective demand General Theory of Keynes Great Depression Hyperinflation Inflation Interest Interest rate Investment IS–LM model Microfoundations Monetary policy Money NAIRU National accounts PPP Rate of profit Rational expectations Recession Saving Stagflation Supply shock Unemployment Macroeconomics publications

Microeconomics	Aggregation problem Budget set Consumer choice Convexity Cost–benefit analysis Deadweight loss Distribution Duopoly Economic equilibrium Economic shortage Economic surplus Economies of scale Economies of scope Elasticity Expected utility hypothesis Externality General equilibrium theory Indifference curve Intertemporal choice Marginal cost Market failure Market structure Monopoly Monopsony Non-convexity Oligopoly Opportunity cost Preference Production set Profit Public good Returns to scale Risk aversion Scarcity Social choice theory Sunk costs Supply and demand Theory of the firm Trade Uncertainty Utility Microeconomics publications

Sub-disciplines	Behavioral Development Economic sociology Environmental Experimental Family Financial Geography Industrial organization Information Institutional International Labour Law and economics Macroeconomics Microeconomics Political economy Public finance Welfare

Methodology	Computational economics Econometrics Economic data Experimental economics Heterodox economics Mainstream economics Mathematical economics Normative economics Positive economics Methodological publications

History of economic thought	Ancient economic thought Austrian School Chicago school of economics Classical economics Feminist economics Institutional economics Keynesian economics Marxian economics Neoclassical economics

Notable economists and thinkers within economics	Kenneth Arrow Francis Ysidro Edgeworth Milton Friedman Ragnar Frisch Friedrich Hayek Harold Hotelling John Maynard Keynes Tjalling Koopmans Paul Krugman Jacob Marschak Alfred Marshall Karl Marx John von Neumann Vilfredo Pareto David Ricardo Paul Samuelson Joseph Schumpeter Amartya Sen Herbert A. Simon Adam Smith Robert Solow Léon Walras more

International organizations	Asia-Pacific Economic Cooperation Economic Cooperation Organization European Free Trade Association International Monetary Fund Organisation for Economic Co-operation and Development World Bank World Trade Organization

Category Index Lists Outline Publications Business and economics portal

This article is issued from Wikipedia. The text is available under the Creative Commons Attribution/Share Alike; additional terms may apply for the media files.

Tobit model

Consistency

Interpretation

Variations of the Tobit model

Type I

Type II

Type III

Type IV

Type V

The likelihood function

See also

References

Further reading

External links