Collapse Sidebar

Students / Subjects



Forgot password?



A B C D E F G H I J K L M N O P Q R S T U V W X Y Z (Show all)


Stands for "earnings before interest and taxes" which is used as a measure of earnings performance of firms that is not clouded by changes in debt or equity types, or tax rules.

Source: econterms


An electronic bibliography of economics literature organized by the American Economics Association, derived partly from the Journal of Economic Literature. EconLit is made available through libraries and universities. See for more information.

Source: econterms

econometric model

An economic model formulated so that its parameters can be estimated if one makes the assumption that the model is correct.

Source: econterms


A journal whose web site is at .

Source: econterms



Source: econterms


Econometrics is the field of economics that is concerned with the application of mathematical statistics and the tools of statistical inference to the empirical measurement of relationships postulated by economic theory. That is, econometrics (hopefully) uses some clever combination of economic theory and mathematical statistics. Typically, application of econometric methods involves the following elements:

formulating an economic model appropriate to the questions to be answered;
reviewing the available statistical models and the assumptions underlying these models, and selecting the form most suitable for the problem at hand;
obtaining appropriate data, properly defined and matching the concepts of the economic model;
finding suitable computer software to enable the calculations necessary for estimating and testing the econometric model.

The ultimate goal of an econometric exercise is to see whether an economic model is consistent with empirical (observed) behavior as reflected in the data. Note that econometrics is mostly based on large samples, i.e., on observing economic relationships over a long period of time or for a large number of individuals at the same time (or both, as in the case of longitudinal or panel data). Note also that econometricians usually have to use data that were not created in a controlled experiment (as in natural and some other social sciences). An important aspect of applied work is therefore to assess whether the sample used for estimation is actually a random sample drawn from the population for which the underlying model is supposed to be appropriate ? in other words, whether the relationship of interest is empirically identified. For example, this might not be the case if there are selection problems.

Source: SFB 504

Economic decision rule

A rule in economics asserting that if the marginal benefit of an action is higher than the marginal cost, then one should undertake the action; however if the marginal cost is higher than the marginal benefit of the action, one should not undertake it.

Source: EconPort

economic discrimination

in labor markets: the presence of different pay for workers of the same ability but who are in different groups, e.g. black, white; male, female.

Source: econterms

economic environment

In a model, a specification of preferences, technology, and the stochastic processes underlying the forcing variables.

Source: econterms

economic growth

Paraphrasing directly from Mokyr, 1990: Economic growth has four basic causes:
1) Investment, meaning increases in the capital stock (Solovian growth)
2) Increases in trade (Smithian growth)
3) Size or scale effects, e.g. by overcoming fixed costs, or achieving specialization
4) Increases in knowledge, most of which is called technological progress (Schumpeterian growth).

Further elaboration is in Mokyr's book.

Source: econterms

Economic profit

Profit that takes into account both explicit and implicit costs of production. It is calculated as Total revenues minus implicit and explicit costs.

Source: EconPort

Economic Rent

In equilibrium of a market or game, the traders (or players) participate voluntarily because their payoff exceeds the one from abstaining to engage in the trade (or to play the game): in equilibrium, the participants earn profits. Part of the equilibrium profits is explained by ordinary trading or exchange activity; part of it accrues to a trader (player) by owning a fixed idiosyncratic resource which is not consumed in the transaction (interaction). The last part is called a player's (trader's) rent. An economic rent is thus the 'wage' for some fixed resource which is necessary for and valuable in a transaction but in monopolistic possession of some trader.

Apart from the costs of not using his outside options (i.e. turning to another partner for exchange), it is the rent on such idiosyncratic factors which must be conceded to a player in order to ensure his participation in the exchange process. For example, traders draw a rent from making accessible a fixed resource like 'land' (which is where the term comes from), from uniquely owning a patent or license protecting a technological achievement or professional activity, or from uniquely owning private information about a fact that influences all players' payoffs.

In analogy to rents acrrueing to a trader from possessing idiosyncratic property rights or 'tangible' assets, the equilibrium profits acrrueing to a player solely from possessing payoff-relevant information are called his information rent. The familiar consumer's surplus from microeconomics is a simple example. The consumer is left a surplus from being able to buy all the quantities consumed at the price of the last consumed unit, instead of having to pay higher prices for earlier units consumed. This 'surplus' corresponds to her information rent for knowing her entire schedule of marginal willingnesses-to-pay for different quantities of the object. If the seller instead knew this schedule of marginal valuations, he could squeeze out all the profits from the customer by selling each unit at a different price, just demanding the consumer's marginal valuation for each unit. The fact that the consumer's information is private thus guarantees him a consumer's rent.

If the seller faces a single customer that she knows very well, having observed his choices at varying prices for a long time, she could devise a schedule of quantity discounts that extracts nearly all of the consumer's rent. This changes when the seller faces a set of competing consumers, each of which has private information on his marginal willingness-to-pay for different quantities. The schedule of discounts must now prevent lower-valued consumers from copying the quantity demanded by higher-valued consumers (and thus get much for a small payment). Typically, this is achieved by selling each additional unit of quantity at a slightly lower price than the previous one, thus inducing customers with higher valuations to choose quantities such large that low-valuation customers will find it too costly to mimic choosing a large quantity. (See the entry on incentive compatibility.) In this way, however, the seller thus increases the equilibrium profits of high-valuation customers disproportionately (relative to those for low-valuations customers), i.e. she pays larger information rents to higher marginal valuations (types) of customers.

The upshot of all this is that in a game where the seller designs an incentive compatible price schedule so that the players implicitly 'sell' their private information by revealing it through their equilibrium choices, the players do not loose their information rents. Instead, the very possibility that lower typed customers can mimic the choices of higher typed customers forces the seller to leave higher information rents to higher types, which which own 'more valuable' private information. In this sense, paying information rents to economic agents is intimately related to providing incentives for the revelation private information in strategic contexts.

Source: SFB 504

economic sociology

Piore (1996) writes of two definitions of economics, a narrow one organized around optimization and a broad one organized around scarcity, and suggests that the subjects included by the larger one but not in the smaller one are the subjects of economic sociology discussed in the Handbook (1994).

More specifically, the broad definition of economics is "the study of how people employ scarce resources and distribute them over time and among competing demands" paraphrasing Paul Samuelson (1961). The narrower definition is from Gary Becker (1976): "The combined assumptions of maximizing behavior, market equilibrium, and stable preferences, used relentlessly and unflinchingly . . . [B]ehavior [of] participants who maximize their utility from a stable set of preferences and accumulate an optimal amount of information and other inputs in a variety of markets."

A bit more specifically -- optimization and formal equilibrium are not natural subjects or methods of economic sociology, but the general subjects of economics are. Economic sociology is more likely than economics to use groups or organizations rather than individuals as units of analysis. The practical definition seems to be evolving over time.

Source: econterms


The study of the allocation of scarce (limited) resources.

Source: EconPort

economies of scale

Usually one says there are economies of scale in production of cost per unit made declines with the number of units produced. It is a descriptive, quantitative term. One measure of the economies of scale is the cost per unit made. There can be analosous economies of scale in marketing or distribution of a product or service too. The term may apply only to certain ranges of output quantity.

Source: econterms

Economies of Scale

See increasing returns to scale

Source: EconPort


European Currency Unit

Source: econterms

Editor's comment on time series

A frequent and dangerous mistake for those not familiar with this language is to think that discussion of 'time series' are about data values in a sample. Actually, they are about probability distributions. It has taken this author years to get used to that, which may just be normal.

An example of the error is to think that a discussion about E[Xt] is testable or measurable. Usually it's not. It's assumed in the discussion. A sample has a computable mean, but whether a time series has a trend, or a unit root, or heteroskedasticity are statements about a conjectured process, not statements about data.

Source: econterms

education production function

Usually a function mapping quantities of measured inputs to a school and student characteristics to some measure of school output, like the test scores of students from the school.

For empirical purposes one might assume this function is linear and generate the linear regression:

Y = X'b + S'c + e

where Y is a measure of school outputs like a vector of student test scores, X is a set of measures of student attributes (collectively or individually), S is vector of measures of schools those students attend, b and c are coefficients, and e is a disturbance term.

Source: econterms


An abbreviation for the journal Explorations in Economic History.

Source: econterms


An abbreviation for European Economic Review.

Source: econterms

effective labor

In the context of a Solow model, if labor time is denoted L and labor's effectiveness, or knowledge, is A, then by effective labor we mean AL. In general means 'efficiency units' of labor or 'productive effort' as opposed to time spent.

Source: econterms


Has several meanings. Sometimes used in a theoretical context as a synonym for Pareto efficiency. Below is the econometric/statistical definition. Efficiency is a criterion by which to compare unbiased estimators. For scalar parameters, one estimator is said to be more efficient than another if the first has smaller variance. For multivariate estimators, one estimator is said to be more efficient than another if the covariance matrix of the second minus the covariance matrix of the first is a positive semidefinite matrix. Sometimes properties of the most efficient estimator can be computed; see efficiency bound.

Computation of efficiency is defined on the basis of assumed distributions of errors ('disturbance terms'). It is not calculated directly on the basis of sample information unless the sample information come from a simulation where the actual error distribution was known.

Source: econterms


Analysis of efficiency in the context of resource allocation has always been a central concern of economics, and it is an essential element of modern microeconomic theory. The ends of economic activity are the satisfaction of human needs within resource constraints, preferences, and technological constraints. In this broad sense, an efficient use of scarce resources within a given technological environment is one that maximizes the satisfaction of aggregate needs for a given set of preferences. In a narrower sense, efficiency is a commonly agreed upon criterion to compare the economic desirability of different allocations, or states of the economy, and different allocation mechanisms or institutions. The incomparability of economic preferences gives rise to a criterion that is independent of the distributional characteristics of the allocations (or institutions) compared (Pareto efficiency). Whether construed as a general purpose of economic activity, or as a criterion for evaluating different allocations and exchange institutions, efficiency is a purely technical notion that is neither related to justness or equality criteria, nor to any moral or ethic questions of economic activity.

Source: SFB 504

efficiency bound

The minimum possible variance for an estimator given the statistical model in which it applies. An estimator which achieves this variance is called efficient.

Source: econterms

efficiency units

Usually interpretable as "output per worker per hour."
More generally: An abstract measure of the amount produced for a constant production technology by a worker in some time period. Often the context is theoretical and the time period and production technology do not have to be specified.
But efficiency units can be conceived of (and theorized about) as a function of each worker's characteristics, of the vintage of equipment, of the date in history, of the production technology, and so forth.

Source: econterms

efficiency wage hypothesis

The hypothesis that workers' productivity depends positively on their wages. (For reasons this might be the case see the entry on efficiency wages.)
This could explain why employers in some industries pay workers more than employers in other industries do, even if the workers have apparently comparable qualifications and jobs. A contrasting explanation is that of hedonic models in which these differentials are explained by quality differences in the jobs.

Source: econterms

efficiency wages

A higher than market-clearing wage set by employers to, for example:
-- discourage shirking by raising the cost of being fired
-- encourage worker loyalty
-- raise group output norms
-- improve the applicant pool
-- raise morale

Labor productivity in efficiency wage models is positively related to wage.

By contrast, consider models in which the wage is equal to labor productivity in equilibrium, or models in which wages are set to reduce the likelihood of unionization (union threat models). In these, productivity is not a function of the wage.

Source: econterms


A description of either:
-- an allocation that is Pareto efficient
-- an estimator that has the minimum possible variance given the statistical model; see efficiency bound.

Source: econterms

Efficient capital market

Market efficiency is one of the major paradigms of financial economics, focussing on informational efficiency as opposed to Pareto efficiency in microeconomic theory.

Market efficiency as applied to securities markets means that it is on average impossible to gain from trading on the basis of generally available public information (information-arbitrage efficiency) and that the valuation of an asset reflects accurately the future payments to which the asset gives title (fundamental-valuation efficiency). It is apparent that market efficiency in this sense is only part of overall market efficiency.

Fama (1970) distinguishes three forms of informational efficiency: He defines weak, semi-strong and strong form efficiency as holding when the stock market prices reflect all historical price information, all publicly available information, and all information (including insider information), respectively. In order for the price to reflect exactly all information about an asset, nothing can impede the purchase or sale of securities, such as brokerage, fees, taxes and so on. To the extent that impediments exist to the trading of an asset, the prices will only imperfectly reflect information of relevance to the valuation of the security.

Most financial markets have generally been shown to be efficient in the weak or semi-strong form, although not necessarily so in the strong sense.

Source: SFB 504

efficient markets hypothesis

"A market in which prices always 'fully reflect' available information is called 'efficient.'" -- Fama, p. 383

Source: econterms


Exponential GARCH. The EGARCH(p,q) model is attributed to Nelson, (1991).

Source: econterms


An eigenvalue or characteristic root of a square matrix A is a scalar L that satisfies the equation:

det [ A - LI ] = 0

where "det" is the operator that takes a determinant of its argument, and I is the identity matrix with the same dimensions as A.

Source: econterms

eigenvalue decomposition

Same as spectral decomposition.

Source: econterms


For each eigenvalue L of a square matrix A there is an associated right eigenvector, denoted b that has the dimension of the number of rows of A. The right eigenvector satisfies: Ab = Lb

Source: econterms


An occasional abbreviation for the British academic journal Economic Journal.

Source: econterms


A measure of responsiveness. The responsiveness of behavior measured by variable Z to a change in environment variable Y is the change in Z observed in response to a change in Y. Specifically, this approximation is common:

elasticity = (percentage change in Z) / (percentage change in Y)

The smaller the percentage change in Y is practical, the better the measure is and the closer it is to the intended theoretically perfect measure.

Elasticities are often negative, but are sometimes reported in absolute value (perhaps for brevity) in which case the author is depending on the reader knowing, or quickly applying, some theory. Usually the theory is the theory of supply and demand.

Among the elasticities that show up in the economics literature are:
elasticity of quantity demanded of some product in response to a change in price of that product-- I think this is 'elasticity of demand' or 'price elasticity of demand'. These are ordinarily negative, and when author reports a positive figure it is usually just an absolute value. A reader has to decide whether the true value is negative; hopefully this is obvious.
elasticity of supply, which is analogous
elasticity of quantity demanded in response to a change in the potential consumer's income -- called 'income elasticity of demand'. These are normally positive.

Inventing another kind of elasticity is plausible. Doing so implies a partial theory of behavior -- e.g. that Y creates a reason for the agent to change behavior Z.

Source: econterms

Elimination by aspects

Tversky (1972): This rule begins by determining the most important attribute and then retrieves a cutoff value for that attribute. All alternatives with values below that cutoff are eliminated. The process continues with the most important remaining attribute(s) until only one alternative remains.
Lexicographic Strategy: This strategy first identifies the most important attribute and then selects the alternative that is best on this attribute. In the case of ties, the tied alternatives are compared on the next most important attribute and so on.
Equal Weight Strategy: It examines all alternatives and attribute values but ignores the weights (probabilities). It sums the attribute values for an alternative to get an overall score for that alternative and then selects the alternative with the highest evaluation.
Satisficing Strategy Simon (1955): This strategy considers one alternative at a time, in the order they are presented. Each attribute of the current alternative is compared to a cutoff. If an attribute fails to exceed the cutoff, then the alternative is rejected. The first alternative to pass all the cutoffs is selected.

Source: SFB 504


An occasional abbreviation for the journal Econometrica.

Source: econterms

embedding effect

The tendency of some contingent valuation survey responses to be similar across different survey questions in conflict with theories about what is valued in the utility function.

An example from Diamond and Hausman (1994): A survey might come up with a willingness-to-pay amount that was the same for either (a) one lake or (b) five lakes which include the one that was asked about individually. If lakes have some utility value to the respondent, one would have expected that five lakes would be worth more than one. Possibly the difference arises because the respondent was not expressing a specific preference for the first lake, and/or was not taking a budget constraint into account. Diamond and Hausman argue that for this reason among others contingent valuation surveys cannot arrive at good estimates for values of public goods.

Source: econterms


An attribute of the way technological progress affects productivity. In Solow (1956), any improvement in technology instantaneously affects the productivity of all factors of production. In Solow (1960) however productivity improvements were a property of only of new capital investment. In the second case we say the technologies are embodied in the new equipment, but in the first case they are disembodied.

Source: econterms


European Monetary System -- founded in 1979, its purpose was to reduce currency fluctuations, and evolved toward offering a common currency.

Source: econterms


European Monetary Union.

Source: econterms


A variable is endogenous in a model if it is at least partly function of other parameters and variables in a model. Contrast exogenous.

Source: econterms

endogenous growth model

An endogenous growth macro model is one in which the long-run growth rate of output per worker is determined by variables within the model, not an exogenous rate of technological progress as in a neoclassical growth model like those following from Ramsey (1928), Solow (1956), Swan (1956), Cass (1965), Koopmans (1965). Influential early endogenous growth models are Romer (1986), Lucas (1988), and Rebelo (1991). See the sources for this entry for more information. Hulten (2000) says 'What is new in endogenous growth theory is the assumption that the marginal product of (generalized) capital is constant, rather than diminishing as in classical theories.' Generalized capital includes the result of investments in research and development (R&D).

Source: econterms


In a general equilibrium model, an individual's endowment is a vector made up of quantities of every possible good that the individual starts out with.

Source: econterms

energy intensity

energy consumption relative to total output (GDP or GNP).

Source: econterms

Engel curve

On a graph with good 1 on the horizontal axis and good 2 on the vertical axis, envision a convex indifference curve, and a diagonal budget constraint that meets it at one point. Now move the budget constraint in and out and mark the points where the tangencies with indifference curves are. The locus of such points is the Engel curve -- it's the mapping from wealth into the space of the two goods. That is, the Engel curve is (x(w), y(w)) where w is wealth and x() and y() are the amounts of each of the goods purchased at those levels of wealth.

Hardle (1990) p 18 defines the Engel curve as the graph of average expenditure (e.g. on food) as a function of income. And on p 118, defines food expenditure as a function of total expenditure.

The name refers to 19th century Prussian statistician Ernst Engel, according to Fogel (1979).

Source: econterms

Engel effects

Changes in commodity demands by people because their incomes are rising. A generalization of Engel's law.

Source: econterms

Engel's law

The observation that "the proportion of a family's budget devoted to food declines as the family's income increases."

See also Engel effects.

Source: econterms

English open bid auction

Sequential bidding game where the standing bid wins the item unless another, higher bid is submitted. Bidders can submit bids as often as they want to, and they observe (hear) all previous bids. Often, a new bid has to increase the standing bid by some minimal amount (advance). The English auction is known to have been in use since antique times; from this auction format the word derives: the latin word augere means to increase. With stastitically independent private valuations, an English auction is equivalent in terms of payoffs to a second price sealed bid auction.

Source: SFB 504


A possible description of the actions of managers of firms. Managers can make investments that are more valuable under themselves than under alternative managers. Those investments might not maximize shareholder value. So shareholders have a moral hazard in contracting with managers.

Or, in the phrasing of Weisbach (1988): "Managerial entrenchment occurs when managers gain so much power that they are able to use the firm to further their own interests rather than the interests of shareholders."

The abstract to Shleifer and Vishny, 1989, p 123, is nicely explicit: "By making manager-specific investments, managers can reduce the probability of being replaced, extract higher wages and larger perquisities from shareholders, and obtain more latitude in determining corporate strategy."

Source: econterms


European Options Exchange

Source: econterms

Epanechnikov kernel

The Epanechnikov kernel is this function: (3/4)(1-u2) for -1<u<1 and zero for u outside that range. Here u=(x-xi)/h, where h is the window width and xi are the values of the independent variable in the data, and x is the value of the scalar independent variable for which one seeks an estimate.
For kernel estimation.

Source: econterms


"Of, relating to; or involving knowledge or the act of knowing." An economic theory might take aspects of human understanding or belief as fundamental to economic processes or outcomes.

Source: econterms


"1. The division of philosophy that investigates the nature and origin of knowledge. 2. A theory of the nature of knowledge."

Source: econterms


(Usually written with a true epsilon character.)

In a noncooperative game, for any small positive number epsilon, an epsilon-equilibrium is a profile of totally mixed strategies such that each player gives more probability weight than epsilon only to strategies that are best responses to the profile of strategies the others are playing.

For a more formal definition see sources. This is a rough paraphrase.

Source: econterms

epsilon-proper equilibrium

In a noncooperative game, a profile of strategies is an epsilon-proper equilibrium if "every player is giving his better responses much more probability weight than his worse responses (by a factor 1/epsilon), whether or not those 'better' responses are 'best'."
-- Myerson (1978), p 78.

For a more formal definition see sources. This is a rough paraphrase.

Source: econterms


Some balance that can occur in a model, which can represent a prediction if the model has a real-world analogue. The standard case is the price-quantity balance found in a supply and demand model. If the term is not otherwise qualified it often refers to the supply and demand balance. But there also exist Nash equilibria in games, search equilibria in search models, and so forth.

Source: econterms


In economics, an equilibrium is a situation in which no agent has an incentive to change any of her choices, given the constraints she faces (constraints being interpreted in a broad sense here):

her perceptions of the behavior of other agents;
the terms of trade (prices);
the strategic environment;
her individual characteristics such as perferences (or production technologies), wealth, and computing capabilities.

In addition to this central property, it is also required that every agent makes optimal choices based on correct expectations of these constraints. Important applications of the concept of an (economic) equilibrium are the formation of prices on markets (the competitve market equilibrium), and the strategic equilibria used in game theory.

Source: SFB 504

equity premium puzzle

Real returns to investors from the purchases of U.S. government bonds have been estimated at one percent per year, while real returns from stock ("equity") in U.S. companies have been estimated at seven percent per year (Kocherlakota, 1996). General utility-based theories of asset prices have difficulty explaining (or fitting, empirically) why the first rate is so low and the second rate so high, not only in the U.S. but in other countries too. The phrase equity premium puzzle comes from the framing of this problem (why is the difference so great?) and the attention focused on it by Mehra and Prescott (1985); sometimes the phrase risk free rate puzzle is used to describe the closely related question: why is the bonds rate so low? The problem can be inverted to ask: why do investors not reject the low-returning bonds in order to buy stocks, which would then raise the price of stocks and lower their subsequent returns?

The above is drawn from the excellent review by Kocherlakota (1996) which surveys the substantial literature on this subject. Abbreviating further from it: the theories against which the evidence constitute a "puzzle" (or paradox, which see) tend to have these aspects in common: (1) standard preferences described by standard utility functions, (2) contractually complete asset markets (against possible time- and state-of-the-world contingencies), and (3) costless asset trading (in terms of taxes, trading fees, and presumably information).

Overwhelmingly the discussion in the economics literature has focused on expansions to the formal theory and on refinements and expansions of data sources, rather than survey evidence. A survey of U.S. households would answer (has answered?) the question of why they invest so little in stocks.

[Editorial comment follows.] It is likely (but this is conjecture) that large fractions of the population do not seriously consider investing in stocks, and are thus not rejecting stocks because their returns are low, but rather because they do not know how and think there are some barriers to learning how; and/or they perceive the risks of stocks to be higher than they have historically been; and/or they believe their savings are insufficient to invest. These explanations suggest that as stock trading becomes easier (e.g. over the Web, with heavy marketing and easy interfaces) the theories will fit better because more of the population will buy stocks. Indeed, this has been observed over the last few years. Another class of likely explanations is that people are highly impatient to spend their income (which would conflict with standard constant-discount-rate utility functions, but agree with the evidence; see hyperbolic discounting). Seen this way, the puzzle is not why the evidence looks the way it does, but the hard theoretical problem of getting these factors into the asset pricing models.

Source: econterms


Informally: a stochastic process is ergodic if no sample helps meaningfully to predict values that are very far away in time from that sample. Another way to say that is that the time path of the stochastic process is not sensitive to initial conditions.

Two events A and B (e.g. possible sets of states of the process) are ergodic iff, taking the limit as h goes to infinity:
lim (1/h)SUMfrom i=1to i=h |Pr(A intersection with L-iB)-Pr(A)Pr(B)| = 0
Here L is the lag operator. This definition is like that of 'mixing on average'. A stochastic process is ergodic, I believe, if all possible events in it are ergodic by this definition.

If a random process is mixing, it is ergodic.

Priestly, p 340: A process is ergodic iff 'time averages' over a single realization of the process converge in mean square to the corresponding 'ensemble averages' over many realizations.

Example 1: Let xt (for integer t=0 to infinity) is known to be drawn iid from a standard normal distribution. Then knowing the value of x1 doesn't help predict the value of x2, because they are independently drawn. This time series process is ergodic.

Example 2: Suppose the process is xt=k+sin(t)+et where k is unknown and et is a white noise error. Then any sample of xt for a known t gives information about k and that is enough information to make predictions at remote times in the future that are just as good as predictions at nearby times. This process is not ergodic.

Source: econterms

ergodic properties

means persistent properties

Source: econterms

ergodic set

In the context of a stochastic processes {xt}, set E is an ergodic set if:
(i) it is a subset of the state space S of possible values of xt,
(ii) if xt is in E, then Pr(xt+1 is in E}=1, and
(iii) no proper subset of E has the property in (ii).

Source: econterms


The Employee Retirement Income Security Act of 1974, a major U.S. law which guaranteed certain categories of employees a pension after some period at their employer; there had been more ambiguity before about what rules an employer could put on which employees could get a pension. Also ERISA changed the perceived rules about whether pensions could be invested in venture capital.

Source: econterms

error-correction model

A dynamic model in which "the movement of the variables in any periods is related to the previous period's gap from long-run equilibrium."

Source: econterms

essentially stationary

A time series process {xt} is essentially stationary iff E[xt2] is uniformly bounded. (from Wooldridge)

This definition may not be standard or widely used.

I believe this means that even if the variance wanders around and is different for different t, there is a finite bound to those variances. The variance of the distribution of xt is never infinite for any t and indeed never exceeds that finite bound. Thus an ARCH-type process might be essentially stationary even though its variance is not constant for all t.

Note that there are strictly stationary processes that have infinite second moments; such processes are not essentially stationary.

Source: econterms



Source: econterms


A function of data that produces an estimate for an unknown parameter of the distribution that produced the data.
The way estimators are often discussed, they can be thought of as chosen before the data are seen. This can be hard to understand for the person new to the term. Properties of estimators (such as unbiasedness in finite samples, asymptotic unbiasedness, efficiency, and consistency) are discussed without considering any particular sample, by making assumptions about the distribution of the data, and considering the estimator in the context of the distributions.

Source: econterms

Euler equation

A first order condition that is across a time or state boundary. (Across a state boundary means a tradeoff between uncertain events.) That is, a first order condition that is a relation between a variable that has different values in different periods or different states. E.g. kt = b(1+r)kt+1 is an Euler equation, but 2nt2 - 3kt = 0 is not.

Source: econterms

Euler's constant

May refer to either the natural logarithm base e, approximately 2.71828, or to the Euler-Mascheroni (sp) constant, which is approximately .57721566.

Source: econterms


"Originally, it was a dollar-denominated deposit created either in a European bank or in the European subsidiary of an American bank, usually located in London." Here's why: (1) Americans overseas might want their deposits in dollars; (2) the dollar being the most common international currency, borrowers and lenders internationally may want to make their accounts in it; (3) the Eurodollar market was "exempt from reserve requirements and other regulatory costs imposed on domestic American banks. Superior terms in the Eurodollar market attracted American borrowers and depositors who would have otherwise patronized domestic institutions." An example of such regulation was the US Regulation Q which limited interest banks could pay.

Source: econterms


a name for the 'disease' of rigid, slow-moving labor markets in Europe in contrast to fast-moving markets, e.g. in North America.

Source: econterms

even function

A function f() is even iff f(x)=f(-x).

Source: econterms

event studies

Empirical study of prices of an asset just before and after some event, like an announcement, merger, or dividend. Can be used to discuss whether the market priced the information efficiently, whether there was private information, etc.

This method was developed by Fama, Fisher, Jensen, and Roll (1969) according to Weisbach, 1988, p 455

Source: econterms

evolutionary game theory

Describes game models in which players choose their strategies through a trial-and-error process in which they learn over time that some strategies work better than others.

Source: econterms

ex ante

Latin for "beforehand". In models where there is uncertainty that is resolved during the course of events, the ex antes values (e.g. of expected gain) are those that are calculated in advance of the resolution of uncertainty.

Source: econterms

ex dividend date

Firms pay dividends to those who are shareholders on a certain date. The next day is called the ex dividend date. People who own no shares until the ex dividend date do not receive the dividend. The price of the stocks is often adjusted downward before the start of trading on the ex dividend date because to compensate for this.

Source: econterms

ex post

Latin for "after the fact". In models where there is uncertainty that is resolved during the course of events, the ex post values (e.g. of expected gain) are those that are calculated after the uncertainty has been resolved.

Source: econterms

Excess chance measures

Starting point of the excess chance measures is a target return, for example a one-month market return, defined by the investor. Chance is than to be considered as the possibility to beat the target return. Special cases of excess chance measures are the excess probability, the excess expectation and the excess variance.

Source: SFB 504

excess kurtosis

Sample kurtosis minus 3, which means when 'excess kurtosis' is positive, there is greater kurtosis than in the normal distribution.

Source: econterms

excess returns

Asset returns in excess of the risk-free rate. Used especially in the context of the CAPM. Excess returns are negative in those periods in which returns are less than the risk-free rate. Contrast abnormal returns.

Source: econterms

Excess Supply

A situation in which the quantity supplied exceeds the quantity demanded. It occurs when the market price is greater than the equilibrium price.

Source: EconPort

exclusion restrictions

In a simultaneous equation system -- that some of the exogenous variables are not in some of the equations; often this idea is expressed by saying the coefficient next to that exogenous variable is zero. This way of putting it may make this restriction (hypothesis) testable, and may make a simultaneous equation system identified.

Source: econterms

exclusive dealing

A requirement in a contract that the buyer will only buy goods of a certain type from the stated seller.

Source: econterms


data set from Standard and Poors on compensation to American corporate executives, including stock and options ownership.

Source: econterms

existence value

The value that individuals may attach to the mere knowledge of the existence of something, as opposed to having direct use of that thing. Synonymous with nonuse value.

For example, knowledge of the existence of rare and diverse species and unique natural environments may have value to environmentalists who do not actually see them.

Source: econterms


A variable is exogenous to a model if it is not determined by other parameters and variables in the model, but is set externally and any changes to it come from external forces. Contrast endogenous.

Source: econterms


There are several, overlapping definitions: 1) The mean of a probability distribution. If the probability distribution function is F(x) then the mean would be calculated by integrating dF(x) over the domain of the probability distribution function. The expectation operator, E[], is a linear operator per Hogg and Craig, 1995, page 55.
2) In a model, the agents may have to anticipate the value of variables whose realizations may occur in the future. The values they anticipate are often called their expectations. The agents may generalize only from past realizations in a way that we can call "adaptive expectations" or they may have other information from which they hypothesize a distribution from which the realization will be drawn. From such a distribution they can calculate the mean value, and variance, and so forth. This process is one of "rational expectations." --- Note: the notation Ex[] means the expectation of the expression taken over the random variable X. The result of the expression could still be a random variable if there are other random variables in the expression.

Source: econterms

expected utility hypothesis

That the utility of an agent facing uncertainty is calculated by considering utility in each possible state and constructing a weighted average, where the weights are the agent's estimate of the probability of each state. Arrow, 1963 attributes to Daniel Bernoulli (1738) the earliest known written statement of this hypothesis.

Source: econterms

Expected utility von NeumannMorgenstern utility

An axiomatic extension of the ordinal concept of utility to uncertain payoffs. An agent possesses a von Neumann-Morgenstern utility function if she ranks uncertain payoffs according to (higher) expected value of her utility of the individual outcomes that may occur.

Source: SFB 504

expected value

The expected value of a random variable is the mean of its distribution.
In its technical use this word does not have exactly the same meaning as in ordinary English. For example, people buying a lottery ticket that has a 1/10,000 chance of paying $10,000 can expect to get zero since that is overwhelmingly the likely outcome. They can be certain they won't get $1. But the expected value of their winnings is $1.
Having said this, it is a standard implementation of 'rational expectations' to assume that agents behave in response to the expected values of the distributions they face.

Source: econterms

expenditure function

e(p,u) -- the minimum income necessary for a consumer to achieve utility level u given a vector of prices for goods p. (The consumer is presumed to get utility from the goods.)

Source: econterms


In the context of studies of employees, length of time employed anywhere. Sometimes narrowed to include only length of time employed in relevant jobs. Contrast tenure.

Source: econterms


An empirical research method used to examine a hypothesized causal relationship between independent and dependent variables. The antecedent event in a proposed causal sequence is called the independent variable. The measured effect in the causal sequence is called the dependent variable. The main methodological rule of experimentation is that the experimenter must have precise control over the experimental situation. Control involves the creation and variation (manipulation) of the independent variables. The values of the independent variable(s) define the experimental conditions or the design of the experiment. At a minimum, the design involves the application of a treatment to one group of participants and the withholding of the treatment from a comparison or control group. Aside this manipulation the situation in both groups must be hold identical. Accordingly, the dependent variable(s) has (have) to be assessed by consistent measures. In order to render the group samples comparable, participants must be randomly assigned to conditions. Differences that then occur on the measures of the dependent variable can be attributed to the factor which differentiates the groups systematically, the presence, absence or level of the independent variable. Without randomization the method is quasi-experimental (e.g. if gender is used as a factor in the design). Experiments can be conducted in the laboratory or in natural settings (field). Because it is easier to precisely control the experimental situation in the laboratory, this kind of experiments allow the experimenter to achieve a higher level of internal validity than in field experiments. However, if it is possible to sufficently control the experimental situation in natural conditions, field experiments are more likely to be externally valid.

Source: SFB 504

Experimental design

A plan for collecting and treating the data of a proposed experiment. It is important that the experimental design provides the opportunity to make appropriate inferences and decisions relating to the hypothesis from the data.

Source: SFB 504

Experimental group

In an experimental design, which is contrasting two or more groups, the experimental group of subjects is given the treatment whose effect is under investigation.

Source: SFB 504

Explicit Costs

The accounting costs involved in the production of a good or service. Explicit costs include fixed costs and variable costs, but do not include opportunity costs.

Source: EconPort

exponential distribution

A particular function form for a continuous distribution with parameter k, a scalar real greater than zero. Has pdf f(x)=ke-kx.
The mean is E[x]=1/k, and variance var(x)=1/k2. Moment-generating function is (1-kt)-1.

Source: econterms

exponential family

A distribution is a member of the exponential family of distributions if its log-likelihood function can be written in the form below.

ln L(q | X) = a(X) + b(q) + c1(X)s1(q) + c2(X)s2(q) + . . . + cK(X)sK(q) where a(), b(), and cj() and sj() for each j=1 to K are functions; q is the vector of all parameters; X is the matrix of observable data; and L() is the likelihood function as defined by the maximum likelihood procedure.

The members of the exponential family vary from each other in a(), b(), and the cj()s and sj()s. Most common named distributions are members of the exponential family.

Quoting from Greene, 1997, page 149: "If the log-likelihood function is of this form, then the functions cj() are called sufficient statistics [and] the method of moments estimators(s) will be functions of them," Those estimators will be the maximum likelihood estimators which are asymptotically efficient here.

Source: econterms

exponential utility

A particular functional form for the utility function. Some versions of it are used often in finance.

Here is the simplest version. Define U() as the utility function and w as wealth. a is a positive scalar parameter.
U(w) = -e-aw

is the exponential utility function.

Now consider events over time. An agent might have a utility function mapping possible streams of consumption into utility values. Here is one way this is often parameterized:
Define (b) as a constant discount rate known to the agent. It's a scalar that is between zero and one, and usually thought of as near one.
Define t as a time subscript that starts at zero and increases over the integers, either to some fixed T or to infinity.
Define c(t) as the amount the agent gets to consume at each t, and {c(t)} as the series of consumptions for all relevant t. c(t) is random here. its value is not known but its distribution is assumed known to the agent.
Let E[] be the expectations operator that takes means of distributions.

Using this notation a common dynamic version of exponential utility is:
u({ct} = the sum over all t of (b)tE[-e-ac(t)]

Whether this utility function describes observed investment decisions is discussable and testable. It is not often discussed, however. If clear information on that becomes known to this author, it will be added here.
Most uses of the exponential utility function in finance are driven by these aspects: (a) its analytic tractability; e.g. that it can be differentiated with respect to choice variables that affect future wealth w or consumption c(t); (b) for some applications it aggregates usefully, meaning that if every agent has this exact utility function and they can buy securities then a representative agent can be defined which also has this analytically convenient form and for whom the securities prices would be the same. It's convenient for computing securities prices in some abstract economies to use that representative agent. There are 'no wealth effects' -- that is, the amount of risky securities that the agent wants to hold is not a function of his own wealth, as long as he can borrow infinitely (which is often assumed for tractability in these models.)

Source: econterms


Goods and services that are produced in the home country and sold in other countries.

Source: EconPort

extended reals

Or, extendend real numbers, or extended real line. The set of reals plus the elements (infinity) and (minus infinity). Addition and multiplication can generally be extended to this set; see Royden, p. 36

Source: econterms

extensive margin

Refers to the range to which a resource is utilized or applied. Example: the number of hours worked by an employee. Contrast intensive margin.

Source: econterms

External validity

or criterion related validity is a type of validity that is assessed by the relationship between test scores and an independent, non-test criterion.

Source: SFB 504


An effect of a purchase or use decision by one set of parties on others who did not have a choice and whose interests were not taken into account.
Classic example of a negative externality: pollution, generated by some productive enterprise, and affecting others who had no choice and were probably not taken nto account.
Example of a positive externality: Purchase a car of a certain model increases demand and thus availability for mechanics who know that kind of car, which improves the situation for others owning that model.

Source: econterms


In a general sense, a technological externality is the indirect effect of a consumption activity or a production activity on the consumption or production possibilities available to some other consumer or producer. By the term indirect it is meant that the effect concerns an agent other than the one exerting this economic activity and that this effect does not work through the price system. This effect has first been analyzed by Pigou (1920).

In a non-cooperative game, the utility payoff to one player usually only depends on the profile of strategies taken by all the players, but not on the identity of the players that undertake certain actions or that face certain outcomes. If this is the case, the game is said to contain externalities. In an auction with externalities, for example, the final valuation for the object in sale of each bidder depends on the identity of the player who wins the auction and receives the object. The modification of payoffs due to externalities in a game can be thought of as an immediate consequence of the actions taken during the play of the game, as in production externalities, or as a reduced-form description of expected future interactions among the players (i.e., their equilibrium behavior) after the end of the game, as in the case of an auction with resale possibilities of the object obtained.

Source: SFB 504

Copyright 2006 Experimental Economics Center. All rights reserved. Send us feedback