1,354,858 research outputs found
Why is Consumption More Log Normal Than Income? Gibrat's Law Revisited
No full textSignificant departures from log normality are observed in income data, in violation of Gibrat’s law. We show empirically that the distribution of consumption expenditures across households is, within cohorts,
closer to log normal than the distribution of income. We explain this empirical result by showing that the logic of Gibrat’s law applies not to total income, but to permanent income and to marginal utility
THE INVERSE LEWBEL DEMAND SYSTEM
No full textA new model of consumer preferences is introduced. It is appropriate for modeling perishable commodities which are produced with a lag, where it is reasonable to assume the market-level quantities are fixed by previously made production decisions. The inverse Lewbel system, as it is called, is a flexible nonlinear system of share equations, which nests two other inverse demand systems, the direct translog and the inverse AIDS. Thus, the inverse Lewbel may be employed to test whether these more restrictive preference structures are appropriate. In an application to quarterly U.S. meat consumption, the more restrictive structures are rejected.Consumer/Household Economics,
Nonparametric errors in variables models with measurement errors on both sides of the equation
No full textMeasurement errors are often correlated, as in surveys where respondent's biases or tendencies to err affect multiple reported variables. We extend Schennach (2007) to identify moments of the conditional distribution of a true Y given a true X when both are measured with error, the measurement errors in Y and X are correlated, and the true unknown model of Y given X has nonseparable model errors. After showing nonparametric identification, we provide a sieve generalized method of moments based estimator of the model, and apply it to nonparametric Engel curve estimation. In our application measurement errors on the expenditures of a good Y are by construction correlated with measurement errors in total expenditures X. This problem, which is present in many consumption data sets, has been ignored in most demand applications. We find that accounting for this problem casts doubt on Hildenbrand's (1994) "increasing dispersion" assumption
THE INVERSE LEWBEL DEMAND SYSTEM
No full textA new model of consumer preferences is introduced. It is appropriate for modeling perishable commodities which are produced with a lag, where it is reasonable to assume the market-level quantities are fixed by previously made production decisions. The inverse Lewbel system, as it is called, is a flexible nonlinear system of share equations, which nests two other inverse demand systems, the direct translog and the inverse AIDS. Thus, the inverse Lewbel may be employed to test whether these more restrictive preference structures are appropriate. In an application to quarterly U.S. meat consumption, the more restrictive structures are rejected
LATE with Missing or Mismeasured Treatment
No full textWe provide a new estimator, MR-LATE, that consistently estimates local average treatment effects when treatment is missing for some observations, not at random. If instead treatment is mismeasured for some observations, then MR-LATE usually has less bias than the standard LATE estimator. We discuss potential applications where an endogenous binary treatment may be unobserved or mismeasured. We apply MR-LATE to study the impact of women’s control over household resources on health outcomes in Indian families. This application illustrates the use of MR-LATE when treatment is estimated rather than observed. In these situations, treatment mismeasurement may arise from model misspecification and estimation errors
Supplement for “Identification and Estimation of Games with Incomplete Information Using Excluded Regressors"
No full textThis online supplement to Lewbel and Tang (2014) provides proofs and additional results. Section A provides results regarding existence of required sets of states and assumed support conditions. Section B provides further discussion and formal derivations of our asymptotic results, and Section C gives the proof of Theorem 3. A. N.D.S. Sets and Large Support Conditions This section provides further primitive conditions on model elements that are suffi cient for some of the identifying conditions in the text. These include the existence of a non-degenerate and non-singular (n.d.s.) set for a given ˜x; and the large support condition in A4 and A4’. A.1. Suffi cient condition for the existence of n.d.s. sets In Section 3.2 of Lewbel and Tang (2014), we give conditions under which there exists a n.d.s. set for a generic vector of non-excluded regressors ˜x. Here in this section we give stronger conditions that ensure there exists a n.d.s. set ω where the equilibrium is unique. This is useful for the identification strategy discussed in Section 7.1 of Lewbel and Tang (2014). Consider a game with N = 2 (with players denoted by 1 and 2) and hi(Dj) = Dj for i = 1, 2 and j = 3−i. To begin with, we give conditions under which the game admits a unique equilibrium at a generic state x. Let p ≡ (p1, p2) and define
Measurement Error Models
No full textThis chapter discusses the use of instrumental variables for dealing with measurement
error in regression covariates. Instruments are defined as observed
variables that correlate with the mismeasured covariates, but do not correlate
with the measurement error and with the model error. The approaches considered
here are useful when little is known about the distribution of the measurement
error and when validation data on the mismeasured covariates are not available.
This chapter discusses the usefulness of instruments to account for measurement
error in the case of linear, nonlinear, and nonparametric regression models. This
chapter additionally discusses how to use instruments in empirical work when the
assumption of classical measurement error is violate
Latent separability and price variation in the estimation of demand System
No full textThe aim of this paper is to overcome the problems caused by insufficient price variation in estimating a large demand system. For that, we propose a new form of Stone-Lewbel (SL) cross section prices developed under latent separability that explore individual specific variation in the composition of the bundles of exclusive goods. The estimation of demand system under latent separability needs the choice of at least one exclusive good per group. We estimate Quadratic Almost Ideal Demand System (QAIDS) under weak and latent separability using traditional aggregate price indices and SL prices. Our empirical analysis is based on fifteen non durable goods of a Tunisian Family Expenditure Survey Data. The results show greater differences among effects price and estimates of price elasticities obtained under weak separability and latent separability using both traditional price indices and SL prices. We obtain higher precision of estimates of own price elasticities using SL prices under latent separability.Latent separability, Weak separability, Demand system, Exclusive goods, Price variation, SL prices
The Inverse Lewbel Demand System
No full textLewbel (1989) offered a demand model which nested both the indirect translog (ITL) of Christensen, Jorgensen, and Lau (1977) and almost ideal demand system (AIDS) of Deaton and Muellbauer (1980a, 1980b). It has the advantage, then of allowing the applied demand analyst to test the restrictions which imply the ITL and AIDS models directly. In terms of parametric analysis of demand, the increased generality of Lewbel's demand system should minimize the impact of maintained hypotheses on the outcome of the statistical analysis. All of these models have appealing theoretical properties, they correspond to a well defined preference structures, which is convenient for welfare analysis. These so-called PIGLOL preferences also have the property of consistent aggregation from the micro to the market level, while allowing nonlinear Engel curves. Second, the functional form of the preferences is \"flexible\" in that it can be thought of as a local second order approximation to an unknown preference structure. Third, homogeneity and symmetry restrictions depend only on estimated parameters and so are easily imposed and/or tested. There are commodities for which the assumption of predetermined prices at the market level may not be viable. Some of the earliest applied work in demand for agricultural products took current supplies as fixed and therefore specified ad hoc inverse demand curves for statistical evaluation. This alternative aggregation story is still employed, especially by those building market models, such as Freebairn and Rausser (1975) and Arzac and Wilkinson (1979). So, for example, if modeling demand for a perishabl commodity, the production of which is subject to long biological lags, the researcher might employ inverse demands. Production lags prevent market-level supply response, while perishability requires the commodity be consumed. Thus, price must adjust. Not all previous studies which have employed inverse demand structures have proceeded in an ad hoc manner. Heien, and Chambers and McConnell developed separable inverse demand systems and applied them to food commodities. Barten and Bettendorf developed an inverse Rotterdam system and applied it to the demand for fish. Christensen, et al. develop the direct translog demand system (as well as the indirect system). Both they and Jorgenson and Lau use the direct translog demand system to test demand restrictions. Huang used the theoretical development of Anderson and the distance function to generate a system of inverse demands, which were applied to composite food and nonfood commodities. Eales and Unnevehr also employed a particular distance function to develop an inverse AIDS model. In the sections that follow a model which nests both the direct translog (DTDS) and the inverse almost ideal demand system (IAIDS) is developed. This system will be referred to as the inverse Lewbel (ILDS). These three demand systems are then compared and contrasted and used to model Canadian demand for meat
Advice on using heteroskedasticity-based identification
No full textLewbel (2012, Journal of Business and Economic Statistics 30: 67–80) provides a heteroskedasticity-based estimator for linear regression models containing an endogenous regressor when no external instruments or other such information is available. The estimator is implemented in the command ivreg2h by Baum and Schaffer (2012, Statistical Software Components S457555, Department of Economics, Boston College). In this article, we give advice and instructions to researchers who want to use this estimator
- …
