Showing papers by "Instituto Tecnológico Autónomo de México published in 2004"

Intergenerational Persistence of Earnings: The Role of Early and College Education

Abstract: Recent empirical evidence from the United States indicates a high degree of persistence in earnings across generations. Designing effective public policies to increase social mobility requires identifying and measuring the major sources of persistence and inequality in earnings. We provide a quantitative model of intergenerational human capital transmission that focuses on three sources: innate ability, early education, and college education. We find that approximately one-half of the intergenerational correlation in earnings is accounted for by parental investment in education, in particular early education. We show that these results have important implications for education policy.

Consistent Estimation of Models Defined by Conditional Moment Restrictions

Abstract: In econometrics, models stated as conditional moment restrictions are typically estimated by means of the generalized method of moments (GMM). The GMM estimation procedure can render inconsistent estimates since the number of arbitrarily chosen instruments is finite. In fact, consistency of the GMM estimators relies on additional assumptions that imply unclear restrictions on the data generating process. This article introduces a new, simple and consistent estimation procedure for these models that is directly based on the definition of the conditional moments. The main feature of our procedure is its simplicity, since its implementation does not require the selection of any user-chosen number, and statistical inference is straightforward since the proposed estimator is asymptotically normal. In addition, we suggest an asymptotically efficient estimator constructed by carrying out one Newton–Raphson step in the direction of the efficient GMM estimator.

Monte Carlo Valuation of American Options through Computation of the Optimal Exercise Frontier

Abstract: This paper introduces a Monte Carlo simulation method for pricing multidimensional American options based on the computation of the optimal exercise frontier. We consider Bermudan options that can be exercised at a finite number of times and compute the optimal exercise frontier recursively. We show that for every date of possible exercise, any single point of the optimal exercise frontier is a fixed point of a simple algorithm. Once the frontier is computed, we use plain vanilla Monte Carlo simulation to price the option and obtain a low-biased estimator. We illustrate the method with applications to several types of options.

Circular distributions based on nonnegative trigonometric sums.

Abstract: A new family of distributions for circular random variables is proposed. It is based on nonnegative trigonometric sums and can be used to model data sets which present skewness and/or multimodality. In this family of distributions, the trigonometric moments are easily expressed in terms of the parameters of the distribution. The proposed family is applied to two data sets, one related with the directions taken by ants and the other with the directions taken by turtles, to compare their goodness of fit versus common distributions used in the literature.

Social Security Reform with Uninsurable Income Risk and Endogenous Borrowing Constraints

Abstract: We study the aggregate effects of a social security reform in a large overlapping generations model where markets are incomplete and households face uninsurable idiosyncratic income shocks. We depart from the previous literature by assuming that, because of lack of commitment in the credit market, the borrowing constraint in the unique asset is endogenously determined by the agents' incentives to default on previous debts. We find that a model with fixed borrowing constraints overestimates the positive effect of reforming social security on the capital stock and the saving rate, compared to our model with endogenous borrowing limit. The reason is that, in the latter, the size of precautionary savings is smaller because after the reform the incentives to default on previous debts are lower and consequently households face more relaxed borrowing limits. Adding retirement accounts to the basic model does not change these conclusions, although the quantitative importance of endogenizing borrowing constraints is reduced.

A Simple Test of Normality for Time Series

Abstract: This paper considers testing for normality for correlated data. The proposed test procedure employs the skewness-kurtosis test statistic, but studentized by standard error estimators that are consistent under serial dependence of the observations. The standard error estimators are sample versions of the asymptotic quantities that do not incorporate any downweighting, and, hence, no smoothing parameter is needed. Therefore, the main feature of our proposed test is its simplicity, because it does not require the selection of any user-chosen parameter such as a smoothing number or the order of an approximating model.We are very grateful to Don Andrews and two referees for useful comments and suggestions. We are especially thankful to a referee who provided a FORTRAN code. Lobato acknowledges financial support from Asociacion Mexicana de Cultura and from Consejo Nacional de Ciencia y Tecnologia (CONACYT) under project grant 41893-S. Velasco acknowledges financial support from Spanish Direccion General de Ensenanza Superior, BEC 2001-1270.

Valuation by Simulation of Contingent Claims with Multiple Early Exercise Opportunities

Abstract: This paper introduces the application of Monte Carlo simulation technology to the valuation of securities that contain many (buying or selling) rights, but for which a limited number can be exercised per period, and penalties if a minimum quantity is not exercised before maturity. These securities combine the characteristics of American options, with the additional constraint that only a few rights can be exercised per period and, therefore, their price depends also on the number of living rights (i.e., American-Asian-style payoffs), and forward securities. These securities give flexibility-of-delivery options and are common in energy markets (e.g., take-or-pay or swing options) and as real options (e.g., the development of a mine). First, we derive a series of properties for the price and the optimal exercise frontier of these securities. Second, we price them by simulation, extending the Ibanez and Zapatero (2004) method to this problem.

Normalized random measures driven by increasing additive processes

Abstract: This paper introduces and studies a new class of nonparametric prior distributions. Random probability distribution functions are constructed via normalization of random measures driven by increasing additive processes. In particular, we present results for the distribution of means under both prior and posterior conditions and, via the use of strategic latent variables, undertake a full Bayesian analysis. Our class of priors includes the well-known and widely used mixture of a Dirichlet process.

The Impact of Acquiring Control on Productivity

Abstract: Empirical studies on the importance of control rights on efficiency are hindered by actual - presumably efficient - ownership patterns. Finding settings where the right owner does not own the right asset and where ownership arbitrarily changes is challenging. In this paper I aim at overcoming these problems by investigating the elimination of foreign majority ownership restrictions in Mexico. Specifically, I study the performance of affiliates of multinational corporations for which (1) ownership restrictions appeared to bind before they were lifted, and (2) parent ownership increased from minority to majority as the reform was implemented. Using detailed plant-level information, I find that multinational control leads to large improvements in total factor productivity, particularly in industries that rely on technological innovations from their parent companies. Control is also associated with higher investment - particularly in technology intensive forms of production - and with an improvement in the skill profile of the labor force. Overall, I interpret the evidence as supportive of the property rights theory of the firm.

Seismic performance of laminated timber frames with fiber‐reinforced joints

Abstract: A series of dynamic experiments was performed on two-story glue-laminated timber frames. The tests included sinusoidal sweeps in one direction, arbitrary signals simulating earthquake loads in two directions, and harmonic free vibration at the fundamental frequency. Two experimental frames were manufactured and tested: (1) a control with horizontal laminations and no reinforcement at joint areas, and (2) a new frame design with densified material in the joint area that was further reinforced by glass-fiber composite material. Preliminary tests of scaled and full-size beam-to-column connections were performed to obtain connection characteristics needed for subsequent analytical modeling. Copyright © 2004 John Wiley & Sons, Ltd.

Valuation by Simulation of Contingent Claims with Multiple Early Exercise Opportunities

Abstract: This paper introduces the application of Monte Carlo simulation technology to the valuation of securities that contain many (buying or selling) rights, but for which a limited number can be exercised per period, and penalties if a minimum quantity is not exercised before maturity. These securities combine the characteristics of American options, with the additional constraint that only a few rights can be exercised per period and therefore their price depends also on the number of living rights (i.e., American-Asian-style payoffs), and forward securities. These securities give flexibility-of-delivery options and are common in energy markets (e.g., take-or-pay or swing options) and as real options (e.g., the development of a mine). First, we derive a series of properties for the price and the optimal exercise frontier of these securities. Second, we price them by simulation, extending the Ibanez and Zapatero (2004) method to this problem.

Classification of anatomical structures in mr brain images using fuzzy parameters

Abstract: We present an algorithm that automatically segments and classifies the brain structures in a set of magnetic resonance (MR) brain images using expert information contained in a small subset of the image set. The algorithm is intended to do the segmentation and classification tasks mimicking the way a human expert would reason. The algorithm uses a knowledge base taken from a small subset of semiautomatically classified images that is combined with a set of fuzzy indexes that capture the experience and expectation a human expert uses during recognition tasks. The fuzzy indexes are tissue specific and spatial specific, in order to consider the biological variations in the tissues and the acquisition inhomogeneities through the image set. The brain structures are segmented and classified one at a time. For each brain structure the algorithm needs one semiautomatically classified image and makes one pass through the image set. The algorithm uses low-level image processing techniques on a pixel basis for the segmentations, then validates or corrects the segmentations, and makes the final classification decision using higher level criteria measured by the set of fuzzy indexes. We use single-echo MR images because of their high volumetric resolution; but even though we are working with only one image per brain slice, we have multiple sources of information on each pixel: absolute and relative positions in the image, gray level value, statistics of the pixel and its three-dimensional neighborhood and relation to its counterpart pixels in adjacent images. We have validated our algorithm for ease of use and precision both with clinical experts and with measurable error indexes over a Brainweb simulated MR set.

Bayesian nonparametric survival analysis via Levy driven Markov processes

Abstract: In this paper we present and investigate a new class of nonparamet- ric priors for modelling a cumulative distribution function. We take F (t )=1 � exp{�Z(t)} ,w hereZ(t )= � t 0 x(s )d s is continuous and x(·) is a Markov process. This is in contrast to the widely used class of neutral to the right priors (Doksum (1974)) for which Z(·) is discrete and has independent increments. The Markov process allows the modelling of trends in Z(·), not possible with independent in- crements. We derive posterior distributions and present a full Bayesian analysis.

When the State is Untrustworthy: Public Finance and Private Banking in Porfirian Mexico

Abstract: All sovereign governments face a commitment problem: how can they promise to honor their own agreements? The standard solutions involve reputation or political institutions capable of tying the government's hands. Mexico's government in the 1880s used neither solution. It compensated its creditors by enabling them to extract rents from the rest of the economy. These rents came through special privileges over banking services and the right to administer federal taxes. Returns were extremely high: as long as the government refrained from confiscating all their assets (let alone repaying their debts) less than twice a decade, they would break even. There exists a commonly accepted view about why governments charter banks to serve as their exclusive financial agent. In that view, a government chooses to grant a single institution a monopoly over its borrowing because a monopolist can credibly punish defaults by denying future credit, whereas multiple creditors would be unable to sustain a boycott. The result is more credit for the government, and more security for its creditors. The Bank of England was born in such an arrangement.! We do not know much, however, about cases where the government makes such a deal and then reneges. Should its demand for credit rise, its time horizon fall, or new sources of credit become available, the government has incentives to borrow from third parties, or even default on its debts.2 If the government then defaults on its debts, how can it re

Nominal Wage Rigidities in Mexico: Evidence from Social Security Records

Abstract: This paper analyses the existence and extent of downward nominal wage rigidities in the Mexican labor market using data from the administrative records of the Mexican Social Security Institute (IMSS). This longitudinal, firm-level dataset allows us to track workers employed with the same firm, observe their wage profiles and calculate the nominal-wage changes they experience over time. Based on the estimated density functions of nominal wage changes and other moments of the distribution, we are able to calculate several standard tests of nominal wage rigidity that have been proposed in the literature. Furthermore, we extend these tests to take into account the presence of minimum wage laws that may affect the distribution of nominal wage changes. The densities and tests calculated using these date are similar to those obtained using administrative data from other countries, and constitute a significant improvement over the measures of nominal wage rigidities obtained from household survey data. We find considerably more wage rigidity than previous estimates obtained for Mexico using data from the National Urban Employment Survey suggest. Furthermore, we find evidence that the extent of nominal wage rigidities has been falling over time. We also document the importance of minimum wages in the Mexican labor market, as evidenced by the large fraction of minimum wage earners and the widespread indexation of wage changes to the minimum wage increases.

A model of endogenous political party platforms

Abstract: We develop a model of endogenous party platform formation in a multidimensional policy space. Party platforms depend on the composition of the parties’ primary electorate. The overall social outcome is taken to be a weighted average of party platforms and individuals vote strategically. Equilibrium is defined to obtain when no group of voters can shift the social outcome in its favor by deviating and the party platforms are consistent with their electorate. We provide sufficient conditions for existence of equilibria.

Gaussian kinetic model for granular gases.

Abstract: A kinetic model for the Boltzmann equation is proposed and explored as a practical means to investigate the properties of a dilute granular gas. It is shown that all spatially homogeneous initial distributions approach a universal ”homogeneous cooling solution” after a few collisions. The homogeneous cooling solution (HCS) is studied in some detail and the exact solution is compared with known results for the hard sphere Boltzmann equation. It is shown that all qualitative features of the HCS, including the nature of over population at large velocities, are reproduced semi-quantitatively by the kinetic model. It is also shown that all the transport coefficients are in excellent agreement with those from the Boltzmann equation. Also, the model is specialized to one having a velocity independent collision frequency and the resulting HCS and transport coefficients are compared to known results for the Maxwell Model. The potential of the model for the study of more complex spatially inhomogeneous states is discussed.

An approach to the non-normal behavior of hedge fund indices using Johnson distributions.

Abstract: In this paper we consider Johnson distributions as a tool to analyze and model the non normal behavior of hedge fund indices. After efiectively ob- taining the parameters for these distributions in the case of the CSFB /Tremont indices, we use the Omega parameter to test how e-ciently the Johnson distribu- tions re∞ect the presence of occasional large negative returns. We conclude that, although Johnson distributions yield improved models and conflrm that skewness and kurtosis are relevant in the evaluation of performance, they still fail to capture a substantial part of the tail risk, specially in those cases when we have high kurtosis.

Real Time Vision System for a Small Size League Team

Abstract: The RoboCup Small Size League (SSL) is a challenging environment for computer vision applications. Ball and robots move at speeds up to 2 m/s and the need of tracking them precisely requires the implementation of a high performance vision system. We will present the vision system developed in the ITAM Small Size Team (Eagle Knights) explaining the main algorithms and technical decisions made in the implementation. We pay special attention to the details necessary to give new teams a head start in th eir own implementation based on knowledge gained in two years of experience in SSL.

Composite index for the quantitative evaluation of image segmentation results

Abstract: Medical image segmentation is one of the most productive research areas in medical image processing. The goal of most new image segmentation algorithms is to achieve higher segmentation accuracy than existing algorithms. But the issue of quantitative, reproducible validation of segmentation results, and the questions: What is segmentation accuracy?, and: What segmentation accuracy can a segmentation algorithm achieve ? remain wide open. The creation of a validation framework is relevant and necessary for consistent and realistic comparisons of existing, new and future segmentation algorithms. An important component of a reproducible and quantitative validation framework for segmentation algorithms is a composite index that will measure segmentation performance at a variety of levels. We present a prototype composite index that includes the measurement of seven metrics on segmented image sets. We explain how the composite index is a more complete and robust representation of algorithmic performance than currently used indices that rate segmentation results using a single metric. Our proposed index can be read as an averaged global metric or as a series of algorithmic ratings that will allow the user to compare how an algorithm performs under many categories.

A simple and general test for white noise

Abstract: This article considers testing that a time series is uncorrelated when it possibly exhibits some form of dependence. Contrary to the currently employed tests that require selecting arbitrary user-chosen numbers to compute the associated tests statistics, we consider a test statistic that is very simple to use because it does not require any user chosen number and because its asymptotic null distribution is standard under general weak dependent conditions, and hence, asymptotic critical values are readily available. We consider the case of testing that the raw data is white noise, and also consider the case of applying the test to the residuals of an ARMA model. Finally, we also study finite sample performance