Abstract: This application analyzes the tax revenues for public waste disposal services from households using an economic and environmental assessment with Stata. This talk is based on Hanemann's (1984) approach using contingent valuation to estimate an indirect utility function; the proposal of this application is to reduce selection bias, avoiding the “willingness to pay quantity” using a proxy for a property tax and maintenance fees (ENIGH 2014). A contingent valuation model can be used as a tool for fiscal policies, in assessing environmental resources end ffects on morbidity or discomfort of families, and as an easy way to predict its utility with Stata.

Abstract: DJA stands for the Duclos-Jalbert-Araar (2003) decomposition of redistribution to vertical, horizontal, and reranking inequalities. The theoretical framework allows one to propose a method to decompose the redistribution effect or change in equality into these three components. The command is programmed as an ado-file in Stata; to be implemented, it requires the microdata of income or expenditures in combination with taxes and transfers or any other type of variables describing sources of inequities in a distribution of income. We provide an empirical application to explain its utility and the easy way to perform it, using data of income from the Mexican survey Encuesta Nacional de Ingresos y Gastos de los Hogares 2014.

Abstract: Fractional outcome models are especially designed for continuous dependent variables with values that range between zero and one. For example, fractional outcome models may be used when working with averages of binary outcomes, such as participation rates, or for variables on a zero-to-one scale, such as proportions and fractions. Stata 14 offers new commands for fractional outcome estimators, which fit models to these data using probit, logit, heteroskedastic probit, and beta regression. I will present basic concepts for these methods, and I will provide some examples.

Abstract: In this presentation, I discuss various exercises that include calculus of indicators, graphic facilities, estimation of a Mincerian econometric model and basic procedures to plot maps about household lack of access to food by Mexican states and municipalities, using Stata and Mata commands, from the original microdata of the Encuesta Intercensal 2015 (EIC, or the Intercensal Survey) conducted by the National Institute of Statistics and Geography (INEGI) in 2015. The syntax, matrix results, and templates that are presented show the versatility of Stata as an ideal tool in the management and analysis of large volumes of data with a focus on some basic assessment issues, statistics, and econometrics that require the use of real and recent information.

Abstract: Stata is an interesting alternative for financial analysts who have little or no experience in programming. Although Stata does not offer a great variety of user commands for financial models, it offers a great variety of econometric models that can be applied to any financial tools. Another advantage of using Stata for programming financial models is its script-based programming that makes it easier to learn for students or professionals with no programming experience. In this talk, a structured method for programming finance models will be presented. The design of this method for teaching finance programming in Stata is also presented. Specific examples of portfolio management models will be presented using the mvport package and an Excel interface.

Abstract: Waste management pollution is a public health concern. The aim of this study was to evaluate the relationship between systemic inflammation markers and exposure to endotoxin and (1→3)-β-D-glucan, present in particulate matter less than 10 micrometers (PM₁₀), in workers of a landfill facility (LF) and control populations of nonoccupationally exposed individuals living around the facility. Methods. After an environmental characterization, we conducted a cross-sectional study using Stata to evaluate inflammatory markers in 58 males between ages 18 and 40: 24 LF workers and 34 males living around the site. Interleukin 6 (IL-6) and 8 (IL-8), tumor necrosis factor-α (TNFα), white blood cell (WBC) count, percentages of lymphocytes, neutrophils and monocytes were analyzed with standardized methods in relation to those working in LF and living in downwind or upwind towns. Using Limulus Amebocyte Lysate (LAL), we assess endotoxin and (1→3)-β-D-glucan concentrations associated with PM₁₀. With regression models, adjusted by potential confounders, we found that IL-6 and neutrophils were significantly lower for LF workers compared with the upwind population, otherwise lymphocytes are higher. Lymphocytes, neutrophils, and monocytes have to do with endotoxin content in PM₁₀. Conclusions. We suppose that endotoxin content in PM₁₀ decreases immune response in landfill workers. This suggests that inflammation could be at other levels. It is important to use health and safety items during work and to study particle quantity induced by urban solid waste management.

Abstract: A considerable number of time series can be characterized by data-generating processes (DGP) that may be affected by particular events that lead to changes in the parameters. The new conditions for the DGP may remain in place for a period of time until the change is reversed to the previous state or until a new event leads to a new state, with the corresponding change in the parameters. In Stata 14, we introduce the mswitch command to model those kinds of time series by characterizing the transitions between unobserved states with a Markov chain. I will briefly introduce the basic concepts of Markov-switching models, and I will use a couple of examples with Mexican data to illustrate the implementation provided by mswitch.

Abstract: Marginal effects are commonly used to interpret linear and nonlinear regression models. Most simply, a marginal effect (ME) computes the change in the outcome for a fixed amount of change in one predictor while holding other predictors constant. This presentation considers a variety of nonstandard applications of MEs in a single model and compares effects across models. For a single model, MEs can be computed that allow proportional changes in a predictor, changes in multiple predictors that are mathematically linked (for example, polynomials, interactions), and changes in multiple variables that are substantively linked. When a predictor is the product of several variables, such as the BMI index, an ME can estimate the effect of changing one component of the predictor while holding other components constant. If odds ratios are viewed as MEs, one can compute odds ratios when a variable is included as a polynomial or when nonlogit models (for example, probit) are used. To compare effects across models, one can use MEs when comparing regression coefficients is inappropriate or misleading. For example, regression coefficients from logit models should not be used to compare the effects of a predictor across groups, but MEs can be compared. Or while comparing regression coefficients across nested models is a common method of interpretation in linear models, it is misleading in nonlinear models where the comparison of MEs is preferred. Each of these applications of MEs is explored using the Stata commands margins, lincom, nlcom, and suest, along with several SPost commands. For simplicity, models for binary outcomes are used, but the methods apply generally to other regression models.

Abstract: In this talk, I will discuss how to implement maximum likelihood and GMM estimators using Mata's main optimization engine: moptimize. As examples, I implement the linear regression model by maximum likelihood and a two-equation system with endogenous variables by GMM.

Abstract: The analysis on stationarity of time series and the simultaneous possibility of structural change is an easy task to deal with Stata. Programming proposals in this work are based on Bai and Perron (2003) and Gómez-Zaldivar and Ventosa-Santaulària (2010). Four possible scenarios are considered: Divergence, catching-up (lagging behind), loose catching-up (loose lagging-behind), and convergence with structural changes. The use of Stata commands for dealing with convergence problems of gross domestic product per capita in Mexico and income inequality is shown. The sub-modules of a larger command that includes several additional tests for stationarity and structural change are applied as well.

Abstract: Tobacco consumption in adolescence is a public health priority as well as an important risk that contributes substantially to the growing epidemic of nontransmissible diseases. Using Stata and corresponding commands, we estimate the factors associated with consumption of tobacco in adolescents for a town from Morelos, Mexico, studying those with a high school level of education. It is a cross-sectional observational epidemiological study. The research was conducted on 269 students during the 2014–2015 school year. A self-applied detailed sociodemographic database of characteristics and consumption of tobacco questionnaire is used in the calculations. We found that factors associated with consumption of tobacco are eased if a friend offers smoking a cigar (OR = 4. 98 95% CI 1.9–12.95), alcohol consumption (OR = 2.51 95% CI 1.01–6.22), if exposed to smoking tobacco in public places (OR = 2.44 95% CI 1.02–5.83), only those with smoking friends is weakly significant (OR = 2.81 95% CI 0.94–8.38) and those who remain smoking during the next 12 months (OR = 95% CI 0.93–6.28 2.41).