EM Algorithm Estimation of Simultaneous Equation Model with Limited Variables: An Example of Cigarette Consumption

Marginal Effects and Significance Testing with Heckman's Sample Selection Model: A Methodological Note

This article illustrates two techniques for calculating the statistical significance of the marginal effects derived from Heckman's sample selection model, an increasingly common econometric specification in economics and political science. The discussion draws on an analysis by Sweeney (2003) of the incidence and intensity of interstate disputes. After replicating his results, the article presents the delta method and the nonparametric bootstrap as alternative techniques for obtaining SEs of the marginal effects, which themselves are calculated from a transformation of the model parameters. The analysis reveals two variables for which misleading inferences are drawn with respect to the precision of the estimated coefficients in the original study, suggesting that significance testing of the derived marginal effects is warranted.

On Interaction Effects: The Case of Heckit and Two-Part Models

The linear regression model is a statistical tool used to model the causal relationship of a dependent variable based on one or several independent or explanatory variables. In scenarios where the dependent variable is a censored variable and there is potential to exist sample selection, the sample selection model can be an alternative in analyzing this relationship. In the Heckman sample selection model, independent variables have the possibility of having an endogeneity effect, where they should be treated as endogenous variables in both the outcome equation and the selection equation instead of as exogenous variables. In result, by including endogenous covariates in the Heckman sample selection model, the sample selection model equation will have more than one equation and makes it a simultaneous equation. To estimate simultaneous equations, simple estimation methods such as the maximum likelihood estimator method are no longer appropriate. In this study, we will discuss the estimation of sample selection models with endogenous covariates utilizing the full information maximum estimator (FIML) approach. The sample selection model with endogenous covariates was then applied to the women labor supply data of Tomas Mroz's research and compared with several models. Based on the MSE and SSE values obtained from the linear regression model, Tobit regression model, Heckman sample selection model, and sample selection model with endogenous covariates, it was concluded that the Heckman sample selection model is the best model that fit the dataset since it yields the best results with the smallest MSE and SSE values

Using data from a survey sample of people 65 years of age and older living in Seoul and Chuncheon, Korea, this paper assesses whether the level of social capital affects elderly individuals' use of medical care. As an econometric model, Heckman's Sample Selection model and the 2SLS method were used to control the endogeneity problem of patient's trust on doctors. The results of our estimations indicate that the level of social capital exerts a positive effect on elderly individuals' use of medical care indirectly, via its positive effect on the level of trust in doctors.

Market Discipline and the Russian Interbank Market

With a focus on individual motorists in car-owning households in Germany, this analysis econometrically investigates the determinants of automobile travel for nonwork service activities against the backdrop of two questions: (a) Does gender play a role in determining the probability of car use and the distance driven? and (b) If so, how is this role mitigated or exacerbated by other socioeconomic attributes of the individual and the household in which he or she resides? Drawing on a panel of data collected between 1996 and 2003, Heckman's sample selection model is specified to control for biases that otherwise could arise from the existence of unobservable variables that determine both the discrete and the continuous choices pertaining to car use. The results indicate that although women, on average, undertake more nonwork travel than men, they undertake less such travel by car, implying a greater reliance on other modes. Moreover, employment status, age, the number of children, automobile availability, and the proximity to public transit are all found to have significantly different effects on the probability of nonwork car travel between men and women but—with the exception of automobile availability—not on the distance driven. Taken together, these results suggest that policies targeted at reducing automobile dependency and associated negative externalities, such as congestion, are unlikely to have uniform effects across the sexes. These findings have implications for both policy evaluation and travel demand forecasting.

Market Discipline and the Russian Interbank Market

PurposeFarmer organizations (FOs) can elevate many of the production- and marketing-related challenges by ensuring access to technology, farming information and loan for inputs and mechanization. This study assesses the major factors that affect the participation in FOs by the smallholder rice farmers in Bangladesh and evaluates the impacts of the participation on rice yield and production efficiency.Design/methodology/approachThe present study used primary data collected from 1,000 smallholder rice farmers in northwest Bangladesh, consisting of 250 farmers those participated in an organization. This study utilized a sample selection stochastic production frontier (SPF) method, a combination of the conventional SPF and the Heckman's sample selection model, to control for biases stemming from observed and unobserved attributes.FindingsThis study demonstrates that participation in an organization is significantly affected by smallholder rice farmers' education, occupation, family size, household income, land ownership and the location where they reside. At the same time, the participation status affects the productivity of smallholder farmers. Findings indicate that farmers who participated in an organization had higher rice yield (11% more) and were technically more efficient (1.4 percentage points higher) compared to farmers who did not participate.Research limitations/implicationsSince this study was carried out with representative sampled farmers from northwest Bangladesh, the findings may not represent all farmers' perceptions of FOs in the country.Originality/valueEven though more than 200,000 FOs are currently in operation, knowledge regarding the effectiveness of Bangladesh's FOs is limited. Notably, this study used a relatively new method, sample selection SPF model, to investigate the impact of FOs on the production efficiency of smallholder rice farmers in northwest Bangladesh. Therefore, this study contributes to the literature in elucidating the factors affecting participation in FOs and its impact on rice yield and efficiency. Since FOs have been somewhat ineffective in their role as service providers in Bangladesh, this study’s results will guide policymakers to intervene more successfully regarding the changes needed.

Selling hunting access may supplement household income, yet only a small proportion of nonindustrial private (NIP) landowners in United States lease the right to hunt their land. Based on a survey of Mississippi landowners, the decision to lease hunting rights and factors influencing lease revenue per leased hectare were analyzed. The two issues were jointly modeled consistent with Heckman's sample selection model, and the lease revenue was specified in accordance with hedonic pricing theory. Empirical results showed that landowner concerns about loss of privacy, accident liability, and conflicts with personal use of land reduced the likelihood of leasing; total landownership and specific landowner characteristics increased it. With regards to factors explaining differences in lease revenue per leased hectare, bottomland hardwoods commanded a greater premium than many other land uses. In addition, lease revenue per hectare was distinctly higher where a landowner had expertise in managing a hunting lease enterprise. These findings have implications for landowners interested in managing wildlife-associated enterprises and public agencies engaged in the provision of natural resource-based recreation.

Heckman's sample selection model is used to examine the role of education on household purchase decisions and expenditures of tobacco products in Malaysia. Results of the marginal effects of education, segmented by ethnic and gender groups, suggest that education decreases the probability, conditional levels and unconditional levels of tobacco expenditures amongst Malaysian households. Specifically, an additional year of education of the household head, irrespective of ethnic or gender considerations, decreases smoking probability by 1.5 percent. However, the negative effect of education seems to be higher for Chinese (US$1.07) than Malay (US$0.26) households in terms of conditional expenditures. Furthermore, education significantly decreases conditional tobacco expenditures within male‐headed households.

The purpose of this paper is to examine the factors affecting the likelihood of consuming and the amount spent on alcoholic beverages in Thailand. Heckman's sample selection model is applied to data from the 2009 Socio‐Economic Survey of Thailand. Results suggest that household size, tenure and occupation have significant impacts on both the probability of alcohol consumption and spending levels. Income also plays a key role in explaining the amount spent on alcoholic beverages. Demand elasticities are calculated under the Extended Linear Expenditure System. Demand for alcoholic beverages is found to be inelastic for Thai drinkers.

BackgroundChina's welfare system including social health insurance has been closely linked to its unique household registration system, despite high population mobility over the past few decades. This study aimed to determine the pattern of health insurance usage from internal migrants in mainland China for hospital care.MethodsData were extracted from the 2018 China Migrants Dynamic Survey. The respondents who enrolled in a social health insurance program and reported illness or injury over the past year were eligible for this study (n = 15,302). Two groups of outcome indicators were calculated assessing the use (incidence and settlement location) of insurance funds for hospital care and the burden of hospital expenditure (total hospital expenditure, out-of-pocket payments, and share of insurance reimbursement), respectively. Logit regression and Heckman's sample selection models were established to determine the predictors of insurance fund usage and the burden of hospital expenditure, respectively.ResultsMost respondents enrolled in a social health insurance program outside of their residential location (70.72%). About 28.90% were admitted to a hospital over the past year. Of those hospitalized, 72.98% were admitted to a hospital at their migration destination, and 69.96% obtained reimbursement from health insurance, covering on average 47% of total hospital expenditure. Those who had a local insurance fund aligned with residency (AOR = 2.642, 95% CI = 2.108–3.310, p < 0.001) and enrolled in employment-based insurance (AOR = 1.761, 95% CI = 1.348–2.301, p < 0.001) were more likely to use insurance funds for hospital care, and paid less out-of-pocket (β = −0.183 for local funds, p = 0.017; β = −0.171 for employment-based insurance, p = 0.005) than others. A higher share of insurance reimbursement as a proportion of hospital expenditure was found in the employment-based insurance enrollees (β = 0.147, p < 0.001). Insurance claim settlement at the residential location was associated with lower total hospital expenditure (β = −0.126, p = 0.012) and out-of-pocket payments (β = −0.262, p < 0.001), and higher share of insurance reimbursement (β = 0.066, p < 0.001) for hospital expenditure.ConclusionLow levels of health insurance benefits for hospital care are evident for internal migrants in mainland China, which are associated with the funding arrangements linked to household registration and inequality across different funds.

This paper presents a mathematical model used to determine jointly a worker's decision to participate in a nonhome and nonwork activity and the decision on how long to participate. With the household interview survey data from the New York City area, the workers’ participation and duration decisions were estimated for each of five periods in a worker's day: before morning commute, morning commute, midday, evening commute, and after evening commute. To account for the censored nature of the duration data (i.e., a large number of observations clustered at zero), Heckman's sample selection model was used together with the full information maximum likelihood estimation method. To enhance the behavioral basis for the models, extensive statistical tests were given to the model specifications and the assumptions underlying the model structure. The empirical results provide useful insights into the effects of socio-demographics, land use–transportation measures, and activity duration characteristics on workers’ daily scheduling of nonwork activities and travel in a highly urbanized environment. This study also provides exploratory methodologic evidence that could lead to an approach for predicting the change in a worker's nonwork activity patterns (participation and duration) as a result of changes in future demographic conditions and land use–transportation scenarios.

Our analysis, based on the ICRISAT panel survey of villages in the semi-arid region of south India, confirms the income stabilizing effect of the Employment Guarantee Scheme in India. Variability of household income is measured by an unconditional variance of residuals of an income equation. A (variant) of Heckman's sample selection model is employed to allow for the endogeneity of EGS participation and to assess its income stabilizing role. The (instrumented) EGS participation reduces the residual variance of household income, implying consequent income stabilization.