Inverse Mills Ratio Research Articles

Determinants of FDI in BRICS Countries: Panel Data Approach

Abstract We empirically investigates the factors that affect Foreign Direct Investment (FDI) inflows in five BRICS countries for the period 1990–2015. We address the selection bias and unobserved heterogeneity by estimating a panel Heckman selection method and attempts to account for both selection and endogeneity within the new two-stage method. After addressing the above mentioned econometric issues, the infrastructure and GDP per capita variables under the new two-stage method remain positive and significantly similar to the coefficient of infrastructure and GDP per capita under the panel Heckman selection model. In addition, the inverse Mills ratio maintain its level of statistical significance, confirming the presence of both sample selection bias and endogeneity.

International Real Estate Review

The purpose of this paper is to estimate the impact of tenure choice on saving for Chinese households. Since housing expenditure usually accounts for a large portion of household spending, households need to consider how much they have in order to save for future housing spending when they decide on their daily expenditures, including food, clothing, transportation, education, leisure, and so on and so forth. We estimate the tenure choice behavior of Chinese households by first, applying a data set from the China Household Finance Survey and separating households into three types of tenure choice, namely, renters, owner-occupied with a mortgage, and owner-occupied without a mortgage. Then, we estimate the actual impact of tenure choice on saving by applying the Heckman two-stage model. Our estimations show two important results. First, the coefficients of the inverse Mills ratio are significant which implies that a two-stage estimation model is appropriate. Secondly, the estimated coefficients for the factors that affect saving behavior under the two-stage model are significantly different from those under a conventional model. The result shows that, without considering tenure choice, the conventional method of estimation for factors that affect saving behavior will be biased.

Exact Bounds on the Inverse Mills Ratio and Its Derivatives

The inverse Mills ratio is $$R:=\varphi /\Psi $$ , where $$\varphi $$ and $$\Psi $$ are, respectively, the probability density function and the tail function of the standard normal distribution. Exact bounds on R(z) for complex z with $$\mathfrak {R}z\geqslant 0$$ are obtained, which then yield logarithmically exact upper bounds on high-order derivatives of R. These results complement the many known bounds on the (inverse) Mills ratio of the real argument. The main idea of the proof is a non-asymptotic version of the so-called stationary-phase method. This study was prompted by a recently discovered alternative to the Euler–Maclaurin formula.

Health IT and inappropriate utilization of outpatient imaging: A cross-sectional study of U.S. hospitals

ObjectiveTo determine whether the use of information technology (IT), measured by Meaningful Use capability, is associated with lower rates of inappropriate utilization of imaging services in hospital outpatient settings. Research designA retrospective cross-sectional analysis of 3332 nonfederal U.S. hospitals using data from: Hospital Compare (2011 outpatient imaging efficiency measures), HIMSS Analytics (2009 health IT), and Health Indicator Warehouse (market characteristics). Hospitals were categorized for their health IT infrastructure including EHR Stage-1 capability, and three advanced imaging functionalities/systems including integrated picture archiving and communication system, Web-based image distribution, and clinical decision support (CDS) with physician pathways. Three imaging efficiency measures suggesting inappropriate utilization during 2011 included: percentage of “combined” (with and without contrast) computed tomography (CT) studies out of all CT studies for abdomen and chest respectively, and percentage of magnetic resonance imaging (MRI) studies of lumbar spine without antecedent conservative therapy within 60days. For each measure, three separate regression models (GLM with gamma-log link function, and denominator of imaging measure as exposure) were estimated adjusting for hospital characteristics, market characteristics, and state fixed effects. Additionally, Heckman’s Inverse Mills Ratio and propensity for Stage-1 EHR capability were used to account for selection bias. Principal findingsWe find support for association of each of the four health IT capabilities with inappropriate utilization rates of one or more imaging modality. Stage-1 EHR capability is associated with lower inappropriate utilization rates for chest CT (incidence rate ratio IRR=0.72, p-value <0.01) and lumbar MRI (IRR=0.87, p-value <0.05). Integrated PACS is associated with lower inappropriate utilization rate of abdomen CT (IRR=0.84, p-value <0.05). Imaging distribution over Web capability is associated with lower inappropriate utilization rates for chest CT (IRR=0.66, p-value <0.05) and lumbar MRI (IRR=0.86, p-value <0.05). CDS with physician pathways is associated with lower inappropriate utilization rates for abdomen CT (IRR=0.87, p-value <0.01) and lumbar MRI (IRR=0.90, p-value <0.05). All other cases showed no association. ConclusionsThe study offers mixed results. Taken together, the results suggest that the use of Stage-1 Meaningful Use capable EHR systems along with advanced imaging related functionalities could have a beneficial impact on reducing some of the inappropriate utilization of outpatient imaging.

Measuring respondent uncertainty in discrete choice experiments via utility suppression

Discrete choice experiments (DCEs) are an important methodology in survey research. Although DCEs assume that respondents know their true preferences and can make choices precisely, in reality, respondents may be uncertain. Respondent uncertainty has long been recognized as an issue in the DCE context, but unfortunately, its explicit modelling has received only limited attention. This paper proposes using hierarchical Bayes multinomial probit models to construct respondent uncertainty measures based on the utility difference and a concept called utility suppression, which is quantified by the derivative of the inverse Mills ratio. Two empirical studies with different DCE design formats are conducted to assess and compare the performance. The first includes a no-choice option, and the second is forced-choice formatted with a simpler design. The hold-out validation shows that the utility difference and our proposed measure, which are consistent with the random utility maximization assumption, have better overall performance, particularly in identifying the high-uncertainty respondents. We further show how the information about respondent uncertainty can be utilized in practice.

Seller ethnicity and property characteristics: foreclosure sales in Malaysia

PurposeThe purpose of this paper is to examine the determinants of property sale probability and sale price at auction in Kuala Lumpur, Malaysia.Design/methodology/approachA two-stage Heckman sample selection model is used for this research. The first stage involves the estimation of a probit model on a successful sale. The second stage introduces an additional selection variable, the inverse Mills ratio, as an explanatory variable to the sale price estimation equation.FindingsThe authors find that Chinese-owned auctioned properties have higher sale probability and are sold at higher prices as compared to Indian and Malay-auctioned properties. Properties auctioned by the largest auction house outperformed other smaller auctioneers. Auction characteristics such as proximity to the city center, number of previous auction attempts and number of online viewers are positively related to sale price and sale probability.Social implicationsThe findings on the substantially lower sale price obtained by Malay and Indian borrowers compared to their Chinese counterparts imply that it is much harder for these borrowers to be relieved from financial distress. The two plausible explanations offered in this paper for this price differential, i.e. racial residential segregation and ownership restriction, warrant further study.Originality/valueFirst, the authors consider the explanatory power of seller ethnicity, number of online viewers and auctioning route which are new to the literature. Second, they use a Heckman model that addresses possible selection bias of sold properties. This methodology is unexplored in the auction literature.

A general model for treatment of protests and no-answer responses in contingent valuation method

This study formulates a general model to account for the protest responses and no-answer replies. The distinct characteristics of this model are general in three ways: simultaneously accounting for protest and no-answer responses, applicable to all kinds of elicitation formats in contingent valuation, and simplicity in estimation. Creation of inverse Mills ratio is the distinctive step in this general model. The inverse Mills ratio is continuously carried in the subsequent estimation for the modification of different types of elicitation formats in this general model. The results generally indicate that these ratios are significantly different from zero. This means that accounting for these Mills ratios does have an important role in such modification when protest responses and/or no-answer responses are both taken into account. The results show that overall total willingness to pay from the general model with inclusion of protest and no-answer responses under different types of elicitation formats are higher than those estimated by traditional treatment. The degree of underestimation of traditional treatment is between 26% and 67%. That is, the general model proposed here for treating protest and/or no-answer responses in contingent valuation method can account for the full information, which might be potentially omitted or inappropriately dealt with in the estimation.

Clawback provision adoption, corporate governance, and investment decisions

AbstractWe examine the effect of corporate governance on the likelihood of clawback provision adoption, and its consequences in terms of corporate investment practices and risk‐taking behavior. We find that firms with strong governance (as proxied by board independence, diligence, and size) are positively associated with the firm's adoption of a clawback provision; whereas firms with weak governance (as proxied by management entrenchment, i.e., CEO duality status and tenure) are negatively associated with clawback provision adoption. Using the propensity‐score matching, difference‐in‐differences research design, and inverse Mills ratio to mitigate omitted variables and self‐selection biases, we find that after adopting a clawback provision, firms’ abnormal investment decreases and the firms’ investments are less risky.

Price elasticity of the demand for soft drinks, other sugar-sweetened beverages and energy dense food in Chile

BackgroundChile is the second world’s largest per capita consumer of caloric beverages. Caloric beverages are associated with overweight, obesity and other chronic diseases. The objective of this study is to estimate the price elasticity of demand for soft drinks, other sugar-sweetened beverages and high-energy dense foods in urban areas in Chile in order to evaluate the potential response of households’ consumption to changes in prices.MethodsWe used microdata from the VII Family Budget Survey 2012–2013, which collects information on expenditures made by Chilean urban households on items such as beverages and foods. We estimated a Linear Approximation of an Almost Ideal Demand System Model to derive own and cross price elasticities of milk, coffee, tea and other infusions, plain water, soft drinks, other flavored beverages, sweet snacks, sugar and honey, and desserts. We considered the censored nature of the data and included the Inverse Mills Ratio in each equation of the demand system. We estimated a Quadratic Almost Ideal Demand System and a two-part model as sensitivity analysis.ResultsWe found an own price-elasticity of −1.37 for soft drinks. This implies that a price increase of 10% is associated with a reduction in consumption of 13.7%. We found that the rest of food and beverages included in the demand system behave as substitutes for soft drinks. For instance, plain water showed a cross-price elasticity of 0.63: a 10% increase in price of soft drinks could lead to an increase of 6.3% of plain water. Own and cross price elasticities were similar between models.ConclusionsThe demand of soft drinks is price sensitive among Chilean households. An incentive system such as subsidies to non-sweetened beverages and tax to soft drinks could lead to increases in the substitutions for other healthier beverages.

An investigation into the effect of alcohol consumption on health status and health care utilization in Ireland

This paper presents a study of the effect of alcohol consumption on individual health status and health care utilization in Ireland using the 2007 Slán National Health and Lifestyle Survey, while accounting for the endogenous relationship between alcohol and health. Drinkers are categorized as those who never drank, non-drinkers, moderate drinkers, or heavy drinkers, based on national recommended weekly drinking levels in Ireland. The drinking-status equation is estimated using an ordered probit model. Predicted values for the inverse mills ratio are generated, which are then included in the health and health-care utilization equations. Differences in health status for each category of drinker are examined, and the relationship between both alcohol consumption and health with a host of other personal and socio-economic variables is also identified. Given that the measure of health status available is self-assessed, the effect of alcohol consumption on health-care utilization is also analyzed as an alternative measure of health. Findings show that in Ireland, moderate drinkers enjoy the best health status. More moderate drinkers report having very good or excellent health compared with heavy drinkers, non-drinkers, or those who never drank. While heavy drinkers do not report having as good a health status as moderate drinkers, they are better off in terms of health when compared with non-drinkers and those who are lifetime abstainers.

Impact Analysis of Small Scale Irrigation Schemes on Household Food Security the Case of Sibu Sire District in Western Oromia, Ethiopia

This study was aimed at examining analysing the impact of small scale irrigation on household food security in Sibu Sire district of Western Oromia regional state. Data were collected from 105 irrigation users and 45 non-users. Multi-stage stratified random sampling was applied to select the respondents. Descriptive statistics, Heckman twostage model and Greer and Thorbecke methods were applied to analyse the data. And the analysis research result implies that 44% of the non-users and 27% of users were food insecure. The result was also revealed that 56% of nonusers and 73% of the users were food secured. The model’s first stage revealed the significance effects of distance from the water source, size of cultivated land, access to credit service, access to extension, livestock holding and soil fertility factors in irrigation utilization. The second stage of the model discovered that access to irrigation, household size, dependency ratio, size of cultivated land, proximity of the households to a water source, distance from the market, livestock holding, crop pest infestation, and the Inverse Mills ratio are the determinants of household food security. The study analysis concluded that small scale irrigation utilization is one of the viable solutions to secure household food needs in the study area.

Private information and limitations of Heckman's estimator in banking and corporate finance research

Private information is a common problem in banking and corporate finance research. Heckman's (1979) two-step estimator is commonly used to test for sample selection using a simple t-test on the inverse Mills ratio (IMR) coefficient. Following Puri (1996), this test is often interpreted as a test for private information. We conduct a series of Monte Carlo simulations to show that researchers can reliably use the Heckman estimator to test for private information when this private information is random. However, private information often takes the form of an omitted variable with a deterministic relationship to selection and outcomes. In this case, we show that the IMR coefficient is biased and inconsistent and that t-tests lead to incorrect conclusions regarding the significance of private information as well as its impact on selection and outcomes. We illustrate our results using a unique case in prior literature in which a bank's prior information was revealed. In conclusion, the Heckman model cannot be interpreted as a test for private information (or sample selection) when private information takes the form of an omitted variable in the first-stage regression.

Analyst Replacement and Analyst Quality

Research Question: I examine why a brokerage firm replaces an analyst who follows a specific firm for a long run period.Research Method: The major population for this study comes from the merged I/B/E/S actual/ detail file over the period January 1, 1996 to December 31, 2009. Regression analysis is used to test hypotheses about the effect of analyst coverage on CEO compensation. Additionally, to address the potential endogeneity of analysts switching their jobs, we implement Heckman’s (1979) two-step regression model, which includes the inverse Mills ratio in the second step as a control.Main Findings: I find that the possibility of analyst replacement increases with the existence of a competing analyst. Analysts with higher quality competitors are more likely to be replaced. Moreover, brokerage firms are more likely to stop following a specific firm that is covered for a long-run period, when the analyst covering this firm switches his served brokerage. For cases with lower quality of alternative analysts, the effect is stronger. Contribution: This study contributes to the literature on financial analysts in two ways. First, it provides new insights into factors that affect the analyst replacement. Second, my results also significantly expand the rather limited literature on the influence of analysts on the brokerage firm’s decision to withdraw.

기업의 부채조달원 선택에 관한 연구: 패널표본선택모형의 적용

우리 기업의 타인자본조달에 관한 통계지표를 살펴보면 대기업은 은행의 기업대출과 회사채 등의 직접금융을 함께 사용하여 자본을 조달하고 있는 반면, 중소기업은 은행대출에 계속 의지하고 있음을 확인할 수 있다. 이러한 현실을 감안하여 본 연구에서는 기업의 타인자본조달을 실증분석하는데 있어 표본선택편의가 존재하고 이를 감안한 연구모형을 사용해야 한다고 주장한다. 이러한 주장을 뒷받침하기 위해 1990년부터 2013년까지의 상장기업 자료를 통해 부채구조를 설명하는 실증분석을 수행한 결과 선행연구에서와 마찬가지로 기업의 회사채사용에 있어 기업규모, 1대주주 지분율, 유형자산 구성비, 수익성, 배당성향 등이 영향을 미치고 있음을 확인할 수 있었으며, 패널표본선택모형에 투입된 Inverse Mills Ratio 변수가 유의하게 나타나 패널표본선택모형을 사용하는 것이 타당함을 확인하였다. 이러한 결과는 기업의 타인자본조달에 있어 표본선택편의가 존재하며 이에 관한 연구에서 이를 반드시 감안해야 함을 의미한다. When I examined the corporate financing statistics in Korea, I have recognized that there are several trends of them. First, large enterprises use bank loan and direct financing like corporate bond as debt. Second, small and medium companies mainly use bank loan only. So I argue that there is sample selection bias in corporate debt choice and using sample selection methodology is more adequate when analysing the behavior in corporate debt choice. Therefore I have tested panel sample selection model, using the listed korean firm data from 1990 to 2013 and I have found that the panel sample selection model is appropriate.

Corporate fraud culture: Re‐examining the corporate governance and performance relation

AbstractWe analyse the corporate governance and performance relation, when conditioning on corporate fraud, for fraud firms during 2000 – 2007. Fraud firms are identified as either self‐ reported fraud events, or subject to regulatory investigation. We use the inverse Mills ratio procedure to account for firms' (unobservable) fraud culture in the dynamic system GMM model of the performance‐ governance relation. We find that corporate governance is an endogenously determined characteristic that has no causal impact on firm performance when conditioning on fraud. Fraud is a significant regulatory event but its overall economic impact at the firm level is highly variable.

Does Religious Similarity Matter in International Trade in Services?

AbstractImpacts of religion on trade in services were examined using gravity models with two estimation methods: Poisson pseudo‐maximum likelihood (PPML), to correct for heteroscedasticity and bias due to the exclusion of zero trade flow, and standard Heckman correction to compensate for sample selection bias. We found that religious similarity contributes to increased international trade in services by establishing ‘trust‐related institutions’. Second, religious pluralism (variety) fosters trade even more than religious similarity, suggesting that in facilitating trade, religious openness matters more than religious similarity. Third, a strong dominant religion discourages trade in services, whereas the presence of religious minorities encourages trade in services. A highly significant inverse Mill's ratio (IMR) result indicated the presence of sample selection bias, indicating that controls are necessary to prevent estimation bias. This study concluded that religion is an important determining factor for trade in services, positively affecting interpersonal trust and thereby reducing institutional distance between countries.

STRUCTURAL THRESHOLD REGRESSION

This paper introduces the structural threshold regression (STR) model that allows for an endogenous threshold variable as well as for endogenous regressors. This model provides a parsimonious way of modeling nonlinearities and has many potential applications in economics and finance. Our framework can be viewed as a generalization of the simple threshold regression framework of Hansen (2000, Econometrica 68, 575–603) and Caner and Hansen (2004, Econometric Theory 20, 813–843) to allow for the endogeneity of the threshold variable and regime-specific heteroskedasticity. Our estimation of the threshold parameter is based on a two-stage concentrated least squares method that involves an inverse Mills ratio bias correction term in each regime. We derive its asymptotic distribution and propose a method to construct confidence intervals. We also provide inference for the slope parameters based on a generalized method of moments. Finally, we investigate the performance of the asymptotic approximations using a Monte Carlo simulation, which shows the applicability of the method in finite samples.

Rates of Returns to Vocational and General Upper Secondary Education, and to Experience in Addressing the Hollow Middle in lndonesia, 2007

.

Consistent Estimation in Pseudo Panels in the Presence of Selection Bias

Abstract In the presence of selection bias the traditional estimators for pseudo panel data models are inconsistent. This paper discusses a method to achieve consistency in static linear pseudo panels in the presence of selection bias and a testing procedure for sample selection bias. The authors’ approach uses a bias correction term proportional to the inverse Mills ratio with argument equal to the “normit” of a consistent estimation of the conditional probability of being observed given cohort membership. Monte Carlo analysis shows the test does not reject the null for fixed T at a 5% significance level in finite samples. As a “side effect” the authors utilize the enlarged pseudo panel to provide a GMM consistent estimation of the pseudo panel parameters under rejection of the null and apply the procedure to estimate the rate of return to education in Colombia.

Assessing self-selection and endogeneity issues in the relation between activity-based costing and performance

Studies investigating the relation between ABC adoption and performance are inconclusive and plagued with econometric problems. This study extends prior research to investigate the association between ABC adoption and four manufacturing plant performance measures (cycle-time improvement, quality improvement, cost improvement, and profitability) and to assess selection bias and the endogenous nature of their relationship. I use the Heckman (1979) model to assess sample selection bias and the Wooldridge (2002) 2SLS-IV approach, to investigate endogeneity. After controlling for sample selection bias and endogeneity, the coefficient of ABC under the Heckman method and ABCfit under the 2SLS-IV method becomes significantly higher compared to the coefficient of ABC under the OLS method. In addition, both the inverse Mills ratio, under the Heckman model, and Hausman F-test, under the Wooldridge 2SLS approach, are positive and significant, confirming the presence of both sample selection bias and endogeneity. Overall, I find that controlling for sample selection bias and endogeneity is essential in properly assessing the significance of ABC-performance association.