Random-effects Probit Model Research Articles

Chartering practices in liner shipping

Chartering rather than owning a vessel is a recurrent question for liner operators. This article aims at identifying the extent of such chartering practices, the characteristics of vessels chartered and if an impact on liner profitability can be found. To do so, an initial data set collected in 2009 on 510 liner operators and 5005 vessels is used. Results from random-effect probit models point out first that chartering rates are not different between small and large operators. Furthermore, findings suggest that chartering of small and young vessels is more common and that chartering rates have increased for companies subject to higher fleet growth from 2007 to 2009. An analysis using a fixed-effect logit model on intrafleet management of 17 selected liner companies further stresses that larger companies have chartered more small vessels during the last 2 years, a result that may be explained by the need to allocate financing to new larger vessels. We then study whether the chartering rate and the size of these 17 liner companies have had an influence on their observed profitability in 2007, 2008 and 2009. Our results suggest that those variables impact profitability, but in variable ways over time.

AN AUSTRALIAN DISCRETE CHOICE EXPERIMENT TO VALUE EQ‐5D HEALTH STATES

Conventionally, generic quality-of-life health states, defined within multi-attribute utility instruments, have been valued using a Standard Gamble or a Time Trade-Off. Both are grounded in expected utility theory but impose strong assumptions about the form of the utility function. Preference elicitation tasks for both are complicated, limiting the number of health states that each respondent can value and, therefore, that can be valued overall. The usual approach has been to value a set of the possible health states and impute values for the remainder. Discrete Choice Experiments (DCEs) offer an attractive alternative, allowing investigation of more flexible specifications of the utility function and greater coverage of the response surface. We designed a DCE to obtain values for EQ-5D health states and implemented it in an Australia-representative online panel (n = 1,031). A range of specifications investigating non-linear preferences with respect to time and interactions between EQ-5D levels were estimated using a random-effects probit model. The results provide empirical support for a flexible utility function, including at least some two-factor interactions. We then constructed a preference index such that full health and death were valued at 1 and 0, respectively, to provide a DCE-based algorithm for Australian cost-utility analyses.

A Pilot Discrete Choice Experiment to Explore Preferences for EQ-5D-5L Health States

The EQ-5D-5L has recently been developed to improve the sensitivity of the widely used three-level version. Valuation studies are required before the use of this new instrument can be adopted. The use of discrete choice experiments (DCEs) in this area is a promising area of research. To test the plausibility and acceptability of estimating an Australian algorithm for the newly developed five-level version of the EQ-5D using a DCE. A choice experiment was designed, consisting of 200 choice sets blocked such that each respondent answered 10 choice sets. Each choice set presented two health state-duration combinations, and an immediate death option. The experiment was implemented in an online Australian-representative sample. A random-effects probit model was estimated. To explore the feasibility of the approach, an indicative algorithm was developed. The algorithm is transformed to a 0 to 1 scale suitable for use to estimate quality-adjusted life-year weights for use in economic evaluation. A total of 973 respondents undertook the choice experiment. Respondents were slightly younger and better educated than the general Australian population. Of the 973 respondents, 932 (95.8 %) completed all ten choice sets, and a further 12 completed some of the choice sets. In choice sets in which one health state-duration combination dominated another, the dominant option was selected on 89.5 % of occasions. The mean and median completion times were 17.9 and 9.4 min, respectively, exhibiting a highly skewed distribution. The estimation results are broadly consistent with the monotonic nature of the EQ-5D-5L. Utility is increasing in life expectancy (i.e., respondents tend to prefer health profiles with longer life expectancy), and mainly decreases in higher levels in each dimension of the instrument. A high proportion of respondents found the task clear and relatively easy to complete. DCEs are a feasible approach to the estimation of utility weights for more complex multi-attribute utility instruments such as the EQ-5D-5L.

The Recreation Value of Piatra Craiului National Park: A Combined Travel Cost and Contingent Valuation Survey

This paper describes the steps of estimating the recreation value of a national park bycombining two non-market valuation methods: the Travel Cost Method and the Contingent ValuationMethod. The questionnaire was designed according to both methods by formulating specific questions.The survey was implemented in Piatra Craiului National Park during the summer of 2011, one of themost visited national parks from Romania. The recreation value in terms of consumer surplus per tripis derived from the Poisson model and in terms of willingness to pay per trip from the random effectsprobit model. It was found that tourists gain significant benefits from recreating in the park. Theestimated recreation values are expressed in monetary units and provide justification for attractingfunds for developing tourism opportunities in a sustainable way. Moreover, the findings show thattourists are aware of the value of the park and do appreciate the opportunities offered by the parkadministrations. However, the process of revealing the benefits people gain after visiting the parkshould receive more attention from the decision makers when developing the management plans.

EFFICIENCY AND EQUITY: A STATED PREFERENCE APPROACH

Outcome measurement in the economic evaluation of health care considers outcomes independent of to whom they accrue. This article reports on a discrete choice experiment designed to elicit population preferences regarding the allocation of health gain between hypothetical groups of potential patients. A random-effects probit model is estimated, and a technique for converting these results into equity weights for use in economic evaluation is adopted. On average, the modelling predicts a relatively high social value on health gains accruing to nonsmokers, carers, those with a low income and those with an expected age of death less than 45 years. Respondents tend to favour individuals with similar characteristics to themselves. These results challenge the conventional practice of assuming constant equity weighting. For decision makers, whether a formal equity weighting system represents an improvement on more informal approaches to weighing up equity and efficiency concerns remains uncertain.

Do fertilizer subsidies crowd out organic manures? The case of Malawi

AbstractWe assess the impacts of the Malawian Farm Input Subsidy Program on manure use at the farm plot level using more than 3,000 farm plot observations from six districts in central and southern Malawi over three years (2006, 2007, and 2009). The probabilities and intensities of manure use were investigated with the correlated random effects (CRE) probit and tobit models. The endogeneity of access to fertilizer subsidies and fertilizer use intensity was controlled for with a control function approach. Both the probability of manure use and intensity of manure use were positively correlated with the intensity of fertilizer use. A 1% increase in fertilizer use intensity is associated with a 1.94%–1.96% increase in the intensity of manure use outside the subsidy program and a 0.62%–1.66% increase in manure use with the subsidy program. A 1% increase in average fertilizer price was associated with a 0.43%–0.76% increase in the probability of manure use and a 3.5%–5.3% increase in the intensity of manure use.

Regional Difference and Counterfactual Decomposition of Pro-Poor Growth: An Application to Rural Ethiopia

Previous literature, ignoring regional heterogeneity, has mainly explored the interrelationship among growth, inequality, and poverty. Exploring the incidence of poverty and growth, we classify rural Ethiopia into three regions based on the difference of production technologies and climates. We find evidence that regional heterogeneity exists across the three regions. To find sources of the heterogeneity, we estimate a pseudo-fixed effect probit model controlling for household fixed effects within a random effect probit model across regions. We find that poverty is determined by different sources across the three regions each with different farming systems. Moreover, poor households can escape poverty only when their expected level of well-being is improved by increases in asset holdings and/or returns to assets. Hence, we propose, using counterfactual decomposition, that pro-poor growth can be decomposed into two components: changes in the amount of attributes such as observable household assets or capital, and changes in ‘aggregate marginal product’ of the attributes. We find that the impacts of the changes in the aggregate marginal product on pro-poor growth are significant in hoe area, but the changes in attributes do not significantly affect growth in this region. The aspects of growth are determined heterogeneously across regions: in the highland area, both components works together; in the hoe area, the changes in aggregate marginal product mainly determine the growth; and in the enset area, it is changes in attributes that mainly determine the positive growth. We find evidence that pro-poor growth in a relative sense appear in the hoe area, where the impact of changes in the aggregate marginal product on growth is heterogeneous along the income distribution; the larger impact appears in the lower tails, while smaller impact can be seen in the upper tails. However, we find no evidence of pro-poor growth in the highland and the enset areas, where the impacts of the marginal product on growth are anti-poor and insignificant, respectively. Therefore, since the impact of changes in productivity on growth differs across regions, disseminated technology to increase household productivity should be tested whether it could generate pro-poor growth based on the recipient’s environment. For example, in providing agricultural technologies through the Korea Project on International Agriculture (KOPIA) or new knowledge via the Knowledge Sharing Program (KSP), we have to deliberate carefully on whether the knowledge and technology do indeed have pro-poor aspects.

Eliciting Public Preference for Health-Care Resource Allocation in South Korea

Objectives To identify the principles the public considers important and the trade-offs between different values in health-care resource allocation practices. Methods This study approached the issue in both qualitative and quantitative ways. In a qualitative study, two focus groups discussed the issues of resource allocation in health care. To facilitate the discussion, a simple ranking task and a series of pairwise choice practices were implemented. A discrete choice experiment survey questionnaire was also administered to a sample of the general population. Attributes and levels were determined through literature reviews and the results from the focus group interview. We used a random-effect probit model to assess the effects of each attribute. Results Through the focus group interviews, we found strong public support for the principle of equal opportunity. The participants thought that the severity of disease was the most important criterion when setting priorities. The majority supported the idea that the most disadvantaged should have the highest priority even when their health gains are less than those of others. The discrete choice experiment results showed that the severity of disease, health gains, and patients' socioeconomic status significantly influence their choices, with each parameter having an expected sign. Conclusion The results showed that Koreans support not only health maximization but also equal opportunity, fair resource allocation, and equality.

Equity in health care use among older people in the UK: an analysis of panel data

This article uses panel data to investigate the extent of income-related inequity in the likelihood of visiting a General Practitioner (GP), specialist, dentist and hospital among individuals aged 65 years and over in the UK. The probability of accessing health care is predicted with separate random effects probit panel models using data from the British Household Panel Survey (BHPS) for the period 1998 to 2006. We use well-established methods based on the concept of the concentration curve to compare the cumulative distribution of health care utilization with the cumulative distribution of the population ranked by income. The results find evidence for inequity in specialist and dental care, but only slight inequity for GP care and not significant inequity in hospital admissions. Levels of inequity are highest for specialist and dental care, even when users of the private sector are excluded from analyses. The Mobility Index (MI) is also used to compare short- and long-run estimates of inequities and show that upwardly income mobile individuals contribute to inequity in the long run.

Low-Wage Jobs – Stepping Stone or Poverty Trap?

We examine whether low-paid jobs have an effect on the occupational advancement probability of unemployed persons to obtain better-paid jobs in the future (stepping-stone effect). We make use of data from the German Socio-Economic Panel (SOEP) and apply a dynamic random-effects probit model. Our results suggest that low-wage jobs can acts stepping stones to better-paid work. The improvement of the chance to obtain a high-wage job by accepting low-paid work is particularly large for less-skilled persons and for individuals with longer unemployment experiences. Low-paid work is less beneficial if the job is also associated with a low social status.

The Impact of Vocational Training on the Unemployment Duration

In this study, we estimate a semiparametric hazard model to assess the impact of vocational training on the duration of unemployment spells. Our research work is based on the data obtained from the District Labor Office in Slupsk, Poland from 1999 to 2007. We employ matching methods to take into account the potential sample selection problem, i.e., finding an adequate control group for the group of trainees. Therefore, we estimate a random-effects probit model for the likelihood of participation in a training program. The main question of this paper is whether the training significantly raises the transition rate from unemployment to employment in the short and the long run.

Regional Difference and Counterfactual Decomposition of Pro-Poor Growth: An Application to Rural Ethiopia

Previous literature, ignoring regional heterogeneity, has mainly ex-plored the interrelationship among growth, inequality, and poverty. In exploring the incidence of poverty and growth, we classify rural Ethiopia into three regions based on the difference of production technologies and climates. We find evidence that regional heterogeneity exists across the three regions. To find the sources of heterogeneity, we estimate a pseudo-fixed effect probit model controlling for household fixed effects within a random effect probit model across regions. We find that poverty is determined by different sources across the three regions, each with different farming systems. Moreover, poor households can escape poverty only when their expected level of well-being has been improved by increases in asset holdings and/or returns to assets. Hence, we propose, using counterfactual decomposition, that pro-poor growth can be decomposed into two components: changes in the amount of attributes such as observable household assets or capital, and changes in ‘aggregate marginal product’ of the attributes. We find that the impacts of the changes in the aggregate marginal product on pro-poor growth are significant in the hoe area, but the changes in attributes do not significantly affect growth in this region. The aspects of growth are determined heterogeneously across regions: in the highland area, both components work together; in the hoe area, growth is mainly determined by changes in the aggregate marginal product; and in the enset area, it is changes in attributes that mainly determine the positive growth. We find evidence that pro-poor growth in a relative sense appear in the hoe area, where the impact of changes in the aggregate marginal product on growth is heterogeneous along the income distribution; the larger impact appears in the lower tails, while smaller impact can been seen in the upper tails. However, we find no evidence of pro-poor growth in the highland and the enset areas, where the impacts of the aggregate marginal product on growth are anti-poor and insignificant, respectively. Therefore, since the impact of changes in productivity on growth differs across regions, technology disseminated to increase household productivity should be tested whether it could generate pro-poor growth in the recipient’s environment. For example, in providing agricultural technologies through the Korea Project on International Agriculture (KOPIA) or new knowledge via the Knowledge Sharing Program (KSP), we have to deliberate carefully on whether the knowledge and technology do indeed have pro-poor aspects.

Heterogeneous Evacuation Responses to Storm Forecast Attributes

This paper investigates the variation in the effects of key storm forecast factors on hypothetical evacuation decisions collected from a mail survey using a random-effects probit model with heterogeneity. Results indicate that once heterogeneity is accounted for, wind speed and landfall time are the only two significant storm forecast attributes. Further, through the use of interaction terms between the forecast attributes and individual-specific indicators, the impact of the forecast factors were found to vary significantly across race, gender, the presence of disabled persons, and geography.

Common determinants of currency crises: the role of external balance sheet variables

This article investigates the role of external balance sheet variables as determinants of currency crises in emerging market (EME) and advanced economies. A random effect probit model is used in a panel of 40 countries with monthly data during the January 1980–December 2004 period. The main results of the article are as follows. First, size and, particularly, the composition of a country's external balance sheet are found to play an important role in the onset of crises. Second, EMEs seem to be more sensitive to external balance sheet variables than developed countries, and so too do economies with fixed or quasi-fixed exchange rate regimes. Third, international capital flows seem to have become a dominant force in explaining the onset of currency crises in EMEs. Fourth, further support is provided to standard theoretical explanations of currency crises. The likelihood of a crisis is found to increase with: the extent to which the real exchange rate rises above its trend; faster growth in broad money (relative to the level of international reserves); larger current account and budget balance deficits; lower GDP growth; and, if a neighbouring country already has a crisis. Economic fundamentals are also found to be a more important explanation of the onset of currency crises during the 1980s than during the 1990s, suggesting that more recent crises are less ‘fundamentally’ driven. Copyright © 2010 John Wiley & Sons, Ltd.

선택실험법을 이용한 의약품 급여결정기준에 대한 선호분석

The purpose of this study is to elicit preference for drug listing decision criteria and to estimate the ICER threshold in South Korea using the discrete choice experiment (DCE) method. To collect the data, a DCE survey was administered to a subject sample either educated in the principle concepts of pharmacoeconomics or were decision makers within that field. Subjects chose between alternative drug profiles differing in four attributes: ICER, uncertainty, budget impact and severity of disease. The orthogonal and balanced designs were determined through computer algorithm to take the optimal set of drug profiles. The survey employed 15 hypothetical choice sets. A random effect probit model was used to analyze the relative importance of attributes and the probabilities of a recommendation response. Parameter estimates from the models indicated that three attributes (ICER, Impact, Severity of disease) influenced respondents` choice significantly(p0.001). In addition, each parameter displayed an expected sign. The Lower the ICER, the higher the probability of choosing that alternative. Respondents also preferred low levels of uncertainty and smaller impact on health service budget. They were also more likely to choose drugs for serious diseases rather than mild or moderate ones. Uncertainty however is not statistically significant. The ICER threshold, at which the probability of a recommendation was 0.5, was 29,000,000 KW/QALY in expert group and 46,500,000 KW/QALY in industry group. We also found that those in our sample were willing to accept high ICER to get medication for severe diseases. This study demonstrates that the cost-effectiveness, budget impact and severity of disease are the main reimbursement decision criteria in South Korea, and that DCE can be a useful tool in analyzing the decision making process where a variety of factors are considered and prioritized.

A note on the existence of the posteriors for one-way random effect probit models

The existence of the posterior distribution for one-way random effect probit models has been investigated when the uniform prior is applied to the overall mean and a class of noninformative priors are applied to the variance parameter. The sufficient conditions to ensure the propriety of the posterior are given for the cases with replicates at some factor levels. It is shown that the posterior distribution is never proper if there is only one observation at each factor level. For this case, however, a class of proper priors for the variance parameter can provide the necessary and sufficient conditions for the propriety of the posterior.

Bidding behavior in dynamic auction settings: An empirical analysis of eBay

We study the impact of dynamic environment of eBay auctions on bidding behavior. Due to high-speed Internet and practically costless search possibilities, bidding behavior is no longer a function of characteristics of a single auction but depends on auctions running simultaneously, completed auctions, available Buy-It-Now prices as well as various outside options. We study how this dynamic market affects a bidder’s choice of participating in an auction or leaving eBay for an outside alternative. We analyze Texas Instruments (TI-83) Graphing Calculator auctions presented on eBay. We estimate a random-effects probit model to study bidders’ probability of staying in eBay, while controlling for unobservable individual-specific heterogeneity. Our main result shows that Market Tightness – the ratio of bidders to sellers – has a negative and significant effect on bidders’ decision to remain in eBay. Moreover, variables containing information from other eBay auctions significantly affect bidders’ participation decision, thus emphasizing the importance of the dynamic, multi-auction environment in eBay marketplace for potential buyers.

Common Determinants of Currency Crises: Role of External Balance Sheet Variables

This paper investigates the role of external balance sheet variables as determinants of currency crises in emerging market (EME) and advanced economies. A random effect probit model is used in a panel of 40 countries with monthly data over the January 1980-December 2004 period. The main results of the paper are as follows. First, size and, particularly, the composition of a country’s external balance sheet are found to play an important role in the onset of crises. Second, EMEs seem to be more sensitive to external balance sheet variables than developed countries, and so too do economies with fixed or quasi-fixed exchange rate regimes. Third, further support is provided to standard theoretical explanations of currency crises. The likelihood of a crisis is found to increase with: the extent to which the real exchange rate rises above its trend; faster growth in broad money (relative to the level of international reserves); larger current account and budget balance deficits; lower GDP growth; and, if a neighbouring country already has a crisis. Economic fundamentals are also found to be a more important explanation of the onset of currency crises during the 1980s than during the 1990s, suggesting that more recent crises are less fundamentally driven.

Are quality and innovation management conflicting activities?

It is commonplace to assert that hard components of quality management inhibit innovation. In fact, the relationship between these two activities (quality and innovation management) is complex and, as claimed in this paper, bi-directional. Some recent works suggest that innovative companies are used to change management and therefore would find adopting quality management routines less demanding and less expensive than non-innovative companies. This paper builds on this view and proposes that innovation capabilities linked to certain valuable resources (better conditions) favor the implementation of hard components of total quality management. Innovation capabilities are key dynamic capabilities accumulated over time. In order to capture their complexities, they are modeled in a broad manner including both product and process innovations, as well as R&D and high technological level. For verifying the hypotheses, a random-effect probit model is tested on a large multi-industry panel of Spanish firms. Unobservable individual heterogeneity and time effects are controlled with this approach, which means a noteworthy improvement over previous research. The results strongly confirm the positive link between innovation capabilities and quality management. It is also demonstrated that some resources of the firm facilitate standardization and quality control activities. These findings have important managerial implications. On the one hand, developing innovation capabilities will permit companies to be proactive in the adoption of standardized management systems. On the other hand, quality and innovation departments should cooperate in order to ease the standardization of new products and processes.

The Wooldridge Method for the Initial Values Problem is Simple: What About Performance?

The Wooldridge method is based on a simple and novel strategy to deal with the initial values problem in the nonlinear dynamic random-effects panel data models. This characteristic of the method makes it very attractive in empirical applications. However, its finite sample performance is not known as of yet. In this paper we investigate the performance of this method in comparison with an ideal case in which the initial values are known constants, the worst scenario based on exogenous initial values assumption, and the Heckman's reduced-form approximation method which is widely used in the literature. The dynamic random-effects probit and tobit (type1) models are used as the working examples. Various designs of Monte Carlo Experiments with balanced and unbalanced panel data sets, and also two full length empirical applications are provided. The results suggest that the Wooldridge method works very well for the panels with moderately long durations (longer than 5-8 periods). In short panels Heckman's reduced-form approximation is suggested (shorter than 5 periods). It is also found that all methods perform equally well for panels of long durations (longer than 10-15 periods).