Continuous-time Hazard Models Research Articles

Do indigenous or foreign innovation efforts matter for the middle-income economies transition to the higher-income rank? An empirical evidence

ABSTRACT This study analyzes whether indigenous or foreign innovation efforts are more important for the transition of middle-income economies to the higher-income rank. Data on 61 countries in both lower middle-income and upper middle-income categories between 1980 and 2018 are used. Discrete-time hazard models are employed. The unobserved heterogeneity is controlled in the estimations to improve the continuous-time hazard model applied in existing studies on innovation at the national level. Results show that foreign sources of innovation measured by nonresident patents and international R&D spillovers through the FDI channel are more important for the lower middle-income group to move up the income ladder. For the upper middle-income group, domestic source of innovation measured by R&D capital stock is the most important, followed by foreign innovation diffused through the import channel. Institutional quality supports the upper middle-income economies to obtain the high-income rank.

Effects of Prices on Youth Cigarette Smoking and Tobacco Use Initiation in Ghana and Nigeria.

Background: Population growth in the African region is set to outpace the rate of decline in smoking prevalence, leading to a projected increase in the total number of smokers. As most tobacco users initiate during their adolescent years, tobacco prevention strategies targeting youth will be particularly important. Methods: This study estimated the impact of cigarette prices on youth cigarette smoking and tobacco use initiation in Ghana and Nigeria using the Global Youth Tobacco Survey data. First, we used cross-section data and logit models to estimate the effects of prices on youth cigarette smoking. Second, we created pseudo longitudinal data and used continuous-time hazard models to evaluate the impact of cigarette prices on tobacco use initiation. Results: We found that higher cigarette prices decreased both 30-day cigarette smoking and tobacco use onset significantly in both Ghana and Nigeria. Additionally, the price elasticity of cigarette smoking and tobacco use initiation ranged from −0.44 to −1.13, and −1.04 to −3.66, respectively. Conclusions: As one of the first studies on youth tobacco consumption in Sub-Saharan Africa, this study strongly suggests that policies that increase real cigarette prices can lower both cigarette smoking and tobacco use initiation among youth in Ghana and Nigeria.

The Correct Use of Hazard Models: A Comparison Employing Different Definitions of SMEs Financial Distress

This study aims to shed light on the debate concerning the choice between discrete-time and continuous-time hazard models in making bankruptcy or any binary prediction using interval censored data. Building on theoretical suggestions from various disciplines, we empirically compare widely used discrete-time hazard models (with logit and clog-log links) and continuous-time Cox Proportional Hazards (CPH) model in predicting bankruptcy and financial distress of the United States SMEs. In line with the theoretical arguments we report that discrete-time hazard models are superior to continuous-time CPH model in making binary predictions using interval censored data. Moreover, hazard models that employ failure definition of SMEs based on bankruptcy laws and firms’ financial health exhibits superior goodness of fit and classification measures, in comparison to hazard models that employ failure definitions that are based either on bankruptcy laws or firms’ financial health.

Empirical comparison of hazard models in predicting SMEs failure

This study aims to shed light on the debate concerning the choice between discrete-time and continuous-time hazard models in making bankruptcy or any binary prediction using interval censored data. Building on the theoretical suggestions from various disciplines, we empirically compare widely used discrete-time hazard models (with logit and clog-log links) and the continuous-time Cox Proportional Hazards (CPH) model in predicting bankruptcy and financial distress of the United States Small and Medium-sized Enterprises (SMEs). Consistent with the theoretical arguments, we report that discrete-time hazard models are superior to the continuous-time CPH model in making binary predictions using interval censored data. Moreover, hazard models developed using a failure definition based jointly on bankruptcy laws and firms’ financial health exhibit superior goodness of fit and classification measures, in comparison to models that employ a failure definition based either on bankruptcy laws or firms’ financial health alone.

Child mortality inequalities across Rwanda districts: a geoadditive continuous-time survival analysis.

Child survival programmes are efficient when they target the most significant and area-specific factors. This study aimed to assess the key determinants and spatial variation of child mortality at the district level in Rwanda. Data from the 2010 Rwanda Demographic and Health Survey were analysed for 8817 live births that occurred during five years preceding the survey. Out of the children born, 433 had died before survey interviews were carried out. A full Bayesian geo-additive continuous-time hazard model enabled us to maximise data utilisation and hence improve the accuracy of our estimates. The results showed substantial district- level spatial variation in childhood mortality in Rwanda. District-specific spatial characteristics were particularly associated with higher death hazards in two districts: Musanze and Nyabihu. The model estimates showed that there were lower death rates among children from households of medium and high economic status compared to those from low-economic status households. Factors, such as four antenatal care visits, delivery at a health facility, prolonged breastfeeding and mothers younger than 31 years were associated with lower child death rates. Long preceding birth intervals were also associated with fewer hazards. For these reasons, programmes aimed at reducing child mortality gaps between districts in Rwanda should target maternal factors and take into consideration district-specific spatial characteristics. Further, child survival gains require strengthening or scaling-up of existing programmes pertaining to access to, and utilisation of maternal and child health care services as well as reduction of the household gap in the economic status.

Bridge Unemployment in Germany: Response in Labour Supply to an Increased Early Retirement Age

This study examines an increase in the early retirement age from 60 to 63 for the group of older unemployed men in Germany. As consequence of this policy reform, the time to retirement is increased from the perspective of recently unemployed individuals and therefore serves as a source of exogenous variation. We estimate continuous time hazard models for individuals at risk of leaving the state unemployment into employment or into early retirement due to exceptional rules. We find a positive impact of an increase in the early retirement age on the reemployment probability whereas the probability to retire early due to exceptional rules is not affected.

Do siblings’ fertility decisions influence each other?

Individuals' fertility decisions are shaped not only by their own characteristics and life course paths but also by social interaction with others. However, in practice, it is difficult to disentangle the role of social interaction from other factors, such as individual and family background variables. We measure social interaction through the cross-sibling influences on fertility. Continuous-time hazard models are estimated separately for women's first and second births. In addition to individual socioeconomic variables, demographic variables, and an unobserved factor specific to each sibling pair, siblings' birth events and their timing enter as time-varying covariates. We use data from longitudinal population-wide Norwegian administrative registers. The data cover more than 110,000 sibling pairs and include the siblings' fertility, education, income, and marital histories. Our results indicate that cross-sibling influences are relatively strong for the respondents' first births but weak for the second parity transition.

An investigation of commuting trip timing and mode choice in the Greater Toronto Area: Application of a joint discrete-continuous model

The trip timing and mode choice are two critical decisions of individual commuters mostly define peak period traffic congestion in urban areas. Due to the increasing evidence in many North American cities that the duration of the congested peak travelling periods is expanding (peak spreading), it becomes necessary and natural to investigate these two commuting decisions jointly. In addition to being considered jointly with mode choice decisions, trip timing must also be modelled as a continuous variable in order to precisely capture peak spreading trends in a policy sensitive transportation demand model. However, in the literature to date, these two fundamental decisions have largely been treated separately or in some cases as integrated discrete decisions for joint investigation. In this paper, a discrete-continuous econometric model is used to investigate the joint decisions of trip timing and mode choice for commuting trips in the Greater Toronto Area (GTA). The joint model, with a multinomial logit model for mode choice and a continuous time hazard model for trip timing, allows for unrestricted correlation between the unobserved factors influencing these two decisions. Models are estimated by occupation groups using 2001 travel survey data for the GTA. Across all occupation groups, strong correlations between unobserved factors influencing mode choice and trip timing are found. Furthermore, the estimated model proves that it sufficiently captures the peak spreading phenomenon and is capable of being applied within the activity-based travel demand model framework.

"Skin in the Game": Zero Downpayment Mortgage Default

Previous research has focused on equity as a prime determinant of mortgage default propensities. This paper extends the analysis of mortgage default to include mortgages that require no down payment from the purchaser. The results indicate that borrowers who provide down payments from their own resources have significantly lower default propensities than do borrowers whose down payments come from relatives, government agencies, or non-profits. Borrowers with down payments from seller-funded non-profits, who make no down payment at all, have the highest default rates. Additionally, borrowers who do not make down payments from their own resources tend to have higher loss given default in the small subset of loans that had completed the property disposition process. A competing risks, continuous time hazard model with Heckman-Singer non-parametric heterogeneity is used to estimate the conditional probability of a serious delinquency, or a claim, as a function of a host of standard control variables, and indicators for the presence and source of the down payment. The data consist of a nationally representative random sample of about 5,000 FHA insured single family mortgages endorsed in Fiscal Years 2000, 2001, and 2002, observed through September 30, 2006, and samples of about 1,000 FHA loans each from the Atlanta, Indianapolis, and Salt Lake City MSAs in the same time period.

Gifts, Down Payments and Mortgage Default

Previous research has focused on equity as a prime determinant of mortgage default propensities. This paper extends the analysis of mortgage default to include the source of initial equity for the purchaser. A continuous time hazard model is used to estimate the conditional probability of a serious delinquency, or a claim, as a function of a host of standard control variables, and indicators for the source of the down payment. The data consist of a nationally representative random sample of about 5,000 FHA insured single family mortgages endorsed in Fiscal Years 2000, 2001, and 2002, observed through January 31, 2006, and samples of about 1,000 FHA loans each from the Atlanta, Indianapolis, and Salt Lake City MSAs in the same time period. The results indicate that borrowers who provide down payments from their own resources have significantly lower default propensities than do borrowers whose down payments come from relatives, government agencies, or non-profits. Borrowers with down payments from seller-funded non-profits have the highest default rates. Additionally, borrowers who do not make down payments from their own resources tend to have higher loss given default in the small subset of loans that had completed the property disposition process.

Modeling Skeletal Components of Workers' Daily Activity Schedules

Four sets of econometric models are presented for time use decisions (durations and start times) regarding the basic, regular, and committed components (skeleton) of workers' daily life: gap before work, work, gap after work, and night sleep. Two types of models are compared for each component: multilevel linear models and continuous-time hazard models. The multilevel models consider three-level random effects (temporal, personal, and household) and the hazard models consider individual-based unobserved heterogeneity. On the basis of performance in fitting observed data, hazard models are selected for the first three components, and a multilevel model is selected for the last component. For parametric hazard models, the Gompertz distribution shows promising performance in fitting activity data. The models are estimated by using 2002-2003 Toronto [Canada] Computerized Household Activity Scheduling Elicitor data.

Modeling Skeletal Components of Workers’ Daily Activity Schedules

Four sets of econometric models are presented for time use decisions (durations and start times) regarding the basic, regular, and committed components (skeleton) of workers’ daily life: gap before work, work, gap after work, and night sleep. Two types of models are compared for each component: multilevel linear models and continuous-time hazard models. The multilevel models consider three-level random effects (temporal, personal, and household) and the hazard models consider individual-based unobserved heterogeneity. On the basis of performance in fitting observed data, hazard models are selected for the first three components, and a multilevel model is selected for the last component. For parametric hazard models, the Gompertz distribution shows promising performance in fitting activity data. The models are estimated by using 2002–2003 Toronto [Canada] Computerized Household Activity Scheduling Elicitor data.

Why do Couples with Highly Educated Parents have Higher Divorce Rates?

Some studies have found that having highly educated parents or having a middle- or upper-class background increases the risk of divorce, but the literature offers little theoretical reasoning on the nature of this effect. This paper extends earlier research on this relationship by considering several competing explanations of the elevated divorce risk for couples from such backgrounds. Continuous-time hazard models are estimated using register-based data on 20 complete Norwegian first marriage cohorts (n = 276326). Results show that the higher dissolution risk for couples with highly educated parents cannot be attributed to the parents' marital history, to their economic resources, or to an urban environment. I speculate that some socio-cultural factor contributes to raise these couples' risk of marital dissolution.

Do State Medicaid Policies Affect the Risk of Nursing Home Entry among the Elderly?

This study investigates the effects of five state Medicaid policies on the risk of nursing home entry among a recent cohort of older persons. Longitudinal data from the first (1993) and second (1995) waves of the Asset and Health Dynamics Among the Oldest Old Survey and continuous-time hazard models are used for the analyses. Separate models are developed for women and men, each controlling for a variety of demographic, socioeconomic, and health-related characteristics of sample persons. Findings suggest that the state Medicaid policies included in the analyses do not significantly affect the risk of nursing home use among women or men and that the strongest predictors of the risk are health-related characteristics. Circumstantial evidence is presented to argue that the lack of policy effects might be due to changes in recent decades in health care policies driven by cost containment efforts.

The impact of individual and aggregate unemployment on fertility in Norway

Continuous-time hazard models are estimated from register-based birth, migration, education and unemployment histories for the complete Norwegian population, linked with aggregate data for municipalities. The analysis covers the period 1992-98. First-birth rates are slightly higher among women who had been unemployed twelve months before than among others, whereas higher-order birth rates are slightly lower. Although men’s unemployment has a more pronounced negative effect, according to paternity rate models, the overall conclusion is that unemployment in Norway has had a negligible impact on fertility through individual-level effects. Aggregate-level effects are more important. Higher-order birth rates are lower in municipalities where men’s or women’s unemployment is high than elsewhere. All in all, the peak unemployment level of 6% experienced in 1993 is found to be associated with a reduction of about 0.08 in total fertility. The results accord well with economic theories for first and higher-order births that are based on the assumption that women are still the primary caretakers.

Paths to Success

:Since the collapse of state socialism, self-employment has become a path to relative material success in Russia. Fifteen hypotheses regarding the effects of individual and regional characteristics on the hazard rate of entry to self-employment among Russians are tested using 3,356 work/activity histories spanning 1991–97 from the Survey of Employment, income, and Attitudes in Russia. Descriptive results confirm that the self-employed are better off materially and more supportive of market reforms than any other occupational class in contemporary Russia. Among the self-employed, those with employees rank higher on these criteria. Continuous-time hazard models reveal that some determinants of entry to self-employment observed in developed capitalist societies–human capital, gender, and labor market position–operate in a similar fashion in Russia, although the effect of human capital is essentially linear and the effect of sex has not diminished over time. Other variables-age and family background-operate differently in Russia. Communist Party affiliation originally facilitated becoming self-employed, but the effect diminished and even changed sign by the mid-1990s. Measures of regional economic performance and the effectiveness of law enforcement have fairly predictable effects, while regional economic policies have little or no net influence on rates of entry to self employment. The findings reflect the hybrid character of Russian economic institutions and culture during the period of market transition.

Agreeing to Disagree: Domestic Institutional Congruence and U.S. Dispute Behavior

Domestic and international politics are linked so that domestic structure and domestic political preferences affect international conflict behavior. In particular, states' policy preferences can be characterized by the congruence of policy preferences between political institutions. As a state's policy preferences become more congruent, as policy preferences between political institutions are more similar, the state's conflict behavior changes. I hypothesize that as the domestic political institutions in a state share similar policy preferences or policy goals, disputes become more likely and tend to last longer. I employ event count and continuous-time hazard models to analyze U.S. conflict propensity and conflict duration during the period 1945-1992. Using the Militarized Interstate Dispute data set (version 2.1), I model U.S. militarized dispute behavior as a function of congruence between the policy preferences of the U.S. President and the Congress. The models reveal a strong relationship between preference congruence and both the amount of conflict and the duration of the disputes in which the U.S. engages. The congruence hypotheses are robust across a variety of measures. The results add substantial strength and substance to claims that domestic political characteristics affect international conflict. The theory and empirical analyses also refine the domestic-international linkage by allowing scholars to consider the effect of normal political change in a single state on that state's foreign policy decisions.

Time-aggregation effects on the baseline of continuous-time and discrete-time hazard models

In this study we reinvestigate the effect of time-aggregation for discrete- and continuous-time hazard models. We reanalyze the results of a previous Monte Carlo study by ter Hofstede and Wedel (1998), in which the effects of time-aggregation on the parameter estimates of hazard models were investigated. Whereas the previous study shows that the estimates of the baseline parameters of continuous- and discrete-time models suffer a lot from time-aggregation, we show that the effects on the baseline hazard itself are much smaller, in particular for the continuous-time model.

The Importance of Accommodation on the Timing of Disability Insurance Applications: Results from the Survey of Disability and Work and the Health and Retirement Study

Using data from the 1978 Survey of Disability and Work and the 1992 Health and Retirement Study, we test the importance of accommodation and other policy variables on the timing of application for Social Security Disability Insurance benefits following the onset of a work-limiting condition. We correct for choice-based sampling in the Survey of Disability and Work by extending the Manski and Lerman (1977) correction to the likelihood function of our continuous time hazard model, we find that this correction significantly affects the results. Similar findings emerge from these two data sets. Accommodation significantly reduces the speed of application and more generous benefits increase the speed of applying for SSDI.

A Monte Carlo study of time aggregation in continuous-time and discrete-time parametric hazard models

This study investigates the effects of time aggregation in discrete and continuous-time hazard models. A Monte Carlo study is conducted in which data are generated according to various continuous and discrete-time processes, and aggregated into daily, weekly and monthly intervals. These data are analyzed with flexible continuous-time and discrete-time parametric hazard models. The estimates of the structural parameter and baseline hazard seem robust to the form of the distribution of the data generation process when the time-aggregation window is small. Both estimates of continuous-time models and of discrete-time models suffer from time aggregation, but the estimates of the discrete-time model are more sensitive to aggregation.