Positive Duration Dependence Research Articles

A Journey Home: What Drives How Long People are Homeless?

This paper uses survival analysis to model exits over time from two alternative notions of homelessness. We are unique in being able to account for time-invariant, unobserved heterogeneity. We find that duration dependence has an inverted U-shape with exit rates initially increasing (indicating positive duration dependence) and then falling. Like previous researchers, we find results consistent with negative duration dependence in models which ignore unobserved heterogeneity. Exit rates out of homelessness fall with age and with the education level of mothers. Women are more likely than men to exit homelessness when it is broadly conceived, but appear to be less likely to exit when it is narrowly defined. Finally, higher paternal education and exemptions from welfare-related activity requirements due to either mental or physical health conditions are all associated with higher exit rates.

Distribution of Unemployment Spells and Survival Analysis: The Case of Bosnia and Herzegovina

It is not enough to study the labour market only by analysing the static variables like rates of employment and unemployment. For the decision making on the labour market it is essential to see the movement of people into and out of jobs, the extent to which they can or cannot quickly find alternative employment and to which extent different groups of the labour force are more affected than others. Using the frailty models, as a specific area in survival analysis, we show that the durations of unemployment are sensitive to the educational level and gender. First, females are experiencing significantly longer durations of finding job than man. Second, better-educated individuals appear to find job more quickly than the less-educated. Finally, there is evidence of duration dependence in unemployment in Bosnia and Herzegovina. The re-employment hazard exhibits positive duration dependence in the first 59 months, and then declines approximately 60 months.

The Portuguese stock market cycle

This paper tries to identify, for the first time, a chronology for the Portuguese stock market cycle and test for the presence of duration dependence in bull and bear markets. A duration-dependent Markov-switching model is estimated over monthly growth rates of the Portuguese stock index for the period January 1989 to April 2012. Six episodes of bull/bear markets are identified during that period, as well as the presence of positive duration dependence in bear but not in bull markets.

Crossing at a Red Light: Behavior of Cyclists at Urban Intersections

To investigate the relationship between cyclist violation and waiting duration, the red‐light running behavior of nonmotorized vehicles is examined at signalized intersections. Violation waiting duration is collected by video cameras and it is assigned as censored and uncensored data to distinguish between normal crossing and red‐light running. A proportional hazard‐based duration model is introduced, and variables revealing personal characteristics and traffic conditions are used to describe the effects of internal and external factors. Empirical results show that the red‐light running behavior of cyclist is time dependent. Cyclist’s violating behavior represents positive duration dependence, that the longer the waiting time elapsed, the more likely cyclists would end the wait soon. About 32% of cyclists are at high risk of violation and low waiting time to cross the intersections. About 15% of all the cyclists are generally nonrisk takers who can obey the traffic rules after waiting for 95 seconds. The human factors and external environment play an important role in cyclists’ violation behavior. Minimizing the effects of unfavorable condition in traffic planning and designing may be an effective measure to enhance traffic safety.

The duration of business cycle expansions and contractions: are there change-points in duration dependence?

The business cycles literature shows that the likelihood of an expansion or contraction ending increases with its age, i.e. they exhibit positive duration dependence. This evidence rests on the assumption that the magnitude of duration dependence is the same over time. However, we assume that the degree of likeliness of an expansion or contraction ending as it gets older might indeed change after a specific duration. Estimating a continuous-time Weibull model for a group of 13 industrial countries over the period 1948–2009 and allowing for the presence of a change-point in the duration-dependence parameter; we conclude that the evidence of positive duration dependence is no longer present when an expansion surpasses 10 years of duration.

The Portuguese Stock Market Cycle: Chronology and Duration Dependence

This paper tries to identify, for the first time, a chronology for the Portuguese stock market cycle and test for the presence of duration dependence in bull and bear markets. A duration dependent Markov-switching model is estimated over monthly growth rates of the Portuguese Stock Index for the period 1989-201'3 Six episodes of bull/bear markets are identified during that period, as well as the presence of positive duration dependence in bear but not in bull markets.

Tricked by Truncation: Spurious Duration Dependence and Social Contagion in Hazard Models

We show both analytically and through Monte Carlo simulations that applying standard hazard models to right-truncated data, i.e., data from which all right-censored observations are omitted, induces spurious positive duration dependence and hence can trick researchers into believing to have found evidence of social contagion when there is none. Truncation also tends to deflate the effect of time-invariant covariates. These results imply that not accounting for right truncation can lead managers to rely too much on word of mouth in generating new product adoption and to poorly identify the customers most likely to adopt early. Not accounting for right truncation can also lead to suboptimal pricing decisions and to erroneous assessments of variations in customer lifetime value. We assess the effectiveness of four possible solutions to the problem and find that only using an analytically corrected likelihood function protects one against truncation artifacts inflating coefficients of contagion and attenuating coefficients of time-invariant covariates.

Duration analysis of technology adoption effects of dissemination pathways: A case of ‘push–pull’ technology for control of Striga weeds and stemborers in Western Kenya

Since its introduction, the ‘push–pull’ technology (PPT) has been promoted as an innovative approach for addressing yield losses caused by Striga weeds and stemborer infestations on farmers’ maize fields in western Kenya. The technology has been disseminated through different pathways each of which might have different uptake enhancement capabilities. This study was concerned with how different dissemination pathways impacted on farmers’ time to adoption while controlling the effects of selected socio-economic and regional factors. This assessment was necessary in order to recommend which dissemination pathway or combination of pathways that could provide an effective dissemination strategy that speed up adoption, while ensuring continued practice of the technology. Data were analyzed using a parametric (Weibull) functional form to specify time to adoption from the first date of a farmer being aware of the technology. The findings indicate positive duration dependence and a positive rate of change in the adoption process. Field days (FD) had the highest effect on the speed of uptake, followed by Farmer Teachers (FT). Other variables that accelerated adoption were education, household size and high-income level. Gender, Tropical Livestock Units (TLUs), group membership, and district dummy variables had delaying effects on the adoption process. The findings suggest the need to fast track the scale up and uptake of PPT through pathways which allow for interactive information sharing and learning as exemplified through farmer teachers and field days.

Modeling Pedestrian Violation Behavior at Signalized Crosswalks in China: A Hazards-Based Duration Approach

Objective: Pedestrian violation is a major cause of traffic accidents involving pedestrians. The research objectives were to investigate the relationship between waiting duration and pedestrian violation and to provide a qualitative and quantitative analysis of the effects of human factors and external environmental factors on street-crossing behavior. Methods: Pedestrians’ street-crossing behavior was examined by modeling the waiting duration at signalized crosswalk. Pedestrian waiting duration was collected by video cameras and it was assigned as censored and uncensored data to distinguish between normal crossing and violating crossing. A nonparametric baseline duration model was introduced, and variables revealing personal characteristics, traffic conditions, and trip features were defined as covariates to describe the effects of internal and external factors. Results: Pedestrians’ crossing behaviors represented positive duration dependence that the longer the waiting time elapsed the more likely pedestrians would end the wait soon. The violation inclination of most pedestrians increased with the increasing waiting duration, but about 10 percent of pedestrians were at high risk of violation to cross the street. About half of pedestrians would still obey the traffic rules even after waiting for 50 s by the street. Human factors and the external environment played an important role in street-crossing behavior, especially for factors that involved pedestrians’ subjective willingness. Conclusions: The street-crossing behavior of pedestrians was time dependent. Pedestrians behave differently under the effects of various factors. Pedestrian safety interventions that aim at reducing pedestrian injuries may need to consider these effects. The pedestrians’ behavioral modifications, such as enhancing the safety awareness, might be the most efficient means to reducing the likelihood of pedestrian violation, though environmental modifications also worked well in improving pedestrian safety.

The Portuguese Business Cycle: Chronology and Duration Dependence

This paper tries to identify, for the first time, a chronology for the Portuguese business cycle and test for the presence of duration dependence in the respective phases of expansion and contraction. A duration dependent Markov-switching vector autoregressive model is employed in that task. This model is estimated over monthly and year-on-year (monthly) growth rates of a set of relevant economic indicators, namely, industrial production, a composite leading indicator and, additionally, civilian employment. The estimated specifications allow us to identify four main periods of contraction during the last three decades and the presence of positive duration dependence in contractions, but not in expansions.

Duration Dependence in Bull and Bear Stock Markets

Testing duration in stock markets concerns the ability to predict the turning points of bull and bear cycles. The Weibull renewal process has been used in previous studies to analyze duration dependence in economic and financial cycles. A goodness-of-fit test, however, shows that this model does not fit data from U.S. stock market cycles. As a solution, this study fits the modulated power law process that relies on less restrictive assumptions. Moreover, it measures both the long term properties of bull and bear markets, such as the tendency of the cycles to become shorter (or longer), as well as the short term effects, such as duration dependence. The results give evidence of negative duration dependence in all samples of bull markets and evidence of positive duration dependence in complete, peacetime and post WWII samples of bear markets. There is no evidence of any structural change in duration dependence after WWII in either bull or bear markets. The results show that bull and bear markets tend to get progressively shorter, but for bull markets this trend has accelerated since WWII whereas for bear markets this trend has decelerated since WWII. Goodness-of-fit tests suggest that the modulated power is a suitable model for U.S. stock market cycles.

Competing Risks in a Time on the Market Analysis

Theoretical models on the selling process in the housing market are scarce. Taylor (1999) specifies a model where time-on-the-market gives a quality signal of the house to potential buyers if inspection outcomes of the house are not public. We specify a duration model with competing risks, where the competing risks are a sale or a withdrawal from the market. We use a unique administrative dataset from the Netherlands. We find negative duration dependence in the hazard of sale and positive duration dependence in the hazard of withdrawal confirming the empirical predictions from Taylor (1999).

A survival analysis approach to the duration of union membership in Italy

Thanks to direct access to union databases, this article applies survival analysis to a sample of 47 637 Italian workers trying to explain the determinants of the duration of union membership. The results show that union membership duration is a positive, though declining, function of age. Furthermore, women, flexible workers, foreign ones and those working in cities tend to show less attachment to union membership than other workers. The estimated median duration is about 6 years. Positive hazard duration dependence is also found. Unobserved heterogeneity is detected but it does not significantly affect model estimates.

The Duration of Business Cycle Expansions and Contractions: Are There Change-Points in Duration Dependence?

The issue of whether the likelihood of an expansion or contraction ending is dependent on its age, i.e whether they are duration dependent, is widely addressed in the business cycles literature and evidence of positive duration dependence is found in several studies. However, there is an important issue that has not been explored in this literature yet: the presence of change-points in duration dependence. All the studies in this field depart from the assumption that the magnitude of duration dependence is the same over time. However, we conjecture that the degree of likeliness of an expansion or contraction ending as it gets older might change after a specific duration. To test for that possibility, this paper will allow for the presence of a change-point in the analysis of the duration of expansions and contractions for a group of 13 European and Non-European industrial countries over the period 1948-2009.The evidence provided by the estimation of a continuous-time Weibull duration model shows strong support for the presence of positive duration dependence, which is stronger for contractions than for expansions. Results also show that contractions have become longer over time and that their length is negatively affected by the length of the previous expansion. Most importantly, this paper provides quite interesting evidence for the presence of a change-point in duration dependence for expansions, but not for contractions. Results show that the magnitude of the duration dependence parameter decreases significantly when an expansion surpasses 10 years of duration. In particular, evidence of positive duration dependence is no longer found when an expansion surpasses that threshold.

An analysis of new firm survival using a hazard function

A unique data set is used to provide a detailed examination of the survival of newly established manufacturing firms in north-east England. Using data on 781 firms established between 1973 and 2001, log-logistic hazard models are estimated separately for (i) micro-enterprises and (ii) small and medium establishments (SMEs). Both micro-enterprises and SMEs show clear evidence of positive duration dependence, followed by negative duration dependence. We find the two firm types are differentially affected by firm-specific and macro-economic variables. Increases in initial plant size impact negatively on micro-enterprise survival and positively on SME survival.

The duration of economic expansions and recessions: More than duration dependence

One widespread idea in the business cycles literature is that the older is an expansion or contraction, the more likely it is to end. This paper tries to provide further empirical support for this idea of positive duration dependence and, at the same time, control for the effects of other factors like leading indicators, the duration of the previous phase, investment, price of oil and external influences on the duration of expansions and contractions. This study employs a discrete-time duration model to analyse the impact of those variables on the likelihood of an expansion and contraction ending in a panel of thirteen industrial countries over the last fifty years. The evidence provided in this paper suggests that the duration of expansions and contractions is not only dependent on their actual age: the duration of expansions is also positively dependent on the behaviour of the variables in the OECD composite leading indicator and on private investment, and negatively affected by the price of oil and by the occurrence of a peak in the US business cycle; the duration of a contraction is negatively affected by its actual age and by the duration of the previous expansion.

Modelling the duration of interest rate spells under inflation targeting in Canada

We use survival models to analyse the duration of the spells associated with the interest rate used by the Bank of Canada as its monetary policy instrument. Both nonparametric and parametric models are estimated, allowing for right-censoring of the data, and time-varying covariates. We find that the data are explained well by an accelerated failure time Weibull model, with the annual rate of inflation and the quarterly rate of growth in Gross domestic product (GDP) as covariates. The model indicates that there is positive duration dependence in the interest rate spells, and that unemployment and exchange rate effects are insignificant.

The Transition to Work for Italian University Graduates

AbstractThis study is focused on the transition from university to first job, taking into account the graduates’ characteristics and the effects relating to degree subject. A large data set from a survey on job opportunities for the 1998 Italian graduates is used. The paper uses a non‐parametric discrete‐time single‐risk model to study employment hazard. Alternative mixing distributions have also been used to account for unobserved heterogeneity. The results obtained indicate that there is evidence of positive duration dependence after a short initial period of negative duration dependence. In addition, a competing‐risk model has been estimated to characterize transitions out of unemployment.

Modeling the time to an initial public offering: When does the fruit ripen?

This paper uses a duration model to examine credit unions who have converted to mutual savings institutions. We investigate the characteristics that they possess which might influence the probability that they subsequently demutualize and become publicly traded banks. Our key findings are as follows. First, we find evidence of positive duration dependence in the data we examine. Second, we find that the hazard of an IPO issue increases in two waves. The first occurs between 3 1/2 years to 4 years reflecting the increased transition intensity for those who wish to demutualize early. However, the majority of the institutions in our sample do so by approximately 8 years after conversion, and this is where the second spike in the hazard occurs. Finally, upon estimating the model with Cox’s (Biometrika 62:269–276, 1975) semi-parametric partial likelihood approach, we find the probability that a converted institution will issue an IPO is influenced to a large extent by various measures that capture asset quality, in particular through measures that are equity driven. However, contrary to what is found in the literature on bank transformation, we find that the size of an institution does not appear to influence the hazard rate within this industry.

Successful Employer Search? An Empirical Analysis of Vacancy Duration Using Micro Data

This paper provides the first estimates of the determinants of the duration of employer search in the UK. We model duration until a vacancy is either successfully filled or withdrawn from the market. The econometric techniques deal with multiple vacancies and unobserved heterogeneity (dependent risks), using flexible and parametric baseline hazards. The hazards to filling and withdrawing exhibit negative and positive duration dependence respectively, implying that the conditional probability of successful employer search decreases with duration. We also find that nonmanual vacancies are less likely to fill, consistent with there being skill shortages in the sample period.