Dynamic Probit Model Research Articles

Getting stuck, falling behind or moving forward: Rural livelihood movements and persistence in Nepal

Research on household livelihood dynamics is central to rural poverty reduction. In this paper, we adopt a three-wave panel dataset to explore the persistence of and transitions in household livelihoods in three districts of Nepal using duration and dynamic probit models. The results demonstrate that the livelihood strategies of rural households are dynamic: approximately 16 per cent of the sample households transitioned up to high-remunerative livelihood strategies, 10 per cent of the households shifted down to either low- or medium-remunerative livelihood strategies, and 69 per cent remained trapped in low-remunerative livelihood strategies. The major upward transition occurred from medium-remunerative strategies to high-remunerative strategies (14 per cent). Overall, 70 per cent of the households persisted in one of the three livelihood strategies, and the remaining 30 per cent changed their strategy at least once. This dynamic is associated with the households’ duration in a particular livelihood strategy and the various characteristics of households and household heads. Understanding livelihood movement, livelihood persistence and the associated covariates and targeting the poor on this basis is crucial to combating rural poverty and dismantling poverty traps. To this end, the present study suggests (i) strengthening physical and financial asset endowments to address low-remunerative strategies, (ii) improving infrastructure connectivity, particularly in remote and inaccessible areas, (iii) insuring against shocks, iv) enhancing opportunities for generating remittances and enabling petty trade, and (v) supporting the establishment of business ventures.

Residential housing bubbles in Hong Kong: identification and explanation based on GSADF test and dynamic probit model

We aim to investigate whether the current high housing price in Hong Kong contains a bubble and to identify the causes of such housing bubble if it exists by combining the generalised sup augmented Dickey–Fuller (GSADF) test and dynamic probit models. Empirical results indicate that the current Hong Kong housing market contains a bubble, and it’s the investors’ speculative demand and the increase of monetary supply that lead to the housing bubbles in Hong Kong. Moreover, speculative investors would prefer the mass markets to the luxury ones considering the outstanding performance of the former. Such preference would contribute to more bubbles in the mass markets than in the luxury ones. In view of these, Hong Kong Government should retrospect the linked exchange rate system and be alert to the impacts of the US monetary policies on Hong Kong residential market. To offset the housing bubble, a targeted and effective approach to the Hong Kong Government is to constrict the speculative demand from investors particularly in the mass markets.

DPB: Dynamic Panel Binary Data Models in gretl

This paper presents the gretl function package DPB for estimating dynamic binary models with panel data. The package contains routines for the estimation of the randomeffects dynamic probit model proposed by Heckman (1981b) and its generalisation by Hyslop (1999) and Keane and Sauer (2009) to accommodate AR(1) disturbances. The fixed-effects estimator by Bartolucci and Nigro (2010) is also implemented. DPB is available on the gretl function packages archive.

The U-shaped effect of R&D intensity on employee downsizing: evidence from Spanish manufacturing firms (1994–2010)

This study explores an empirically untested issue: the relationship between the level of a firm’s R&D intensity and employee downsizing. Basing our conceptual development on the Resource-Based View of the firm, we argue that a linear relationship is a poor approximation of the proposed relationship. Instead, we find theoretical insights supporting the logic that employee downsizing decreases as firms shift from low to moderate levels of R&D intensity but increases as firms change from moderate to high levels. This prediction suggests that a U-shaped link is a better representation of the way in which a firm’s R&D intensity can affect downsizing. This hypothesis is tested and confirmed using a sample of Spanish manufacturing firms during the 1994–2010 period. We also propose a novel empirical tool (i.e. dynamic probit models) that is especially useful for addressing the potential endogeneity and simultaneity problem in studying this relationship. Implications for future research and practice are presented along with the conclusions of our findings.

Long‐Term Health Spending Persistence among the Privately Insured in the US

AbstractThere is little current information regarding the long‐term persistence of health spending in the United States, in particular among the population aged under 65 (pre‐Medicare eligibility). We describe and model the extent of persistence over a six‐year period (2003–08) using medical and pharmacy claims for over 3 million employees, retirees and dependants derived from the Truven Health MarketScan database. Overall, substantial persistence in spending exists, particularly at the extremes of the distribution and for pharmaceutical spending. Error components models are estimated to separate transient from persistent variation in spending, and dynamic probit models are estimated to assess the predictive power of demographic and co‐morbid conditions and prior high spending in determining the likelihood of future high spending. A better understanding of the persistence of health spending can inform the selection and evaluation of appropriate interventions to address high costs, and can help forecast the likelihood and severity of adverse selection in public and private programmes.

Estimating Dynamic Binary Panel Data Model with Random Effects: A Computational Note

Recently, Gao et al. (J Time Ser Anal, 2016 doi: 10.1111/jtsa.12178) propose a new estimation method for dynamic panel probit model with random effects, where the theoretical properties of estimator are derived. In this paper, we extend their estimation method to the \(T\ge 3\) case, and some Monte Carlo simulations are presented to illustrate the extended estimator.

Gender Difference in Path Dependency and Spousal Bargaining in Lifestyle Formation: Evidence from Korea

We investigate a direct spousal influence on individual risky behavior such as sedentariness, smoking and drinking. We also assess its economic importance relative to that of an individual's own path dependency, and we compare the results between women and men. Using the longitudinal data from the Korean Labor & Income Panel Study (KLIPS) from 2005 to 2014, we apply a bivariate dynamic probit model with random effects to control for individual unobserved heterogeneity and contemporaneous shocks that are shared between partners. While prior studies reveal that the spousal spillover effect is muted once assortative mating and shared environment are considered, our results demonstrate that intra-household bargaining still plays a role. We also find that both females and males receive an equivalent magnitude of peer pressure from their spouses regarding sedentariness and drinking. Social learning or altruism explains females' behavior better, whilst path dependency is more pronounced among males. Our findings suggest that healthy lifestyle can be effectively promoted if policies target male populations, considering their stronger addictive tendency. Nonetheless, due to the modest magnitude of spousal peer effect and path dependency, the impact of such an intervention would be limited.

Determinants of persistence in innovation of electronics enterprise in Ho Chi Minh city

This study analyzes the factors affecting the persistence in innovation of enterprises in electronics industry in Ho Chi Minh City using DPM (Dynamic Probit Model) with MLE (Maximum Likelihood Estimation) and CML (Conditional Maximum Likelihood) estimation method. Data are collected from annual enterprise surveys in the period between 2007 and 2013 by GSO. The results show that firm size and foreign ownership affect the persistent innovation of the electronic enterprises in Ho Chi Minh City. This indicates that enterprises in the electronic industry in the upcoming time need to strengthen and enhance size, attract foreign investment to facilitate this innovation performance for the stable and sustainable development

Persistence in generating and adopting product innovations

Purpose The purpose of this paper is to test the existence of true persistence in the generation and adoption of product innovations in the context of a developing country. Design/methodology/approach A dynamic probit model with random effects is used to test true persistence relying on a panel data set constructed from three waves of the Colombian innovation survey (Encuesta de Desarrollo e Innvovación Tecnológica) covering the time span from 2003 to 2008. Findings This paper empirically shows the existence of true innovation persistence for two of the three types of product innovation studied: the adoption of product innovation that is new to the firm; and the adoption of product innovation that is new to the national market. However, the study could not confirm true persistence in the generation of product innovation. Originality/value To the best of our knowledge, this is the first study that systematically tests innovation persistence differentiating between the adoption of innovations that are new to the firm and innovation that is new to the national market. It is also the first study in this research area that uses a dynamic probit model with random effects according to the original specification by Wooldridge (2005).

Predicting the direction of US stock markets using industry returns

In this paper, we examine the directional predictability of US excess stock market returns by lagged excess returns from industry portfolios and a number of other commonly used variables by means of dynamic probit models. We focus on the directional component of the market returns because, for investment purposes, forecasting the direction of return correctly is presumably more relevant than the accuracy of point forecasts. Our findings suggest that only a small number of industries have predictive power for market returns, meaning that we find little evidence of stock markets reacting with a delay to information contained in industry returns. We also find that the binary response models outperform conventional predictive regressions in forecasting the direction of the market return. Finally, we test trading strategies and find that some of the industry portfolios do contain information that can be used to improve investment returns.

The impact of precarious employment on mental health: The case of Italy

Although there has been a sizeable empirical literature measuring the effect of job precariousness on the mental health of workers the debate is still open, and understanding the true nature of such relationship has important policy implications.In this paper, we investigate the impact of precarious employment on mental health using a unique, very large data set that matches information on job contracts for over 2.7 million employees in Italy followed over the years 2007–2011, with their psychotropic medication prescription. We examine the causal effects of temporary contracts, their duration and the number of contract changes during the year on the probability of having one or more prescriptions for medication to treat mental health problems. To this end, we estimate a dynamic Probit model, and deal with the potential endogeneity of regressors by adopting an instrumental variables approach. As instruments, we use firm-level probabilities of being a temporary worker as well as other firm-level variables that do not depend on the mental illness status of the workers.Our results show that the probability of psychotropic medication prescription is higher for workers under temporary job contracts. More days of work under temporary contract as well as frequent changes in temporary contract significantly increase the probability of developing mental health problems that need to be medically treated. We also find that moving from permanent to temporary employment increases mental illness; symmetrically, although with a smaller effect in absolute value, moving from temporary to permanent employment tends to reduce it. Policy interventions aimed at increasing the flexibility of the labour market through an increase of temporary contracts should also take into account the social and economic cost of these reforms, in terms of psychological wellbeing of employees.

Dynamics of female labour force participation in France

ABSTRACTThis article formulates and estimates a structural intertemporal model of labour force participation. Relying on theoretical characterizations derived from an economic model of lifetime behaviour, we estimate a dynamic probit model with correlated random effects using longitudinal data to allow for a dynamic structure. The model is applied to a panel of married women drawn from the 1997–2002 French Labour Force surveys in order to represent their participation behaviour. It is estimated by maximum simulated likelihood. Our results show that women’s decisions to go out to work are characterized by significant state dependence, unobserved heterogeneity and negative serial correlation in the transitory error component. In addition, we find that the presence of young children in a woman’s household reduces her labour participation, but unequally according to their age and number. As expected, educational level has a positive impact on women’s participation. Last, a rise in the husband’s wage is found to raise female participation initially and to lower it subsequently. This empirical finding suggests that leisure may not systematically be a normal good, contrary to what is frequently assumed in the literature.

Estimation for Dynamic and Static Panel Probit Models with Large Individual Effects

For discrete panel data, the dynamic relationship between successive observations is often of interest. We consider a dynamic probit model for short panel data. A problem with estimating the dynamic parameter of interest is that the model contains a large number of nuisance parameters, one for each individual. Heckman proposed to use maximum likelihood estimation of the dynamic parameter, which, however, does not perform well if the individual effects are large. We suggest new estimators for the dynamic parameter, based on the assumption that the individual parameters are random and possibly large. Theoretical properties of our estimators are derived, and a simulation study shows they have some advantages compared with Heckman's estimator and the modified profile likelihood estimator for fixed effects.

Forecasting Banking Crises with Dynamic Panel Probit Models

Banking crises are rare events, however when they occur they often have dramatic consequences. The aim of this paper is to contribute to the toolkit of early warning models available to policy makers by exploring the dynamics and non-linearities embedded in a panel dataset covering several countries over three decades. The in-sample and out-of-sample forecast performance of several dynamic probit models is evaluated, with the objective of developing a common vulnerability indicator with early warning properties. The results obtained evidence that adding dynamic and non-linear components to the models substantially improves the ability to forecast banking crises.

Predicting exchange rate cycles utilizing risk factors

This paper attempts to predict the cyclical behavior of exchange rates by using five risk factors, viz., violations of uncovered interest rate parity (UIP), relative purchasing power parity (RPPP) and pseudo-parity for equity returns, relative (cross-country) TED spreads and relative term spreads. These factors are found to forecast periods of depreciation or appreciation and subsequent reversals. The estimates based on a dynamic probit model reveal that violations of UIP, RPPP and equity market pseudo-parity exhibit predictive power for currency cycles albeit only at short horizons. The proposed framework can be utilized by policy makers to smoothen the resulting currency misalignment and by investors to form trading strategies and hedge their positions as well as re-balance their carry trade positions.

Persistence of various types of innovation analyzed and explained

This paper analyzes the persistency in innovation behavior of firms. Using five waves of the Community Innovation Survey in Sweden, we have traced the innovative behavior of firms over a ten-year period, i.e., between 2002 and 2012. We distinguish between four types of innovations: process, product, marketing, and organizational innovations. First, using transition probability matrix, we found evidence of (unconditional) state dependence in all types of innovation, with product innovators having the strongest persistent behavior. Second, using a dynamic probit model, we found evidence of “true” state dependency among all types of innovations, except marketing innovators. Once again, the strongest persistency was found for product innovators.

Dynamics of Firm Participation in R&D Tax Credit and Subsidy Programs

We provide comparative evidence on R&D tax credit and subsidy programs by studying whether firms' participation in each program exhibits state dependence and whether cross program interactions exist and are significant. We use a panel of manufacturing Spanish firms, which could use both types of support, to estimate a random effects bivariate dynamic probit model of program participation. We find that true state dependence of participation in R&D subsidy and tax credit programs accounts respectively for about 55% and 60% of observed persistence. In contrast, we do not find evidence of cross program interaction, suggesting that each tool is used by firms with different profiles. Digging on the role of some observable variables, we find that both programs reach on average stable R&D performers, and that they do not foster participation of young firms relative to older ones. We also identify significant differences across programs: while diversified and commercially successful firms are more likely to use tax incentives, those with high productivity are more likely to obtain subsidies. We discuss some policy implications of these findings.

State dependence in the incidence of strikes: Evidence from Canadian contract data using Heckman’s dynamic probit model

Abstract This paper considers whether there is state dependence in strike incidence using longitudinal contract data from Canada. I obtain my estimates using Heckman’s estimator for dynamic probit models and find that there is no state dependence in strike incidence.

Productivity growth persistence: firm strategies, size and system properties

This paper investigates the dynamics of productivity in a large sample of Italian manufacturing firms, focusing on the determinants of firm-level persistence in time of high total factor productivity (TFP) growth rates relative to the corresponding sectoral distributions. In particular, we assess the impact of both the internal characteristics of companies, including size and management strategies, and external systemic conditions, including business cycles and regional innovation performance. In order to disentangle the effects of the mix of internal and systemic factors in shaping firm-level persistence, we implement both transition probability matrices and dynamic probit models. Results reveal the presence of significant persistence in TFP growth rates. Such persistence turns out to be path-dependent since it is shaped by a number of complementary and contingent factors that locally affect the dynamics of the process.

Poverty Persistence and Intra-Household Heterogeneity in Occupations: Evidence from Urban Ethiopia

Previous studies of poverty in developing countries have to a great extent focussed on the characteristics of the household head and used these as proxies for the underlying ability of the household to generate income. This paper uses five rounds of panel data to investigate the persistence of poverty in urban Ethiopia, with a particular focus on the role of intra-household heterogeneity in occupations. Dynamic probit and system generalised method of moments regression results suggest that international remittances and labour market status of non-head household members are important determinants of households' poverty status. Results also show that controlling for these variables and the “initial conditions problem” encountered in nonlinear dynamic probit models reduces the magnitude of estimated poverty persistence significantly for urban Ethiopia. These findings have important implications for identifying the poor and formulating effective poverty reduction and targeting strategies.