Contributions published at Statistics and Empirical Economic Research (Rainer Winkelmann)

In this paper, we test how reporting behaviors (response time, cognitive effort, questionnaire order) affect reported happiness in a large Dutch internet panel survey. We find that slower responses and higher cognitive effort reduce reported happiness. Moreover, in multivariate happiness equations, these factors moderate the estimated effect of income on happiness, while no interaction effects are found for other determinants of happiness. As a consequence, relative marginal effects may not be invariant to reporting circumstances.

This paper is concerned with the analysis of zero-inflated count data when time of exposure varies. It proposes a new zero-inflated count data model that is based on two homogeneous Poisson processes and accounts for exposure time in a theory consistent way. The new model is used in an application to the effect of insurance generosity on the number of absent days.

Traditional tools of welfare economics identify the envy-related welfare loss from conspicuous consumption only under very strong assumptions. Measured income and life satisfaction offers an alternative for estimating such consumption externalities. The approach is developed in the context of luxury car consumption (Ferraris and Porsches) in Switzerland. Results from household panel data and fixed effects panel regressions suggest that the prevalence of luxury cars in the municipality of residence has a negative impact on own income satisfaction.

null

null

null

null

null

Recently, there has been a renewed interest in modeling economic time series by vectorautoregressive moving-average models. However, this class of models has been unpopularin practice because of estimation problems and the complexity of the identication stage.These disadvantages could have led to the dominant use of vector autoregressive modelsin macroeconomic research. In this paper, several simple estimation methods for vectorautoregressive moving-average models are compared among each other and with purevector autoregressive modeling using ordinary least squares by means of a Monte Carlostudy. Dierent evaluation criteria are used to judge the relative performances of the algorithms.

The paper proposes a new estimator for the fixed effects ordered logit model. In contrast to existing methods, the new procedure allows estimating the thresholds. The empirical relevance and simplicity of implementation is illustrated in an application to the effect of unemployment on life satisfaction.

This paper studies the evolution of life satisfaction over the life course in Germany. It clarifies the causal interpretation of the econometric model by discussing the choice of control variables and the underidentification between age, cohort and time effects. The empirical part analyzes the distribution of life satisfaction over the life course at the aggregated, subgroup and individual level. To the findings: On average, life satisfaction is mildly decreasing up to age fifty-five followed by a hump shape with a maximum at seventy. The analysis at the lower levels suggests that people differ in their life satisfaction trends, whereas the hump shape after age fifty-five is robust. No important differences between men and women are found. In contrast, education groups differ in their trends: highly educated people become happier over the life cycle, where life satisfaction decreases for less educated people.

We bring together some recent advances in the literature on vector autoregressive moving-average models creating a relatively simple specification and estimation strategy for the cointegrated case. We show that in the cointegrated case with fixed initial values there exists a so-called final moving representation which is usually simpler but not as parsimonious than the usual Echelon form. Furthermore, we proof that our specification strategy is consistent also in the case of cointegrated series. In order to show the potential usefulness of the method, we apply it to US interest rates and find that it generates forecasts superior to methods which do not allow for moving-average terms.

The bivariate probit model is frequently used for estimating the effect of an endogenous binary regressor (the "treatment") on a binary health outcome variable. This paper discusses simple modifications that maintain the probit assumption for the marginal distributions while introducing non-normal dependence using copulas. In an application of the copula bivariate probit model to the effect of insurance status on the absence of ambulatory health care expenditure, a model based on the Frank copula outperforms the standard bivariate probit model.

Traditional tools of welfare economics identify the envy-related welfare loss from conspicuous consumption only under very strong assumptions. Measured income and life satisfaction offers an alternative for estimating such consumption externalities. The approach is developed in the context of luxury car consumption (Ferraris and Porsches) in Switzerland. Results from household panel data and fixed effects panel regressions suggest that the prevalence of luxury cars in the municipality of residence has a negative impact on own income satisfaction.

Applications of zero-inflated count data models have proliferated in health economics. However, zero-inflated Poisson or zero-inflated negative binomial maximum likelihood estimators are not robust to misspeci*fication. This paper proposes Poisson quasi-likelihood estimators as an alternative. These estimators are consistent in the presence of excess zeros without having to specify the full distribution. The advantages of the Poisson quasi-likelihood approach are illustrated in a series of Monte Carlo simulations and in an application to the demand for health services.

We propose a direct measure of altruism between parents and adult children, using survey data on happiness from the German Socio-Economic Panel for the years 2000–2004. The question of altruism within families has policy relevance, for example, to understand whether public transfers crowd out private ones. Previous empirical evidence, based on observed transfer behavior, has failed to establish a clear consensus. Using various cross section, panel data, and instrumental variable estimators, we find a robust association between the happiness of parents and that of their adult children. A 1 standard deviation increase in a child’s happiness is associated with the same increase in own happiness as that of a 20–45% increase in household income, depending on specification.

The paper reconsiders existing estimators for the panel data fixed effects ordered logit model, including one that has not been used in econometric studies before, and studies the small sample properties of these estimators in a series of Monte Carlo simulations. There are two main findings. First, we show that some of the estimators used in the literature are inconsistent. Second, the new estimator seems to be more immune to small sample bias than other consistent estimators and is easy to implement. The empirical relevance is illustrated in an application to the effect of unemployment on happiness. Choosing the right estimator avoids a bias of up to 30 percent in key parameters.

The finite-sample size and power properties of bootstrapped likelihood ratio system cointegration tests are investigated via Monte Carlo simulations when the true lag order of the data generating process is unknown. Recursive bootstrap schemes are employed which differ in the way in which the lag order is chosen. The order is estimated by minimizing different information criteria and by combining the corresponding order estimates. It is found that, in comparison to the standard asymptotic likelihood ratio test based on an estimated lag order, bootstrapping can lead to improvements in small samples even when the true lag order is unknown, while the power loss is moderate.

The paper re-examines existing estimators for the panel data fixed effects ordered logit model, proposes a new one, and studies the sampling properties of these estimators in a series of Monte Carlo simulations. There are two main findings. First, we show that some of the estimators used in the literature are inconsistent, and provide reasons for the inconsistency. Second, the new estimator is never outperformed by the others, seems to be substantially more immune to small sample bias than other consistent estimators, and is easy to implement. The empirical relevance is illustrated in an application to the effect of unemployment on life satisfaction.

This article proposes a new class of rating scale models, which merges advantages and overcomes shortcomings of the traditional linear and ordered latent regression models. Both parametric and semi-parametric estimation is considered. The insights of an empirical application to satisfaction data are threefold. First, the methods are easily implementable in standard statistical software. Second, the non-linear model allows for flexible marginal effects, and predicted means respect the boundaries of the dependent variable. Third, average marginal effects are similar to ordinary least squares estimates.