Search

While pseudo‐panels are useful when only repeated cross‐section data are available, estimates are likely to be attenuated and suffer from sampling error if cell sizes (number of individuals grouped together in a cohort) are too few. However, there is no consensus on how large cell size needs to be, with recommendations ranging from 100 to several thousands. This is due to sampling error being affected by both cell size and three important types of variation in the cohort data (across and within cohorts and over time). We combine these into a single metric, called CAWAR, and demonstrate its relationship to sampling error using Monte Carlo simulations and an empirical application. We produce recommended values for CAWAR beyond which sampling error bias is minimal and from these one can easily calculate the required cell size.

We use a pseudo-panel data approach to analyze the relationship between the consumptions of cigarettes, alcohol, and coffee in a rational addiction framework. We find that while cigarette and coffee demands fit well with the rational addiction model, alcohol demand conforms to a model with inventory effects. The results suggest that alcohol consumption increases the marginal utility derived from consumption of cigarettes. Increasing alcohol prices would decrease not only the consumption of alcohol, but also the consumption of cigarettes. On the other hand, increasing cigarette prices do not have the same effect on consumption of alcohol. The cross-price elasticity of coffee with respect to cigarette price is positive and significant which suggests that coffee substitutes for cigarettes when cigarette prices increase. The cross-price elasticity of alcohol with respect to coffee price is found to be negative and significant. On the other hand, Morishima elasticities of substitution indicate that cigarette, alcohol and coffee substitute each other along the indifference curve when relative prices change.

The debate over the association between dietary patterns and obesity has not been settled in the literature. Some studies suggest that there are significant differences in mean body mass index across dietary patterns whereas other studies refute that result. The objective of this study is to test whether dietary patterns have a significant effect on body mass index in Korean adults when controlling for calorie intake and several sociodemographic factors. We present new evidence derived from pseudo panel data created from a series of cross-sections. Our results from the pseudo panel analysis show that some of the dietary patterns that were identified from the Korean adults’ food intake survey have a significant effect on body mass index. Specifically, males with the “beef and processed food” (P < 0.05), or “pork and alcohol” (P < 0.05) dietary pattern had significantly higher BMIs, whereas females with the “fast food” (P < 0.1) or “ramen and bakery” (P < 0.01) dietary pattern had higher BMI.

Summary This paper estimates a hedonic housing model based on flats sold in the city of Paris over the period 1990–2003. This is done using maximum likelihood estimation, taking into account the nested structure of the data. Paris is historically divided into 20 arrondissements, each divided into four quartiers (quarters), which in turn contain between 15 and 169 blocks (ilot, in French) per quartier. This is an unbalanced pseudo‒panel data containing 156,896 transactions. Despite the richness of the data, many neighborhood characteristics are not observed, and we attempt to capture these neighborhood spillover effects using a spatial lag model. Using likelihood ratio tests, we find significant spatial lag effects as well as significant nested random error effects. The empirical results show that the hedonic housing estimates and the corresponding marginal effects are affected by taking into account the nested aspects of the Paris housing data as well as the spatial neighborhood effects.Copyright © 2014 John Wiley & Sons, Ltd.

We investigate the nature and sources of the decline in the level of employment of working age males in Australia in recent decades, drawing on both Australian Bureau of Statistics labour force survey data and census data. Alternative measures of the male employment rate are considered before settling on two complementary measures: the full-time employment rate and the full-time equivalent employment rate. The latter measure weights part-time jobs according to the fraction of a full-time job they represent. We then go on to estimate models of the determinants of these two employment rates using data from the population censuses conducted between 1971 and 2001. We construct a pseudo panel by ‘stacking’ the seven census data sets (Deaton 1997, Kapteyn et al 2005). This facilitates the tracing of birth cohorts over time, in turn making it possible to control for cohort unobserved heterogeneity that may bias cross-sectional estimates of effects of other characteristics, in particular age and year/time period. We produce evidence that a number of factors have contributed to the decline in male employment, including growth in educational enrolment and attainment, the decline in couple households with dependent children, growth in income taxes and welfare replacement rates, growth in labour productivity and changes in the structure of labour demand away from traditionally male-dominated industries. Significantly, we find that, all else (observable) constant, more recent birth cohorts have higher employment rates than earlier birth cohorts.

This paper deals with the estimation of technical efficiency using pseudo panel data constructed from rotating panel data on Swedish crop farmers where each unit is observed for a maximum of four years. Farms are grouped into 89 cohorts, using their common time-invariant characteristics, and the cohorts are used as an individual unit in the empirical analysis. The overall technical inefficiency is decomposed into a persistent and a residual component. An alternative model with time-varying technical inefficiency is also considered. In addition to technical inefficiency, measures of returns to scale and technical change are also obtained.