Pseudo Panels Research Articles

Panel data or pseudo panels for longitudinal research? Cross-national comparisons using the example of firms' training spend

PurposeThe evolution of firm level practices over time has always been a keen area of interest for management scholars. However, in comparison to other social scientists, particularly economists, the relative dearth of firm level panel data sets has restricted the methodological options for exploring inter-temporal changes.Design/methodology/approachThis paper applies a pseudo panel methodology to investigate the evolution of training spend at the firm level over time.FindingsThe analysis is framed within a varieties of capitalism lens and by adopting a more meaningful approach to examining changes over time it leads us to question some of the “truisms” linked to firms expected behaviours within different national institutional frameworks.Research limitations/implicationsAs with any large-scale quantitative analysis, it would always benefits from a larger number of observations and/or a longer time period, in this instance access to annual data rather than 4 or 5 year intervals would have been helpful.Practical implicationsBy adopting a different, and more appropriate, approach to analysing existing cross-sectional data over time this empirical research helps to achieve a deeper understanding of the complex issues that influence decision making at the firm level.Social implicationsAt the firm level, in line with the practical implications above, this will enable decision makers to achieve a deeper understanding of the evolution of the external context in which they operate and the likely influence of that evolution within their own organisation.Originality/valueThis approach enables a more meaningful exploration of inter-temporal changes in situations where longitudinal data does not exist.

Introducing synthetic pseudo panels: application to transport behaviour dynamics

We propose a new approach for constructing synthetic pseudo-panel data from cross-sectional data. The pseudo panel and the preferences it intends to describe is constructed at the individual level and is not affected by aggregation bias across cohorts. This is accomplished by creating a high-dimensional probabilistic model representation of the entire data set, which allows sampling from the probabilistic model in such a way that all of the intrinsic correlation properties of the original data are preserved. The key to this is the use of deep learning algorithms based on the Conditional Variational Autoencoder (CVAE) framework. From a modelling perspective, the concept of a model-based resampling creates a number of opportunities in that data can be organized and constructed to serve very specific needs of which the forming of heterogeneous pseudo panels represents one. The advantage, in that respect, is the ability to trade a serious aggregation bias (when aggregating into cohorts) for an unsystematic noise disturbance. Moreover, the approach makes it possible to explore high-dimensional sparse preference distributions and their linkage to individual specific characteristics, which is not possible if applying traditional pseudo-panel methods. We use the presented approach to reveal the dynamics of transport preferences for a fixed pseudo panel of individuals based on a large Danish cross-sectional data set covering the period from 2006 to 2016. The model is also utilized to classify individuals into 'slow' and 'fast' movers with respect to the speed at which their preferences change over time. It is found that the prototypical fast mover is a young woman who lives as a single in a large city whereas the typical slow mover is a middle-aged man with high income from a nuclear family who lives in a detached house outside a city.

Dynamic profiling through repeated surveys: a customer satisfaction study

Customer Satisfaction management has been long studied as a cross-sectional phenomenon. Today, a shift in emphasis from a static to a dynamic approach appears to be a worthwhile change, although the design and monitoring of panel data is not simple to achieve. In our paper, we propose the usage of pseudo panels, which are less costly and easy to build with the available data for the firms. Our approach is based on a model-free technique: Dual Multi- ple Factor Analysis. The synthesis of the multivariate table is visualised on a common space that sheds light on customers’ trajectories of satisfaction. A real case study of an Italian bank is illustrated. Results related to the dynamic profiling of the clients highlights highly different behaviours that warn the management of the future trends of the services.

Informality and Mobility in the Labor Market: A pseudopanel's approach

One of the main limitations of labor market analysis in developing countries is the lack of appropriate panel data information. This paper extends the methodology of Dang et al. (2014) to examine labor market mobility between the formal and informal sectors in Bolivia and Colombia building consistent pseudo panels from repeated cross-sectional survey data. After testing the robustness of the methodology, we identified confidence intervals for mobility’s group in Bolivia and Colombia (formal to formal, informal to informal, formal to informal and informal to formal). The results provide evidence that in Bolivia and Colombia mobility between labor sectors is relatively low, which can explain the low variability of the informality rate in those countries. Results suggest that the number of people who move to the informal sector tends to be larger than those who scape from the informal market to the formal, or at least there is more variability across years for the first ones. This implies that public policies focused on labor market in Latin America would not lead to significant improvements if mobility patterns do not change.

Pseudo Panel Data Models With Cohort Interactive Effects

When genuine panel data samples are not available, repeated cross-sectional surveys can be used to form so-called pseudo panels. In this article, we investigate the properties of linear pseudo panel data estimators with fixed number of cohorts and time observations. We extend standard linear pseudo panel data setup to models with factor residuals by adapting the quasi-differencing approach developed for genuine panels. In a Monte Carlo study, we find that the proposed procedure has good finite sample properties in situations with endogeneity, cohort interactive effects, and near nonidentification. Finally, as an illustration the proposed method is applied to data from Ecuador to study labor supply elasticity. Supplementary materials for this article are available online.

Preferences for redistribution in Europe

This paper uses pseudo panel techniques and a fixed effects estimator to analyse the determinants of preferences for redistribution in 34 European countries over the period 2002–2012. The data is drawn from the six available waves of the European Social Survey. The main result is that changes in income inequality positively affect changes in preferences for redistribution over time. Though this result is predicted by standard political economy models, it has found little previous empirical support. This study shows that, at least in Europe, growing income inequality leads to more individual support for redistribution. The empirical results hold after performing a variety of robustness checks regarding the construction of pseudo panels, the use of lags and different measures of income inequality.

Consistent Estimation in Pseudo Panels in the Presence of Selection Bias

Abstract In the presence of selection bias the traditional estimators for pseudo panel data models are inconsistent. This paper discusses a method to achieve consistency in static linear pseudo panels in the presence of selection bias and a testing procedure for sample selection bias. The authors’ approach uses a bias correction term proportional to the inverse Mills ratio with argument equal to the “normit” of a consistent estimation of the conditional probability of being observed given cohort membership. Monte Carlo analysis shows the test does not reject the null for fixed T at a 5% significance level in finite samples. As a “side effect” the authors utilize the enlarged pseudo panel to provide a GMM consistent estimation of the pseudo panel parameters under rejection of the null and apply the procedure to estimate the rate of return to education in Colombia.

Consistent Estimation of Pseudo Panels in the Presence of Selection Bias

In the presence of selection bias, traditional estimators of pseudo panel data are inconsistent. In this paper, the authors derive the conditions under which consistence is achieved in pseudo-panel estimation and propose a simple test of selection bias. Specifically, they propose a Wald test for the null hypothesis that there is no selection bias. Under rejection of the null hypothesis, the authors can consistently estimate pseudo-panel parameters. They use cross sections and pseudo-panel regressions to test for selection bias and estimate the returns to education in Colombia. The authors corroborate the existence of selection bias and find that returns to education are around twenty percent.

Keeping ARMS relevant: extracting additional information from ARMS

PurposeThe purpose of this paper is to discuss opportunities to obtain more information from the Agricultural Resource Management Survey (ARMS). Specifically, the paper will explore the issue of survey nonresponse, the development of pseudo panels, and more frequent updating of cost of production data on an enterprise basis.\Design/methodology/approachResearchers from the Land Grant University System and the Economic Research Service have relied on ARMS to evaluate the effect of agricultural, macroeconomic, and other factors on the US farm sector, farm businesses, and the households that manage them. This paper will identify gaps in understanding and proposes approaches to extract additional information from ARMS.FindingsThe relevance of ARMS in the future will depend on the ability to continue to understand potential pitfalls and areas of additional research that can develop new procedures to extract additional information. Three issues which are in need of further study include continuing to examine the issue of non‐response, refining methods to develop pseudo panel data, and examining methods to develop commodity specific financial information between the commodity specific surveys.Originality/valueThe National Research Council completed a review of ARMS to address challenges in keeping the survey relevant into the future. However, research that examines the construction of financial statements and other information had not been conducted since the early 1990s. This study fills part of that gap.

Risk attitudes and premiums of U.S. corn and soybean producers: an empirical investigation

The 2000 Agricultural Risk Protection Act and 2002 Farm Security and Rural Investment Act reduced price and yield risks faced by many U.S. crop producers to low levels. We use a non-structural methodology extended for application to pseudo panels and national survey data to examine the risk attitudes of U.S. corn and soybean producers to test whether, and examine how, risk attitudes varied during the 1996–2001 and 2002–2008 periods by revenue category. We cannot reject the hypothesis of risk neutrality for the entire population, and for each revenue category, for the former period, but can reject risk neutrality, in favor of risk tolerance, for the entire population and for the larger revenue categories for the latter period. Estimated risk premiums for the latter period suggest that U.S. corn and soybean farmers who earn more require larger payments to remain indifferent between receiving their expected income with certainty and receiving an uncertain income from farming and government programs.

Pseudo panels as an alternative study design

Marketing academics and practitioners increasingly recognise the importance of studying the dynamic nature of consumer attitudes and behaviour. However, recent works highlight a dearth of longitudinal studies into consumer dynamics published in marketing academic literature ( Leonidou et al., 2010; Rindfleisch et al., 2008; Williams and Plouffe, 2007). In an attempt to address this gap, this article evaluates the ability of available research designs to meet various research objectives in the study of consumer dynamics. This evaluation highlights the need for a technique capable of modelling gross and individual level change using repeated cross-sectional data. The article proposes the use of pseudo panels to address this gap and advances the use of data fusion techniques for matching independent samples over multiple time periods to create these pseudo panels.

Pseudo Panels and Repeated Cross-Sections

textabstractIn many countries there is a lack of genuine panel data where specific individuals or firms are followed over time. However, repeated cross-sectional surveys may be available, where a random sample is taken from the population at consecutive points in time. In this paper we discuss the identification and estimation of panel data models from repeated cross sections. In particular, attention will be paid to linear models with fixed individual effects, to models containing lagged dependent variables and to discrete choice models.

Unbalanced panel data: A survey

This paper surveys the econometrics literature on unbalanced panels. This includes panels with randomly and non-randomly missing observations. In addition, we survey panels with special features including pseudo panels, rotating panels and censored panels.

A Note on Dynamic Modelling from Short and Heterogeneous Pseudo Panels

In this paper we note the pitfall of cohort‐aggregated dynamic modelling from short pseudo panels when the slope parameters vary across cohorts, and this is not allowed for in estimation. By contrast, quasi‐differencing across pairs of individuals that belong to the same cohort is shown to help circumvent this problem. Some Monte Carlo simulations are then conducted to assess the relative merits of the two types of estimators under alternative data generating mechanisms.