PYXIDA Institutional Repository
and Digital Library
Collections :

Title :Essays in applied microeconometrics : I. Discrete deforestation with spatial interactions II. Employment transitions of older persons in Britain III. Sibling post-schooling educational attainment
Creator :Emmanouilidi, Elpianna
Contributor :Kyriazidou, Ekaterini (Επιβλέπων καθηγητής)
Athens University of Economics and Business, Department of Economics (Degree granting institution)
Type :Text
Extent :145p.
Language :en
Abstract :Chapter one is an environmental application investigating land use changes of forests and semi-natural areas in the Greek region of Western Attica between 1990 and 2000. Its objective is to estimate the spatial equilibrium distribution of individual deforestation actions and determine the degree of coordination in individual behavior. For this purpose, a virtual economic network of 156 agents has been created by laying an ad hoc square grid over the region. Next, dominant forest land use changes have been determined in each land parcel using CORINE land cover maps for the years 1990 and 2000. The economic model is a discrete choice model with endogenous spatial interactions. Even though spatial interactions produce multiple equilibria, the present research proposes a two-stage fixed point estimator yielding a unique solution. Empirical findings suggest that equilibrium deforestation actions are strategic substitutes for the environment and complements for agriculture, and are characterized by a relative lack of coordination in individual behavior. Chapter two uses a dynamic panel in order to study the employment transitions of older persons in Britain in a discrete-time discrete-choice setting in order to identify both observable and unobservable factors which might influence both their labour market participation and status. For this purpose, a sample of 1097 individuals, drawn from the first five waves of the English Longitudinal Study of Ageing (ELSA), born 1946-1956 and aged 45-65 during the 10-year sampling period 2002-2011, is used. Three mutually exclusive employment states are considered with reference to non-participation, namely: full-time employment, part-time employment and self- employment. There is evidence that the 45-65 ELSA respondents under study exhibit significant mobility as they age. Their employment dynamics are characterized by significant positive first order state dependence in individual preferences for work. The long-run determinants of this unobserved heterogeneity indicate, among others, that self-perceived poor general health causes a negative selection out of the more flexible employment states while other household income has a negative impact for the more flexible employment states as opposed to full-time employment. The long- run effects are lower and asymmetric compared to the initial conditions. Next, the study investigates the determinants of the labour market participation of those sample respondents who appear in the sample along with their partner or spouse as well as the presence of any duration dependence in their choices. There is evidence that good as opposed to poor self-perceived health has a positive influence on participation irrespective of the partner's activity status while the effects are asymmetric and twice as high in case of poor health when the partner is inactive. Moreover, couple duration dependence is negative and increases with the length of the spell. Chapter three uses data regarding the children of the 1979 batch cohort of the National Longitudinal Survey of Youth (NLSY79) mothers, namely the Children and Young Adults (CYA) section, in order to investigate, on the one hand, the determinants of sibling post-schooling educational attainment, as measured by highest grade completion and college enrollment, and, on the other hand, the degree of resemblance therein. Empirical interest in sibling educational attainment and resemblance became systematic in the 1970s in the United States, when both sociologists and economists focused on the family as a collective economic decision unit, allocating resources in a competitive way which reduces inequality. In the present context, post-schooling educational achievement is identified by common family background factors as well as sibling own past and present characteristics. Attainment is examined using panel data with family-specific (mother) effects. Siblings are ordered in reverse birth order with a maximum birth lag of 10 years, hence the sibling index stands for the time dimension in a short panel. Next, first order reciprocal resemblance is examined using simultaneous equations methods for sibling data grouped on the mother, namely: 3SLS and endogenous Poisson estimation. Empirical evidence regarding post-schooling educational attainment indicates that sex and race have a positive effect apart from highest grade completion in case of Hispanics. In line with the empirical evidence, family income below the poverty level has a clear negative effect. Moreover, the childhood predictor of future academic achievement has a positive effect, as opposed to the behavioural problems scores. Similar to having experienced a young marriage, current behavioural risk, as measured by the heavy use of marijuana, has a clear negative effect. As far as first order reciprocal resemblance is concerned, estimates are positive but asymmetric and higher in the direction lag-one older to younger sibling, increasing with sibship size in case of linear system as opposed to endogenous Poisson estimation. In conclusion, there seems to be evidence in favour of resource dilution in poor families and confluence in non-poor families.
Subject :Deforestation
Spatial interactions
Employment transitions
Sibling post-schooling educational attainment
Date :05-2013
Licence :

File: Emmanouilidi_2013.pdf

Type: application/pdf