ABSTRACT
The need for familiarity with microeconometric techniques in applied economics has arisen as a result of the increasing use of individual-level data sources in the analysis of economic behaviour. This not only reflects the growing availability of such data but also our increasing ability, as applied economists, to utilize microlevel data sources effectively on microcomputers. Whether it be the study of household behaviour or the study of firms’ behaviour, analysis based at the ‘individual’ level is persuasive. It avoids the problem of aggregation bias as well as identifying the factors that lead to a different distribution and usage of resources across households and firms. However, in microeconometrics distinct issues relating to endogenous selection, censoring and individual heterogeneity have to be faced. Indeed, it may often be difficult to draw useful inferences from cross-section data, especially where history dependence or state dependence is important. Longitudinal data or panel data which follow individual economic agents through time can be utilized so as to avoid some of these drawbacks, and in addition panel data combine the attractive features of time series analysis with individual-level behaviour. Our purpose in this chapter is to provide the applied economist with sufficient background on modern microeconometrics to choose appropriate estimators so as to exploit both the data source and the economic model.
