Researchers have great flexibility in determining how much economic theory should be incorporated into the design and analysis.
The definition of structural econometrics is generally best approached by first understanding the two terms individually in the context of economics. The term econometrics generally means the fusion of economic theory and statistical methods when analyzing data. Structural commonly refers to estimation, which means the deliberate application of economic theory in modeling empirical studies. Therefore, structural econometrics is often defined as empirical studies that incorporate economic theory to model and analyze the results. Some economists find the method useful for drawing concise conclusions between economic, statistical, and institutional relationships and assumptions.
Structural modeling may vary considering when a research project is designed. Researchers have great flexibility in determining how much economic theory should be incorporated into the design and analysis. They also have great flexibility in determining how much to rely on statistical assumptions, and this ultimate usefulness of the method is often based on trade-offs. Disagreement about the options available and in which situations to select these options is routine, although the compromises made in research projects are often debated, as there are no concrete rules. Still, the focus on structural econometrics seems to offer some clear advantages.
Implicitly it results in linking statistical models and economic theory in non-structural approaches, with economic theory often not even present in university econometrics courses. It is explicitly the goal of the structural approach to persuade researchers to make connections between economics, statistics, and the real world. Therefore, structural econometrics potentially offers some distinct advantages.
The estimation of structural parameters is one of these advantages, while allowing the use of counter-experiments, simulations and statistical comparison. It also allows the comparison of different theories applied to a research project, while the explicit assumptions also provide a deeper insight into the mechanisms that impact the results. The application of structural econometrics is often used in game theory projects to understand market supply and demand, as well as a variety of other research endeavors where explicit results are helpful rather than a hindrance.
Nonstructural approaches, however, do one of three things: rely on statistics will have little or no contribution from economic theory, rely on economic theory with little or no contribution from statistics, or incorporate a minimal fusion of the two. . Examples of nonstructural studies include forecasting, which is based on statistics, and measurement studies such as GDP, which is based on economic theory. Policy evaluation infuses economic theory and statistics by estimating causal effects, but is not considered structural econometrics because it generally makes minimal assumptions. Conceptually, this means that nonstructural and structural work are fundamentally different, but not always in application, as the lines blur considerably.