Defining structural econometrics is usually best approached by first understanding both terms individually in the context of economics. The term econometrics usually means the fusion of economic theory and statistical methods when analyzing data. Structural commonly refers to estimation, meaning the deliberate application of economic theory in the modeling of empirical studies. Thus, structural econometrics is often defined as empirical studies that incorporate economic theory for modeling and analysis of the results. Some economists find the method useful for drawing concise conclusions between relationships and economical, statistical and institutional assumptions.
Structural modeling can vary considering when a research project is designed. Researchers have a great deal of flexibility in determining how much economic theory to incorporate into the design and analysis. They also have a great deal of flexibility in determining how much to rely on statistical assumptions, and this final utility of method is usually based on compromise. Disagreement about the choices available and in what situations to select those choices is routine, while often debating compromises made on research projects since there are no concrete rules. Still, focusing on structural econometrics seems to offer some distinct advantages.
Implicitly results in the linking of statistical models and economic theory in non-structural approaches, with economic theory often not even present in econometrics courses in college. Explicitly is the objective with the structural approach, coaxing researchers to make connections between economics, statistics and the real world. Therefore, structural econometrics potentially provides some distinct advantages.
Estimating structural parameters is one such advantage, while allowing for the use of counter experiments, simulations and comparison of statistics. It also allows for the comparison of different theories as applied to a research project, while explicit assumptions will also provide deeper insights into the mechanisms that impact the results. 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 efforts where explicit results are helpful, rather than a hindrance.
Non-structural approaches, however, do one of three things: rely on statistics will little or no input from economic theory, rely on economic theory with little or no input from statistics or incorporates a minimal fusion of the two. Examples of non-structural studies include forecasting, which relies on statistics and measurement studies like GDP, which relies on economic theory.Policy evaluation infuses both economic theory and statistics by estimating casual affects, but is not considered structural econometrics because it usually makes minimal assumptions. From a conceptual point-of-view, this means non-structural and structural work are fundamentally different, but not always in application since the lines blur considerably.