Econometric model

An econometric model is a statistical representation of economic relationships, combining economic theory with data and assumptions to estimate parameters, test hypotheses, or generate forecasts.

Regression analysis: A statistical method for estimating relationships between an outcome variable and one or more explanatory variables. Many econometric models estimate their parameters through regression.
Ordinary least squares: An estimator that chooses regression coefficients to minimize the sum of squared residuals. It provides a common way to fit linear econometric models.
Causal inference: The study of methods for estimating how outcomes change when causes or interventions change. Econometric models can estimate causal effects when their identifying assumptions hold.
Time-series model: A statistical model that represents patterns and dependence in observations ordered over time. Unlike many econometric models, it can prioritize temporal patterns without explicit economic theory.
Random variable: A variable whose possible values are governed by a probability distribution. Random variables represent uncertainty in economic outcomes and model disturbances.
Maximum likelihood estimation: A method that selects parameter values maximizing the probability of the observed data under a model. It estimates models by specifying how the data are distributed.
Difference-in-differences: A research design that compares outcome changes across treated and comparison groups before and after an intervention. It is a model-based approach to estimating policy effects from panel or repeated data.
Machine learning: A set of computational methods that learn patterns from data to make predictions or decisions. Prediction-focused methods can differ from econometrics' emphasis on interpretable parameters and inference.
Parameter: A numerical quantity that specifies a feature of a mathematical or statistical model. Model parameters quantify the strengths of the economic relationships being estimated.
Instrumental variables: A method that uses variables related to an endogenous regressor but excluded from the outcome equation's error. Instruments can identify relationships when regressors are correlated with the error term.

Linked from 8 pages