Current location - Education and Training Encyclopedia - Graduation thesis - Does empirical analysis have to be a model?
Does empirical analysis have to be a model?
Generally speaking, it is necessary. Demonstration+modeling is the most important part of the paper.

Empirical analysis can be divided into several levels according to the degree of difficulty:

The first level, simple charts and indicators, and general questionnaire results will be displayed in this way, with vivid images.

The second level, descriptive statistics, analyzes the characteristics of data distribution.

The third level, econometric analysis, establishes the model. Econometric analysis of extended data can be divided into several levels. The first level is simple regression, including binary and multivariate regression, and the handling of basic econometric problems (* * * linearity, heteroscedasticity and autocorrelation); The second level is more professional, including model specification error checking and model correction, model selection and processing of special data types (time series, dummy variables, panel data, etc.). ), simultaneous equations, VEC model, VAR model, conditional heteroscedasticity model, etc. The third level includes ordered dependent variable, panel VAR, neural network, quantile model and seasonal adjustment model.