Home - Yuan-032/Econometrics_pre GitHub Wiki

Introduction: introduce the research question and give a brief overview of your methodology and findings

Our group found the relationship between minimum wage and unemployment rate very interesting, and we thought it would be worthwhile to filter this relationship by state. We wanted to test if unemployment and minimum wage had a positive relationship, meaning that an increase in minimum wage would result in a decrease in unemployment. This is better tested by regressing the change in unemployment by the change in minimum wage.

Data: describe where your data are sourced from, what the main variables of interest are, and do some preliminary data analysis (e.g. plotting) as appropriate

Methodology: formulate your research question as an econometric model (e.g. a causal model or a BLP model) and explain how you will go about estimating the parameters of interest. Here you should be careful to describe whether or not certain assumptions key to your analysis (e.g. A1, or conditional mean independence for control variables, etc) are plausibly satisfied.

Estimation and inference: estimate the model, perform whichever hypothesis tests are appropriate for your analysis, and interpret the results within the context of your research question. For instance, do the signs and magnitudes of coefficient estimates make economic sense? Do the data support your research question or provide evidence against it? If you have some surprising estimates, what could it be that drives these results?

The models above show that our hypothesis on the positive relationship between Unemployment Rate and Minimum Wage holds true. Model 1 supports this by regressing Unemployment on Minimum Wage in 2020 dollars, showing that an increase in minimum wage of about $0.64 would raise unemployment by 1%. Model 2 continues by splitting the data into three decades, represented by different dummy variables and coefficients. The overall result still demonstrates a positive correlation between our two main variables. Model 3 is similar to Model 1, but splits unemployment rates and minimum wages down to a state level and finda the same result, this time with the coefficient at close to $0.65. Lastly, Model 4 is an estimate that incorporates the poverty rate into the regression model in an attempt to eliminate some endogeneity bias. This also supports our hypothesis by coming up with a slightly reduced coefficient of minimum wage, as well as including a coefficient that shows one's position in relation to the poverty line and that variable’s effect on unemployment.

Conclusion: brief recap and summary of findings.