This work is dedicated to the data analysis using the OLS technique.
The tasks is the following:
T1: Write down the model (equation) for regression analysis of what you are going to estimate. You specify the following: dependent variable = expected wage; independent variable (composite score of gendered attitudes). Estimate the model using R and interpret the results (use summary(model));
T2: Use the model in T1 and show it graphically (fit the line on the graph with s.e of the slope) using ggplot;
T3: Formulate a new multivariate model so that you add at least 2 independent (control) variables. (Hint! These can be gender, education, or other variables that you have in your dataset).
T4: Formulate a new (third) model and add one interaction in your model. (Hint! it is easier to make interactions with dummy variables)
T5: Compare all your three models (T1, T3, T4) using stargazer. Discuss the model fit — which is the best model. Choose your final model.
T6: Run some after estimation analysis on your final model. Discuss non-normality, non-linearity, and multicollinearity.
Set up the slides (presentation) to reveal your answers (visualizations are welcome) to the following questions:
1. How the final choice of the model (out of 3 model specifications) was done?
2. How statistically significant are your beta coefficients? What they show (their practical significance)?
3. Are your attitude composite variable and expected wages correlated or is there causality?
4. Do you see any statistically significant gender effects?
4. What kind of limitations your study has (think of T6)?
"Place your order now for a similar assignment and have exceptional work written by our team of experts, guaranteeing you A results."