Assignment

Question 2 (50 marks)

2. Re-estimate the model (2 marks)

pricei=β0+β1rdi+β2reptrifectai+β3(rdi×reptrifectai)+β4totshareioi+β5totshareio2i+uipricei=β0+β1rdi+β2reptrifectai+β3(rdi×reptrifectai)+β4totshareioi+β5totshareioi2+ui

1. Write down and interpret an equation for the marginal effect of totshareioitotshareioi (5 marks)

iii.         Plot the marginal effect.  Label your plot. (3 marks)

1. Evaluate the marginal effect at the minimum, mean and max level of totshareio in your mydata dataframe. (4 marks)
2. Calculate the ceteris paribus expected change in residential electricity prices when the total share of investor-owned companies increases from 50 to 70  percent. Describe your working.   (5 marks)

For the remainder of this question, you need to use the “full_data” data frame which includes 72,425 observations.

1. Check whether this a balanced panel dataset and report your findings (2 marks)
2. Write down an equation for a regression model explaining residential electricity prices as a function of revenue decoupling without fixed or random effects. (5 marks)
3. Write down another equation which adds utility company fixed effects. (1 marks)

iii.         Explain what utility company fixed effects represent in the model.  What types of variables are they able to capture? (6 marks)

1. Explain what types of regressors cannot be included in a model with utility company fixed effects and why. (5 marks)
2. Estimate the regression equations that you have written down in parts i) and ii).
3. Interpret your regression outputs. (6 marks)
4. Test the joint significance of the entity fixed effects and report your findings. (3 marks)

iii.         Add time fixed effect to your model and test their joint significance.  What is your conclusion? (1 marks)