- Explained Lessons 4.1 and 4.2, and
- concluded the practical work with the Beauty and hprice1 datasets
In the seminar, we solved this exercise sheet and did a forecasting exercise with the Beauty dataset.
The exercise sheet illustrated several ideas:
To prepare it you should:
The exercise sheet illustrated several ideas:
- Econometrics builds and generalizes over the results of previous statistics courses, for example because...
- ...the OLS estimate for the constant term of a model with no additional regressor is the sample mean of the endogenous variables, and also,
- ...assuming that all the variables are mean-centered, the OLS estimate for the slope of a model with a single regressor is related with standard measures of linear relationship such as the sample covariance or the sample correlation coefficient.
- When the variables in the model are re-scaled, some results may change, while others remain unaffected. The same happens when one adds or subtracts arbitrary constants to the value of the variable. These ideas are described mathematically in a short note
The goals for this week are:
- Explain Lessons 5.1 and 5.2
- In the extra seminar of next April 25th we will do a "dry run" for the final exam. It will consist in a short test with 10 questions about Lessons 1-3.
To prepare it you should:
- Review the material corresponding to Lessons 1, 2 and 3 and clarify any doubts
- Revise again the calculations in the exercises: Many least squares calculations done (almost) by hand and the last exercise sheet.
