Description
Econometrics assignment. There are four file for data Please make sure you can handle it and it is important to me
Each of the following 4 data sets contains the n iid pairs of sample data (y, x1)_1, (y, x1)_2, …, (y, x1)_n, randomly draw from the population space X and Y. Suppose the assumption 1-3 hold for all of the data set. Pick up the dataset with heteroskedasticity and use this data set only to answer the following questions. No need to answer this question. This question sets up the parameter of the following questions and serves as a separator. (Round your answer in 3 decimal Places as always).
Which data set exists heteroskedasticity
- 1
- 2
- 3
- 4
What are the bad consequences of heteroskedasticity
- estimator is no longer unbiased
- estimator is no longer consistent
- estimator is no longer efficient
- all of above
Suppose the variance of the uncertainty is a linear function of the x1. Use the algorithm in slides to implement a WLS estimation. What is your estimation of for each of the linear coefficient?
beta0 =
beta1 =
Which of the underlying theorem guarantees the WLS estimator is the most efficient linear estimator?
- Gauss Markov Theorem
- Law of Large Number
- Central Limit Theorem
- Spectral Theorem
If we conduct the OLS estimation, then the original Homoskedasticity standard deviation is no longer valid for statistical inference. Fill in the below the heteroskedasticity-robust standard deviation of each of the linear coefficients:
sd(beta0) = sd(beta1) =