got problem on doing this question. could someone help me doing this question in eviews for full answers. thank you very much!
here is the questions and the datas:

Background

Hedonic pricing models decompose the price of an item into separate components that determine the price. They are often applied to explain and predict house prices in terms of the size and location of the house and other factors.

For this assignment you will construct a hedonic pricing model to explain and predict house prices based on data for 546 houses that were sold in Windsor, Canada, in 1987.

The data is provided in the excel sheet HPRICE1.xls that is provided on the LMS webpage for Assignment 2.

The data set contains the following information (variables):

price:               sales price of the house measured in Canadian Dollars (CAD)

lot-siz:             lot size measured in square feet
bed:                 number of bedrooms in the house
bath:                number of bathrooms in the house
stor:                 number of storeys of the house
d_loc:              indicator for desirable location of house (1 if  desirable, 0 otherwise)
b_ment:           indicator for basement in the house (1 if basement, 0 otherwise)

Complete the following tasks

• 1. Present appropriate descriptive statistics for the quantitative variables that you have downloaded. Remember to only provide appropriate statistics, that is, do not provide tests or measures we have not discussed in this subject. Briefly describe in words what these descriptive statistics show, paying attention to the measures of central location and the range.
• 2. Present appropriate graphs of the relationships between house prices (our outcome of interest) and the four quantitative variables in the sample that theory and common sense would suggest will affect prices: lot size, number of bedrooms, number of bathrooms and number or storeys. Provide a brief description of what your graphs show. Do the relationships between the variables look linear or not?
• 3. Estimate the relationship between house prices and the four quantitative variables that you considered in part (ii) using OLS. Write down the equation for the population model you are estimating. Report the estimates for your model in a table. Provide a full interpretation in words of each of the slope coefficients you have estimated, paying particular attention to whether the relationships are significant or not. Finally, comment on the fit of the model you have estimated.
• 4. The data set contains two additional dummy variables that describe whether the house has a basement and whether it is in a desirable location. We would think that these are  factors that affect the price of a house and should therefore be included in the model. Before we include them in the model we should check for a relationship between house prices and each of these two variables. Note that since these are qualitative variables/dummy variables including them into the OLS regression has no slope effects. Coefficients on dummy variables only lead to a shift in the intercept. (Hence we don't look at scatter plots to check for a linear relationship).
• We can however test wether each of these dummy variables affects the location of distribution of house prices. For the assignment, just test for a difference in location of house prices for houses in a desirable location (population 1) and those not in a desirable location (population 2). Construct the appropriate test with the help of Eviews and report and interpret your results. At the end of assignment we provide some hints how to create samples for the two populations in Eviews.
• 5. Now re-estimate the relationship between house prices and all the six explanatory variables provided in the sample. Report the estimates for your model in a table. Provide a full interpretation of the slope coefficients for the two newly included nominal variables. Comment on the fit of the model you have just estimated and compare it to the fit of the model from part (iii). Are there any noteworthy changes in the results for the coefficients on the previously included variables?
  
• 6.  List ALL the required conditions for using Ordinary Least Squares. Which of these required conditions can you check for your particular model estimated in part (v)? For those required conditions that you cannot check, describe why they cannot be checked. For the required conditions you can check, provide evidence of whether the required conditions hold or not. For any required conditions that you do not believe to hold here, briefly describe the specific consequences for your OLS results.
• 7. Using your estimates from part (v), construct a point prediction and a 95 per cent prediction interval for the price of a house with the following fairly typical characteristics: lot size of 5000 square feet, 3 bedrooms, 1 bathroom, 2 storeys, no basement and not in desirable location. Then construct a 95 per cent prediction interval for the price of a house with the following characteristics: lot size of 10000 square feet, 4 bedrooms, 2 bathroom, 3 storeys, basement and in desirable location. Compare the two prediction intervals and briefly discuss what you find.
• 8. The model in part (v) assumes that the price effect of an additional storey does not depend on whether the house has a basement. To check whether the value of an additional storey depends on the existence of the basement, create an interaction variable (bment_stor = b_ment  x stor using the gen command in Eviews) and re-estimate the model from part (v) adding  the interaction variable.
• Report the coefficient estimates for the storey variable and the interaction variable.
• What is the average value of an additional storey for a house with no basement and      for a house with a basement? Briefly interpret your results.
• DATA:
  HPRICE1.xls
  大小:(55.5 KB)

   马上下载