Content

Suppose, that you are commissioned to develop a forecasting model for residental property selling price in an Adelaide suburb. The sample information availble to you is collected from a credible source on 176 houses recently sold in the suburb.

- SalePrice (Y): The transaction price in dollars, as recorded on the contract of sale.

-LandArea (x^1): The size of the land area in square metres.

Rooms(x^2): The number of main rooms in a dwelling. Main rooms include bedrooms, living areas, and kitchens, but exclude bathrooms, laundries, and water closets. Groups of small rooms that are not fully enclosed may be counted as one room (e.g. kitchen-dinettes).

EQM(^3): Recorded in square metres. This is a weighted average of all buildings on the property (Equivalent Main Area). The main dwelling will have a weighting of 1, lesser structures such as carports will have a lower weighting. These weightings are not fixed and depend upon the discretion of the assessor. The overall value is a reasonable proxy for the building area.

Condition(C): This is a subjective condition code with the value 0 indicating some minor structural problems or the need for repairs, the value 1 indicating good condition and the value 2 indicating excellent condition.

YearsBuilt (x^4): The number of years since the house was built.

Question 1-

Create a dummy variable named D for house condition such that it takes the value 1 if the original condition variable C taking the values 0 and 1, otherwise takes the value 0. Suppose that you begin with exploring a regression model, In (Y) = a + B,In (X,)+ A,*+ AXs+ BX+ + BD+e. Attach an Excel file that contains your regression output of the model and write down the estimated model. 10 marksAfter some statistical investigation you may select a regression model Y = a + BIX, + BzXz + ByX's + P4X3+ BD . Xi+ e. Attach an Excel file that contains your regression output of the model. Interpret the estimate of B 10 marks