You are consulting for a Buffalo, NY realty company. They provide a data set of 100 homes sold...

Question

You are consulting for a Buffalo, NY realty company. They provide a data set of 100 homes sold within the last year in a Buffalo, NY suburb. The variables included are:SalePrice: The price at which the houses sold, to the nearest $1000LotSize: The size of the lot, in acresHouseArea: The size of the house, in square feetGarage: Number of garage baysBasement: 0 = no basement, 1 = basementBasementArea: Size of basement (in square feet)FinishedBasement: 0 = unfinished, 1 = finishedRanch: 1 = Ranch-style house, 0 = not Ranch-styleNumBedrooms: Number of bedroomsNumBathrooms: Number of bathroomsMainFlooring: Main type of flooring in house (Carpet or Hard)Fence: Type of fence installed (No, Privacy, or Other)Corner: Is the house on a corner lot, Yes or NoMainRoad: Is the house on a main road, Yes or NoKitchen: Realtor's rating of kitchen: Great, Good, Average, Below Average, or PoorBathrooms: Realtor's rating of bathrooms: Great, Good, Average, Below Average, or Poor
Build a regression model to predict the SalePrice of the houses. (Hint: the best models I have seen previously have Adjusted-R2 a little over XXXXXXXXXXYou should use methods from this week, and describe how you decided on the variables to include in your model.Explain why you do not need to use all given variables in your model. (Hint: consider multicollinearity.)Interpret a few of your parameter estimates. Do all of the parameter estimates make sense, or are there some that have unexpected values?Predict, along with 95% prediction intervals, the prices of the five houses at the bottom of the data set (houses XXXXXXXXXXAre you concerned about the predictions for any of the houses?The following characteristics might be changed by homeowners: MainFlooring, Fence, Kitchen quality, and Bathroom quality. Explain which of these have an effect on the sale price of the house, and which do not. (Hint: use the adjusted-R2 shortcut, or if you are very ambitious, try partial F tests)Write a case report summarizing your findings.Upload your case report by Sunday night.

house-prices-iiuahm05.xls 0-0mjxpxkg.jpg 1-ysah0kyo.jpg

Monica · Accepted Answer

Given
The data set of 100 homes sold within the last year in a Buffalo, NY suburb
The variables included are
SalePrice: The price at which the houses sold, to the nearest $1000
LotSize: The size of the lot, in acres.
HouseArea : The size of the house, in square feet.
Garage: Number of garage bays
Basement : 0 = no basement, 1 = basement
Basement Area : Size of basement (in square feet)
FinishedBasement : 0 = unfinished, 1 = finished
Ranch: 1 = Ranch – style house, 0 = not ranch – style
NumberBathrooms: Number of bathrooms
MainFlooring: Main type of flooring in house ( Carpet  or Hard )
Fence: Type of fence installed ( No, Privacy, or Other )
Corner: Is the house on a corner lot, Yes or No
MainRoad: Is the house on a main road, Yes or No
Kitchen: Realtor’s rating of kitchen. Great, Good, Average, Below Average, or Poor
Bathrooms: Realtor’s rating of bathrooms, Great, Good, Average, Below Average, or Poor
It is asked to build a regression model for the Salesprice of the house.
The sales price is the independent variable which is dependent on other given independent variable.
Based on Multicollinerity we can decide what variables need to include and what are not.
Multicollinerity
Ragnar Frisch first used the term multicolinerity.The term describes an exact relationship between regression exploratory variables. A linear regression analysis assumes that there are no perfect exact relationships between exploratory variables.  Multicollinearity occurs when this assumption is violated in regression analysis.
1. There is no Multicollinearity in the data when the regression exploratory variables have no relationship with each other.
2. It is a type of low Multicollinearity when there is a relationship among exploratory variables but that relationship is very low.
3. When the relationship among exploratory variables is moderate, it can be described as moderate multicollinearity.
4. In the case of high multicollinearity, the relationship between the exploratory variables is high or their correlation is perfect.
5. There is a problem of very high Multicollinearity when there is a perfect correlation among exploratory variables, which should be removed from the data when regression analysis is performed.
First we have to label the give values into number in order to do the further calculations.
Consider the below column we the numbers are assigned to the following  data set values which are present in the data.
	No
	0
	Yes
	1
	Carpet
	1
	Hard
	0
	Privacy
	0
	Other
	1
	No
	0
	Great
	4
	Below Average
	1
	Average
	2
	Good
	3
	Poor
	0
Now Press Ctrl + F and go to the “Replace tab” and replace each value with above given digits respectively.

You are consulting for a Buffalo, NY realty company. They provide a data set of 100 homes sold within the last year in a Buffalo, NY suburb. The variables included are:SalePrice: The price at which...

Solution

Answer To This Question Is Available To Download

Related Questions & Answers

Submit New Assignment