Homework 2 - mh55345 (1)
pdf
keyboard_arrow_up
School
Nassau Community College *
*We aren’t endorsed by this school
Course
101
Subject
Industrial Engineering
Date
Dec 6, 2023
Type
Pages
2
Uploaded by marisolh5164
●
Baths
: The number of bathrooms
●
Beds:
The number of bedrooms
●
Area
: The livable area, in square feet
●
Fireplaces:
The number of fireplaces
●
Garage_Cars:
The number of cars that fit in the garage
●
Year_Built
: The year the home was built
●
Sale_Price:
The sales price of the home, in dollars
Answer each question using full sentences: do not simply write down a number or paste
in R output. (However, when appropriate, you can paste in an R output that supports the
statements you are making.) Upload a PDF of your responses as well as an R script
that replicates the calculations you have done.
Create a model to predict the sales price of a home from the number of bedrooms,
fireplaces, and the year it was built.
1.
Interpret the intercept of this model. Does the intercept have any meaning in
this context? Explain.
The intercept of this model is the sales price when all dependent variables are at
zero. The intercept in this case is -2499047.196. There is no meaning in this
context Year_Built cannot equal zero.
Is the coefficient for Beds statistically significant? Practically significant? Explain.
The coefficient for Beds is 10425.93 when fireplaces and Year_Built are zero and
it is statistically significant on a 99% confidence interval. As
the pval of <2e-16 <
0.001
2.
Suppose a new client comes to you wanting to sell their 2005-built, 4
bedroom, 1 fireplace home. What is your best estimate for the sales price of
this home? Calculate and interpret the appropriate 95% confidence or
prediction interval.
The fitted value gives you the predictive price given the home specifications
mentioned above. We are 95% confident a home built in 2005, with 4 bedrooms
and 1 fireplace would sell for a predicted price of $259,810.2 with a lower
predicted value of $146,719.4 and an upper prediction price of $372,901. 4.
Now add the number of bathrooms and livable area as predictors to your model above.
4.
Re-interpret the coefficient for the Beds variable in this new model. Explain
any differences you may see, and speculate on the cause of the differences
you saw.
The coefficient of the Beds variable for the original regression model was
$10425.93 and the coefficient for the new model was -$26845.176. I believe the
reason the coefficient for this model turned out to be negative was because of
the addition of area and baths variables as predictors.
5.
What percent of the variation in sale price is explained by the model? If we
added another predictor, say, number of floors (1,2,3, etc.), would this
increase, decrease, or not change this percentage? Note: this variable is not
in the data set; you are being asked to explain what would happen.
59% of the variation in sale price is explained by the model, according to our
R-squared.
If we added another predictor variable, like number of floors, this
would increase the percentage of variation as adding variables now leaves you
with a model containing too many predictor variables, increasing sum of squares,
and in turn r squared.
6.
Now that you've finished your data science work, it's time to put on your
"consultant" hat. The realtors are eager to learn how to accurately price
homes so that they know if a home they are buying or selling is priced
appropriately. How would you explain your findings to the team at Keller
Williams? The realtors haven't taken a class on regression, so you should
avoid using technical terms (e.g. slope, confidence, etc.).
You want to be able to compare homes of the same caliber to each other while
also keeping in mind th
Your preview ends here
Eager to read complete document? Join bartleby learn and gain access to the full version
- Access to all documents
- Unlimited textbook solutions
- 24/7 expert homework help