Scribd
Upload
53 views

Assignment 2 Guide

This document provides instructions for analyzing the relationships between house price (dependent variable) and various predictor variables like beds, baths, square feet, ZIP code, and location from real estate data. It describes using statistical tools in JMP to explore these relationships visually with scatter plots and mean comparisons, and preparing the predictors for modeling by binning categorical variables. Key steps include fitting linear and quadratic lines to price vs. each continuous predictor, filtering for property types, binning ZIP and location by price ranges, and summarizing the findings in tables and figures in the appendix with conclusions about appropriate modeling approaches.

Uploaded by

fbfdb
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
53 views

Assignment 2 Guide

This document provides instructions for analyzing the relationships between house price (dependent variable) and various predictor variables like beds, baths, square feet, ZIP code, and location from real estate data. It describes using statistical tools in JMP to explore these relationships visually with scatter plots and mean comparisons, and preparing the predictors for modeling by binning categorical variables. Key steps include fitting linear and quadratic lines to price vs. each continuous predictor, filtering for property types, binning ZIP and location by price ranges, and summarizing the findings in tables and figures in the appendix with conclusions about appropriate modeling approaches.

Uploaded by

fbfdb
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Your Name Title BADM 7020-Assignment 2

Data Understanding: The objective of this step is to PRICE to the Y axis and ZIP to the X axis. Select the vertical
understand the relationships between dependent bar chart.
variable (Price) and the predictors and to prepare the
This part should explains how you binned ZIP and
predictors. Specifically, look at the predictors BEDS,
LOCATION and what has been achieved by that.
BATH, SQUARE FEET, ZIP and LOCATION and their
relationship to PRICE. Tabulate Mean PRICE by PROPERTY TYPE. If you decide to
bin this variable you can use recode in JMP because it has
Use Analyze>Fit Y by X in JMP to obtain plots of PRICE
versus individual predictors BEDS, BATH, SQUARE FEET. only few levels. What properties should be excluded and
Under the red triangle on the JMP output select Fit Line. why (see assignment 1)?
Additionally Fit Polynomial Quadratic to all three graphs. Conclusion:
Add the three graphs to the appendix. Describe what you
see. Which line (linear or quadratic) seems to be a better The conclusion should be a recap of what you discovered.
fit? State which variables were binned and what type of
model should be used for the continuous predictors, i.e,
BEDS: What is the best functional relationship. Use Local linear or quadratic.
Filter to select only Townhomes, Single Residential, and
Condos. What is the relationship now?
BATHS: What is the best functional relationship. Use Local
Filter to select only Townhomes, Single Residential, and
Condos. What is the relationship now?
SF: What is the best relationship between PRICE and SF.
Use Local Filter to select only Townhomes, Single
Residential, and Condos.
Use Tabulate to display the mean PRICE by ZIP. Do the
same for LOCATION. Describe what you see. How does
mean PRICE change by ZIP and LOCATION?
Data Preparation: To reduce the number of levels for ZIP
and LOCATION perform binning as described in the
module. Create bins with average prices <$100k, ($100k-
$199k), (%200k-$299k), (%300k-$399k), >=$400k. Use
Tabulate to create a table of mean price by ZIP. The steps
for binning ZIP and LOCATION explained in Binning with
JMP in the Moodle Book are:
1. Tabulate Mean PRICE by ZIP
2. Save result as new table
3. Sort table by mean PRICE
4. Create new column for bins
5. Type in manually the bin ranges
6. Use Tables>JMP Query Builder to merge this table
with your original data
Do step 1-6 for mean PRICE by LOCATION.
IMPORTANT: Note, that you do NOT attempt to bin ZIP
codes according to ZIP codes! In the past some students
attempted to bin ZIP codes according to their numbers
like 70800-70810, 70811-70820,….. This type of binning is
not useful in modeling a relationship between PRICE and
ZIP codes. ZIP codes and LOCATION is binned using
average PRICE not ZIP codes to preserve the relationship
between ZIP codes and PRICE. Therefore your labels for
the bins must indicate that. To create the graphs move
Your Name Title BADM 7020-Assignment 2
Table 1: Summary of variables
(Fill out this table)
Figure 3: PRICE BY SQUARE FEET
VARIABLE Binned Relationship Data Modeling
(Yes/No) to Price Type
(Linear/non- (CONTINUOUS
linear, or ORDINAL
NA) NOMINAL)
BEDS
BATH
SQUARE
FEET
PROPERTY
TYPE
ZIP
LOCATION

Figure 4: BAR CHART OF MEAN PRICE BY BINNED ZIP

Figure 1: PRICE BY BEDS

Figure 2: PRICE BY BATH


Figure 4: BAR CHART OF MEAN PRICE BY BINNED LOCATION

You might also like