Sam Easton

What should Easton Realty do next?

DI VITO Simone
GOWANI Twinkle
HANDA Premik
HAMONOU Quentin

Refresher Statistics – Group 3

Sam Easton has to come up with concrete arguments showing that the prices for which both houses
were sold were actually in line with the market

Problem Key points under Consideration

• Sale month
Easton Realty received a letter from local Board of Realtors • Sale price
The Board received complaints from two people that Easton • Size of the home (area)
Realty under-priced the properties in order to sell them faster • Number of bedrooms
• Age of the house
• Location, and
• The real estate company that sold the house

Home Sale in the DFW Area

Average Sales Price Average Size

Properties in picture
$ 104,250 1860 sq. ft.

Refresher Statistics – Group 3

3,300 160,000

Surface area v/s Sale Price Sale Prices in Different Months

140,000
2,800

120,000

2,300

100,000
f(x) = − 4056.67 x + 111459.44
R² = 0.06
1,800

80,000

1,300
60,000

March April May June

800 40,000
40,000 60,000 80,000 100,000 120,000 140,000 160,000
2 3 4 5 6 7

The sales price is strongly related to the surface (area) of the The parameter value of x provides strong evidence that property
property. prices are significantly dependant on the month in which property is
sold.

Refresher Statistics – Group 3

 Area of the House Age of the House Number of Bedrooms
120

Fort Worth 22%

100

2 Bedroom
80

44%
Fort Worth 60
3 Bedroom

40
4 Bedroom

Elsewhere 20
33%

0
0 50 100 150 200 250 1 2 3 4 5 6 7 8 9 10 11 12 13

Area of House Maximum houses are sold in Dallas, which is out of scope of the argument
Age of House Similarly, the houses in question are aged 4 & 9 which is again not near the median
# Bedrooms The bedroom criteria still aligns a bit with the data at 3 bedrooms each

Refresher Statistics – Group 3

 Agency vs Area of Sale Age of House vs Number of Bedrooms
Agency / Area Dallas Fort Worth Elsewhere Total Age / # Bedrooms 2 3 4 Total
Otherwise 236 139 94 469 1 1 1 2
Easton 7 27 15 49 2 4 11 3 18
Grand Total 243 166 109 518 3 3 16 9 28
4 16 39 16 71
Agency vs Month of Sale 5 12 50 14 76
Agency / Months March April May June Total 6 20 53 26 99
Otherwise 129 120 114 106 469 7 11 56 17 84
Easton 11 11 13 14 49 8 8 42 19 69
Total 140 131 127 120 518 9 6 28 10 44
Area vs Month of Sale 10 - 9 8 17
Area / Months March April May June Total 11 2 3 3 8
Dallas 73 60 52 58 243 12 - 1 - 1
Fort Worth 42 43 41 40 166 13 - - 1 1
Elsewhere 25 28 34 22 109 Grand Total 82 309 127 518

• Max houses are sold in Dallas followed by Fort Worth • The bedroom criteria matches with property in question at 3 bedrooms
• More Sales happened in March-April as opposed to May-June each
• In Fort Worth sales are relatively steady in all months, where Easton • The Median Age of the house is 7 with the mode being at 6 years
sold its maximum properties • Heavy concentration of sample lies in Age 4-8 years

Refresher Statistics – Group 3

 Area Size Bedrooms Price Age
Observations
Dallas 1851.127572 3.024691358 104288.0658 6.024691358
• For Area of Sales (Standard Deviations below)
Fort Worth 1844.891566 3.078313253 84192.16867 6.397590361
55,24051316 0,09013319343 5587,079611 0,1285179929

Everywhere 1950.642202 3.23853211 83411.00917 6.174311927 There is significant variation in the price of properties
sold in Dallas
Grand Total 1870.069498 3.086872587 93455.01931 6.175675676
Size of property sold everywhere exceeds the other
Month Size Bedrooms Price Age areas

March 1883.157143 3.078571429 99091.42857 6.085714286 • For Months of Sales (Standard Deviations below)
27,57612518 0,02765759063 3661,423254 0,1009975051
April 1883.51145 3.106870229 95648.85496 6.297709924
May 1883.874016 3.11023622 90972.44094 6.251968504 Price of Properties sold in June is relatively lower than
in the other months
June 1825.516667 3.05 87111.66667 6.066666667
Similarly, size and age of properties sold in June also
Grand Total 1870.069498 3.086872587 93455.01931 6.175675676 varies from the other months
• For Agency of Sales (Standard Deviations below)
Agency Size Bedrooms Price Age
26,25811585 0,06844856045 4485,018795 0,008775456468
Easton 1866.189765 3.076759062 94117.69723 6.176972281
Size of properties & Number of Bedrooms are smaller
Elsewhere 1907.204082 3.183673469 87112.2449 6.163265306 when sold by Easton

Grand Total 1870.069498 3.086872587 93455.01931 6.175675676 Price of sale is higher for Easton sold properties

Refresher Statistics – Group 3

 Price Statistics Size Stats Bedrooms stats Age stats

Mean 93455.01931 Mean 1870.0695 Mean 3.08687259 Mean 6.17567568

Standard Error 789.1776975 Standard Error 14.9024657 Standard Error 0.02767342 Standard Error 0.09159941
Median 92100 Median 1821.5 Median 3 Median 6
Mode 88500 Mode 1478 Mode 3 Mode 6
Standard Deviation 17961.37926 Standard Deviation 339.174357 Standard Deviation 0.62983639 Standard Deviation 2.08476722

Sample Variance 322611145 Sample Variance 115039.245 Sample Variance 0.39669388 Sample Variance 4.34625438
Kurtosis -0.401114776 Kurtosis -0.0705961 Kurtosis -0.4964793 Kurtosis -0.2843845
Skewness 0.316780227 Skewness 0.33012626 Skewness -0.0683109 Skewness 0.10328162
Range 97300 Range 1805 Range 2 Range 12
Minimum 51800 Minimum 1067 Minimum 2 Minimum 1
Maximum 149100 Maximum 2872 Maximum 4 Maximum 13
Sum 48409700 Sum 968696 Sum 1599 Sum 3199
Count 518 Count 518 Count 518 Count 518
Confidence Level(95,0%) 1550.38938 Confidence Level(95,0%) 29.2768341 Confidence Level(95,0%) 0.05436618 Confidence Level(95,0%) 0.17995283

Confidence interval
Lower Higher Lower Higher Lower Higher Lower Higher
91,905 95,005 1,841 1,899 3 3 6 6

Confidence intervals inform about likely population average For Price Statistics, the properties in question fall outside the 95% confidence level displaying
Provide both the location and precision of a measure distortion and hence would require further analysis
The greater the confidence level, the wider the confidence interval Like for Age of the House, properties are way away from the confidence level range
If we assume the confidence level is fixed, the only way to obtain more precise Similarly, one of the properties size lie way outside the confidence interval denoting an outlier
population estimates is to minimize sampling error For Bedrooms, the mean=median=mode and is well in line with the properties challenged

Refresher Statistics – Group 3

The margin of error is a statistic expressing the A margin of error tells you how many percentage
amount of random sampling error in points your results will differ from the real
a survey's results. population value.
The larger the margin of error, the less For example, a 95% confidence interval with a 4
confidence one should have that the poll's percent margin of error means that your statistic
reported results are close to the true figures; will be within 4 percentage points of the real
that is, the figures for the whole population. The
The sample
sample population value 95% of the time
size
size should
should be
be
221
221

If we assume that the confidence interval is too large, we estimate a Margin of Error of (+/- 5%). The
sample size, considering the information previously used, can be computed with the following formula:
 

Refresher Statistics – Group 3

 As per the average price according to
As the number of bedrooms increase,
As the time (year) passes, the the area, we can say that area 1 is
the price of the house increase
prices of the houses also falls priced higher and area 3 is on the
Calculated is the average price for the
down lower end
# bedrooms in a house and the results
Thus, age of the house has an Area 1 might be in the heart of the
denote that price increases as number
impact on the prices town with area 3 being on the
of bedrooms increase in a house
outskirts

Refresher Statistics – Group 3

 Sale Price vs Area in Different locations Observations
150000

• Properties in Dallas have

a higher sale price
130000
• Properties outside Dallas
and Fort Worth have the
lowest sale price

110000
• The sales price is
strongly related to the
surface (area) of the
property
90000 • The sales price is closely
related to the location of
the property

70000
• Both properties appear
to have been sold with a
price lower than the
average market price in
50000
their respective location
1000 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000

Refresher Statistics – Group 3

 Price per square foot
To facilitate comparison we now compare the price per 60
square foot instead of total price.
Summary range
March April May June 55
Minimum 38.53881 36.06045 37.79104 37.37374
25th percentile 46.45174 45.14296 43.07908 42.15816

PPSF (in thousands)

Median 53.32892 51.05597 47.13596 47.36284 50
75th percentile 58.32165 56.83761 54.15611 53.37435
Maximum 71.00271 69.41679 68.27586 66.43258
45

From the given chart we interpret that-

- Price per square foot vary depending on the month 40

- PPSF is lowest in the month of June. PPSF has gone
down by approximately 10% in the last 4 months
35
March April May June

Refresher Statistics – Group 3

 Number of bedrooms Area Number of properties sold

Other locations 120

2 bedrooms 140
4 bedrooms 18%
23%

Fortwort
127
131

Dallas

March April May June

0 10 20 30 40 50 60 70

Total of 120
3 bedroom
houses were
Maximum properties had
sold in the
number of the highest
month of June.
properties were share of total
That is lowest
sold in Dallas. sales in the
amongst the last
month of June.
4 months.

Refresher Statistics – Group 3

 140,000 Plot size vs Sales Price
120,000 3,500

100,000 3,000

80,000 2,500
Sales Price

f(x) = 0.01 x + 627.13

60,000 price 2,000
Linear (price)
40,000 1,500

20,000 1,000

0 500
1 1 2 2 3 3 4
0
Area 40,000 60,000 80,000 100,000 120,000 140,000 160,000

 Dallas is the most expensive area followed by Fortwort and then other locations
Observations  Sales price has a positive relationship with the plot size

Refresher Statistics – Group 3

   Summary range
Other
  Dallas Fortworth locations
Minimum 47.3537604 38.58757 37.37374
25th 51.9721473 40.92862 38.93279
percentile
Median 53.5853573 43.22212 40.3689
75th 56.4694325 44.97847 42.22134
percentile

Maximum 66.4325843 49.57983 47.51339

From the given chart we interpret that-

- Price per square foot has a very strong

relationship with the area in which the property is
located
- PPSF is significantly high in Dallas, followed by
Fortworth and then by other locations

Refresher Statistics – Group 3

 SIGNIFICANT Coefficients Standard Error t Stat P-value Lower 95% Upper 95%
VARIABLES
Intercept 56.839744 0.940107 60.460905 3.336E-234 54.992767 58.686721
Size -0.006560 0.000505 -12.989428 1.631E-33 -0.007552 -0.005567
Bedrooms 0.480867 0.272131 1.767045 7.782E-02 -0.053773 1.015507
Age -0.061319 0.063018 -0.973034 3.310E-01 -0.185128 0.062490
April -1.069916 0.361112 -2.962835 3.191E-03 -1.779373 -0.360459
May -2.926743 0.365134 -8.015542 7.585E-15 -3.644101 -2.209386
June -4.946914 0.370197 -13.362911 4.080E-35 -5.674220 -4.219608
Dallas 13.094238 0.349863 37.426723 3.929E-148 12.406881 13.781595
Fortwort 2.280467 0.369149 6.177630 1.335E-09 1.555220 3.005714
Easton -0.239373 0.456625 -0.524222 6.004E-01 -1.136480 0.657734

SUMMARY OUTPUT
Regression Statistics
ANOVA
Multiple R 0.915186167
R Square 0.83756572   df SS MS F Significance F
Adjusted R Square 0.834687947 Regression 9 23031.70603 2559.078448 291.0465186 4.06E-194
Residual 508 4466.680646 8.792678436
Standard Error 2.965245089
Observations 518 Total 517 27498.38668      

Refresher Statistics – Group 3

PPSF = 56.84 - 0.00656Size + 0.4808Bedrooms - 0.0613Age – 1.07April – 2.927May – 4.9469June + 13.0942Dallas +
2.28Fortworth – 0.2394Easton

For House 1 (Arlington)

• PPSF = 56.84 – (0.00656*2190) + 0.4808*3 – (0.0613*4) – (1.07*0) – (2.927*0) – (4.9469*1) +
13.0942*0 + 2.28*0 – (0.2394*1) = 38.48
For House 2 (Fortworth)
• PPSF = 56.84 – (0.00656*1848) + 0.4808*3 – (0.0613*9) – (1.07*0) – (2.927*0) – (4.9469*1) +
13.0942*0 + 2.28*1 – (0.2394*1) = 42.70

Refresher Statistics – Group 3

 Estimated price per square foot Price per square foot at which both the
properties were sold
House 1 (Arlington) 38.48 88500/2190 = 40.41
House 2 (Fortworth) 42.70 79500/1848 = 43.02

Conclusion
As demonstrated from the above table we can conclude that Easton Realty did not
deliberately under price the properties.

Refresher Statistics – Group 3

 Size   Bedrooms   Age   April   May   June   Dallas   Fortworth  

Mean 1866.19 Mean 3.076759 Mean 6.176972 Mean 0.255864 Mean 0.24307 Mean 0.226013 Mean 0.503198 Mean 0.296375
Standard
15.79494 Standard Error 0.029363 Standard Error 0.096697 Standard Error 0.02017 Standard Error 0.019828 Standard Error 0.019333 Standard Error 0.023112 Standard Error 0.021109
Error
Median 1814 Median 3 Median 6 Median 0 Median 0 Median 0 Median 1 Median 0
Mode 1490 Mode 3 Mode 6 Mode 0 Mode 0 Mode 0 Mode 1 Mode 0
Standard 342.0616 Standard 0.635887 Standard 2.094106 Standard 0.436811 Standard 0.429395 Standard 0.418694 Standard 0.500524 Standard 0.457146
Deviation Deviation Deviation Deviation Deviation Deviation Deviation Deviation
Sample 117006.2 Sample 0.404352 Sample 4.385281 Sample 0.190804 Sample 0.18438 Sample 0.175305 Sample 0.250524 Sample 0.208983
Variance Variance Variance Variance Variance Variance Variance Variance
Kurtosis -0.14228 Kurtosis -0.53486 Kurtosis -0.29 Kurtosis -0.74296 Kurtosis -0.558 Kurtosis -0.2736 Kurtosis -2.00842 Kurtosis -1.20473
Skewness 0.328813 Skewness -0.06501 Skewness 0.083842 Skewness 1.122599 Skewness 1.20183 Skewness 1.314375 Skewness -0.01283 Skewness 0.894667
Range 1805 Range 2 Range 12 Range 1 Range 1 Range 1 Range 1 Range 1
Minimum 1067 Minimum 2 Minimum 1 Minimum 0 Minimum 0 Minimum 0 Minimum 0 Minimum 0
Maximum 2872 Maximum 4 Maximum 13 Maximum 1 Maximum 1 Maximum 1 Maximum 1 Maximum 1
Sum 875243 Sum 1443 Sum 2897 Sum 120 Sum 114 Sum 106 Sum 236 Sum 139
Count 469 Count 469 Count 469 Count 469 Count 469 Count 469 Count 469 Count 469
Confidence 31.03778 Confidence 0.057699 Confidence 0.190014 Confidence 0.039635 Confidence 0.038962 Confidence 0.037991 Confidence 0.045416 Confidence
0.04148
Level(95.0%) Level(95.0%) Level(95.0%) Level(95.0%) Level(95.0%) Level(95.0%) Level(95.0%) Level(95.0%)

Regression Statistics ANOVA

Multiple R 0.912506
  df SS MS F Significance F
R Square 0.832667
Regression 8 20842.32 2605.29 286.1258 3.2E-173
Adjusted R Square 0.829757
Standard Error 3.017516 Residual 460 4188.484 9.1054
Observations 469 Total 468 25030.81      

Refresher Statistics – Group 3

   Coefficients Standard t Stat P-value Lower 95% Upper 95%
Error Residuals
Intercept 56.92489 1.006381 56.56394 2.8E-209 54.94722 58.90257 15

size -0.00691 0.000536 -12.8837 1.2E-32 -0.00796 -0.00585 10

bedrooms 0.61868 0.287975 2.148381 0.032205 0.05277 1.18459
5
age -0.01167 0.06703 -0.17407 0.861885 -0.14339 0.120054
0
april -1.25307 0.383788 -3.26501 0.001176 -2.00727 -0.49888 30 35 40 45 50 55 60 65
-5
may -3.02068 0.39089 -7.72769 6.94E-14 -3.78883 -2.25252

june -5.11377 0.396964 -12.8822 1.22E-32 -5.89386 -4.33368 -10

dallas 13.02129 0.374476 34.77202 8E-131 12.28539 13.75718 -15

fortwort 2.192502 0.407571 5.379438 1.19E-07 1.39157 2.993433 Residuals

Only age is nonsignificant for us because it does not lie between the confidence interval (which is calculated as mean +- confidence level) 
R and R square measure goodness of fit of our model at 0.83 denoting that R explains 83% of PPSF
Arlington Estimated price: 38.253 and Fortworth estimated price: 42.837
In this scenario we are modelling for the rest of the market excluding Easton. The results show that the two houses are under-priced and this
proves our point that Easton is above the market rate

Refresher Statistics – Group 3