I analyze the determinants of property values in Las Vegas, by using a multitude of independent variables. I utilized Stata for my analysis, using several forms of summary and descriptive statistics.
2. Introduction
The Las Vegas housing market is as unpredictable as the slot machines within the city
itself. In In the early 2000’s, Las Vegas faced a major housing bubble due to multiple factors -
demand, speculation, and exuberant spending to name a few. What once was a little big town
with casinos, truly became a world class city, one that would rival nearby metropolises such as
Los Angeles, Seattle, and San Francisco. In 2021 and beyond, Vegas has a new set of parameters
that determine the overall housing market. I put together a table to describe these relationships
between prices and the explanatory variables within a given data set.
First, I ran the data given consisting of said variables (sqft, bed, bath, pool) through Stata
to explain the rationale for the expected signs (+/-/n/a). Once provided, I then constructed a
second table to describe the summary statistics (This would include common terms such as
mean, standard deviation, and min. and max. values for all variables). From there, I ran a matrix
correlation that demonstrates the predictor variables correlate with the target variable
(determinants of the Las Vegas housing market.) I put together and displayed scatter plots that
show the correlation between each of the predictor variables and determinants (dependent
variables) that will be analyzed. Finally, I constructed two separate regression tables displaying
the full and reduced model with average determinants. With this information, we ultimately
determine the slope p-value approach to test the statistical significance of each of the slope
coefficients, interpretations, and test for the overall significance of the model.
3. Model
This model that is used is a function of all predictor variables, since all of them are expected to
influence the determinants of the Las Vegas housing market.
Determinants = f (sqft, bed, bath, pool)
Table 1: Variables, definitions and expected signs
Variables Definitions Expected Sign
Price Sale price of home +
SQFT Square feet, measuring the size of living space +
BED Number of bedrooms +
BATH Number of bathrooms +
POOL 1 if the house has a pool or spa and 0 otherwise +
Table 1 displays a predictor of all variables. In this case, I assumed that every variable
would have a positive impact on the buyer. This may be based off the prospective buyer’s
attitude towards something like a pool, which would include more maintenance, but also allows
them to sell the property after a few years at a higher price to family who may want a pool. Other
variables such as bed, bath and sq ft may also be positives, as many people would prefer more
for less. For example, more bathrooms may not be a benefit for the average single person, but it
usually wouldn’t be a negative (-) as it doesn’t hurt the property value. Overall, I decided to
leave every sign positive (+) due to the mentality that most variables shouldn’t hurt the buyer in
the long run, specifically when to comes to property and re-sale values.
4. Table (2) Descriptive results
The average price is equal to $192600.10, and the table shows as descriptive statistics for
all the variables. As we can see, average price of the home in the data set is $192,600.10 with a
standard deviation equal to $90,684.38. The cheapest price is $71,000, and the price of the most
expensive house in the data set is $674,500. Also, we can see that the Summerlin, Green Valley,
Northwest, and East Las Vegas areas all have close to the same percentage of houses of .26, .30,
.24, and .21 for the study. I also analyzed bed, bath, pool and sqft while following the same
model for price. With the other variables in the descriptive results, the data demonstrates the
trends for other variables within the results.
5. Table (3) The matrix of correlation coefficients
The matrix of correlation coefficient reveals that a strong correlation between (+.71) the
size of the house and the price. Also, the number of beds show a weak positive correlation (+.24)
between the number of beds and the price of the house. The number of bathrooms showed a
moderate positive correlation with the price of the house. Summerlin shows the highest positive
correlation between the price and the house, comparing to Green Valley and especially East and
Northwest Las Vegas, which have negative correlations. As stated prior, the data depends on the
year used and what economic market it represents.
6. Figure (1) Scatter plot of PRICE and SQFT
The scatterplot shown between the price and sq ft of house shows a positive correlation where an
increasing the number sq ft in the house will increase the price.
7. Figure (2) Scatter plot of PRICE and BED
The scatter plot between price and the number of beds shows a slightly positive correlation that
is associated with what we found in correlation matrix, which was expected at the beginning of
the project.
8. Figure (3) Scatter plot of PRICE and BATH
This scatter plot also reflects a slight positive correlation – as the number of baths increases,
prices increase as well. Ultimately, this reflects (generally speaking) what I stated in Table 1.
Empirical Results
Table 4 displays the results of a Full and Reduced regression model. We then test the statistical
significance of the slope coefficients at the .1, .05, and .01, respectively.
9. Pricef=12385.19 +85.07sqft -
15884.94bed+10277.4bath+10058.48pool+57413.72summerlin+24723.08greenvalley-
5745.75east
Pricer=32331.87 +88.16sqft -23975.04bed+10277.4bath+9732.33pool
Table 1: Regression results representing the Full and Reduced models. Parentheses indicate p-
values, * p<0.1, ** p<0.05, *** p<0.01
Variables Full Model Reduced Model
Constant 12385.19 32331.87
-0.65 -0.209
sqft 85.07 88.16
(0.000) *** (0.000)***
bed -15884.94 -23975.04
(0.018)** (0.001)***
bath 10277.4 19248.88
-0.481 -0.213
pool 10058.48 9732.33
-0.4 -0.448
Summerlin 57413.72
(0.000)***
Green valley 24723.08
(0.054)*
East -5745.75
-0.681
N 164 164
R2 0.6141 0.5455
F Fc (7,156) = 35.47 Fc (4,159) = 47.71
10. We can see that the sq ft is a positive significant correlation which tells us that increasing
the house with 1 unit of a sq ft will increase the price by $85, on average. This keeps all other
variables constant. Bed is negative and significant at .05 and .10, but in the reduced model, it’s
strongly significant at .01. Bath is positive, but not significant in both full and reduced model.
Pool is positive and not significant in both models. The variable Summer was strongly and
positively significant at .01. Green Valley is significant as well, but not as strong as Summerlin.
Which obviously means that buyers overwhelmingly prefer Summerlin to Green valley in the
given data set. On the other hand, the East side of the valley has a negative coefficient with no
significance, even at .10. In other the words, the prices in the East Las Vegas will not be affected
in this data set.
Conclusion
Even with the lingering Covid-19 pandemic, the Las Vegas real estate market is red hot.
While many businesses are still struggling with how to get back their labor force after abrupt
layoffs, we can see the data shows that people still like the “bells and whistles” when it comes to
purchasing a house. According to the statistical information, Summerlin was and is still an
expensive market for most, and East Las Vegas was significantly less popular. Supply and
demand are two factors that will always play a role in real estate appreciation, and location is one
of the biggest factors in this equation. Other data shows that extra add-ons such as beds, bath,
and pools don’t really affect the overall price of the house in a way that many would expect.
Overall, the study reflects what many in the Las Vegas valley already know: people want to live
in good neighborhoods, with good amenities and ultimately be able to sell their properties at a
reasonable profit.
