# RECORD baaee624057dd67c85185a4177e6d00e5b13c4a2
# TITLE: The Hedonic Pricing Model Applied to the Housing Market
# YEAR: 2020.0  VENUE: International journal of economics and business administration  DOI: 10.35808/ijeba/526
# AUTHORS: My-Linh Thi Nguyen
# PAGES: 13  CANDIDATE TABLE PAGES (1-based): [7, 8]
# ABSTRACT: Purpose: The article applies the hedonic pricing model to estimate house price in the housing market of Vietnam, which is a country with a fledgeling housing market, so the study is expected to bring interesting findings. Design/Methodology/Approach: By applying the hedonic pricing model, most of the previous studies have reported that house price is significantly influenced by the characteristics of the house itself, its location and surrounding facilities. Based on this, the article adopts the Ordinary Least Squares (OLS) regression in combination with robustness statistics in the model estimation, so the estimated results on house price are reliable and able to be widely applied. Data are collected through the survey into housing projects in Ho Chi Minh and Ha Noi city, which are the two largest cities in Vietnam. Findings: The findings reveal that the hedonic pricing model can be applied to estimate house price in Vietnam’s housing market. This can be said to be a big success in giving first empirical evidence in Vietnam on this matter. Specifically, house price is negatively affected by its proximity to the city center. Also, factors including house size, house type, house structure, number of bedrooms, amenities around the house exert a positive influence on the price. Practical Implications: The estimated results are typical and reliable, which can be applied universally. Originality/Value: The study confirms the hedonic pricing model can be well-applied to accurately estimate the price of houses in Vietnam. More than that, the findings are also essential for other countries, especially those with a nascent housing market like Vietnam.


===== PAGE 1 (score 8, numeric-paren 0, table-words 0, var-words 8) =====

 
 
International Journal of Economics and Business Administration 
Volume VIII, Issue 3, 2020 
  
 pp. 416-428 
 
The Hedonic Pricing Model Applied to the Housing Market   
Submitted 11/04/20, 1st revision 19/05/20, 2nd revision 25/06/20, accepted 30/07/20  
  
  My-Linh Thi Nguyen1 
Abstract:  
 
Purpose: The article applies the hedonic pricing model to estimate house price in the 
housing market of Vietnam, which is a country with a fledgeling housing market, so the study 
is expected to bring interesting findings.  
Design/Methodology/Approach: By applying the hedonic pricing model, most of the 
previous studies have reported that house price is significantly influenced by the 
characteristics of the house itself, its location and surrounding facilities. Based on this, the 
article adopts the Ordinary Least Squares (OLS) regression in combination with robustness 
statistics in the model estimation, so the estimated results on house price are reliable and 
able to be widely applied. Data are collected through the survey into housing projects in Ho 
Chi Minh and Ha Noi city, which are the two largest cities in Vietnam.  
Findings: The findings reveal that the hedonic pricing model can be applied to estimate 
house price in Vietnam’s housing market. This can be said to be a big success in giving first 
empirical evidence in Vietnam on this matter. Specifically, house price is negatively affected 
by its proximity to the city center. Also, factors including house size, house type, house 
structure, number of bedrooms, amenities around the house exert a positive influence on the 
price.  
Practical Implications: The estimated results are typical and reliable, which can be applied 
universally.  
Originality/Value: The study confirms the hedonic pricing model can be well-applied to 
accurately estimate the price of houses in Vietnam. More than that, the findings are also 
essential for other countries, especially those with a nascent housing market like Vietnam. 
 
Keywords: Hedonic pricing model, house price, housing market, real estate, Vietnam. 
 
JEL: C53, E37, R32. 
 
Paper Type: Research article. 
 
 
 
 
 
  
 
1Assoc. Prof. Faculty of Finance and Banking, University of Finance – Marketing (UFM), 
Ho Chi Minh city, Vietnam. E-mail: ntmylinh@ufm.edu.vn.  
ORCID ID: 0000-0001-7475-2502   


===== PAGE 7 (score 32, numeric-paren 1, table-words 5, var-words 16) =====

        The Hedonic Pricing Model Applied to the Housing Market 
  
 422  
 
 
PRICE = β0 + β1 SIZE + β2 BEDROOMS + β3 BATHROOMS + β4 CENTER + β5 
TYPE + β6 POOL + β7 PARK + β8 STRUCTURE + ε 
 
Where: Dependent variable (PRICE): House price (price per square meter). 
 
 
Independent variables: SIZE: House size (square meters); BEDROOMS: Number 
of bedrooms; BATHROOMS: Number of bathrooms; CENTER: Proximity to the 
city center (kilometers); TYPE: A binary variable indicating whether the house has a 
garden (= 1 if yes, = 0 otherwise); POOL: A binary variable indicating whether the 
house has its amenities such as a hot-tub and/or a swimming pool (= 1 if yes, = 0 
otherwise); PARK: A binary variable indicating whether the house has local 
amenities such as parks, squares, schools or shopping malls (= 1 if yes, = 0 
otherwise); STRUCTURE: A binary variable indicating whether the structure is new 
and modern (= 1 if yes, = 0 otherwise); ε: Error term; βi: Coefficients; β0: Constant. 
 
4. Empirical Results  
 
First, the article analyses the correlation among the variables. The result is shown in 
Table 2 as follows: 
 
Table 2. Variable correlations 
 
PRICE 
SIZE 
BEDR
OOMS 
BATH
ROOM
S 
CENTE
R 
TYPE 
POOL 
PARK 
STRUC
TURE 
PRICE 
1.000 
 
 
 
 
 
 
 
 
SIZE 
0.128 
1.000 
 
 
 
 
 
 
 
BEDROOMS 
0.776 
0.146 
1.000 
 
 
 
 
 
 
BATHROOMS 
0.546 
0.167 
0.656 
1.000 
 
 
 
 
 
CENTER 
-0.590 
0.010 
-0.473 
-0.369 
1.000 
 
 
 
 
TYPE 
0.595 
-0.046 
0.429 
0.315 
-0.300 
1.000 
 
 
 
POOL 
0.011 
0.017 
0.025 
0.005 
-0.037 
-0.002 
1.000 
 
 
PARK 
0.776 
0.082 
0.471 
0.334 
-0.413 
0.615 
-0.044 
1.000 
 
STRUCTURE 
0.712 
-0.092 
0.536 
0.370 
-0.422 
0.435 
0.017 
0.661 
1.000 
Source: Own study. 
 
Table 2 reveals that the distance from the house to the center (CENTER) is 
negatively correlated to its price (PRICE) while the other independent variables are 
positively associated with the house price (PRICE). This is in line with what the 
article has been suggested previously.  
 
Table 3. Results of tests on multicollinearity, heteroscedasticity 
Multicollinearity test 
Heteroscedasticity test 
Variable 
VIF 
PARK 
2.52 
Chi2(42) = 89.01 
Prob>chi2 = 0.000*** 
BEDROOMS 
2.36 
STRUCTURE 
2.18 
BATHROOMS 
1.79 


===== PAGE 8 (score 28, numeric-paren 8, table-words 5, var-words 7) =====

    My-Linh Thi Nguyen 
  
423  
TYPE 
1.73 
CENTER 
1.41 
SIZE 
1.14 
POOL 
1.01 
Mean VIF = 1.77 
Note: *** indicates significance at the 1% level. 
Source: Own study. 
 
The model is considered to experience no serious issues of multicollinearity (Mean 
VIF = 1.77). However, heteroscedasticity really exists at the 1% level of significance 
(Prob>chi2 = 0.000) (Table 3). This means that the estimation using the Ordinary 
Least Squares regression (OLS) does not provide a reliable result. Thus, robustness 
statistics is utilized for the estimation, following what have been suggested by White 
(1980) and recently performed by Bui (2020a).  
 
Table 4. Estimation results 
Dependent variable: House price (PRICE) 
Variables 
Coef. 
P>|z| 
Constant 
17.093*** 
0.000 
SIZE 
0.001** 
0.023 
BEDROOMS 
0.065*** 
0.000 
BATHROOMS 
0.003 
0.640 
CENTER 
-0.020*** 
0.000 
TYPE 
0.048** 
0.036 
POOL 
0.002 
0.709 
PARK 
0.187*** 
0.000 
STRUCTURE 
0.050*** 
0.000 
Number of obs 
205 
R-squared 
85.59% 
Significance level 
F(8, 196) = 276.08 
Prob>F = 0.000*** 
Note: ** and *** indicate significance at the 5% and 1% level, respectively. 
Source: own study. 
 
As can be seen from Table 4, the estimation results are significant at the 1% level 
(Prob>F = 0.000). Specifically, the independent variables can account for 85.59% 
variation in house price. It can be concluded that the model is appropriate and valid 
for the practical application. Alternatively, the hedonic pricing model is successfully 
applied in estimating house price in Vietnam. Furthermore, with a high explanation 
level of research model and highly-representative dataset, the results can be widely 
administered.   
 
 
Particularly, CENTER exerts the negative impact (β = -0.020) on PRICE at 1% level 
of significance whereas PRICE is positively influenced by SIZE (β = 0.001, at 5% 
level of significance), BEDROOMS (β = 0.065, at 1% level of significance), TYPE 
(β = 0.048, at 5% level of significance), PARK (β = 0.187, a 1% level of 
significance), STRUCTURE (β = 0.050, at 1% level of significance). Nevertheless, 


===== PAGE 9 (score 13, numeric-paren 0, table-words 1, var-words 11) =====

        The Hedonic Pricing Model Applied to the Housing Market 
  
 424  
 
 
the study has not found the significant influence of BATHROOMS and POOL on 
PRICE. Hence, the estimated results of the model take the following equation:        
 
PRICE = 17.093 + 0.001 SIZE + 0.065 BEDROOMS - 0.020 CENTER + 0.048 
TYPE + 0.187 PARK + 0.050 STRUCTURE + ε 
 
Table 5. Results of hypothesis testing 
No 
Hypothesis 
Results 
H1 
SIZE ⟹ PRICE 
Accepted 
H2 
BEDROOMS ⟹ PRICE 
Accepted 
H3 
BATHROOMS ⟹ PRICE 
Rejected 
H4 
CENTER ⟹ PRICE 
Accepted 
H5 
TYPE ⟹ PRICE 
Accepted 
H6 
POOL ⟹ PRICE 
Rejected 
H7 
PARK ⟹ PRICE 
Accepted 
H8 
STRUCTURE ⟹ PRICE 
Accepted 
Source: Own study. 
 
- The impact of house size (SIZE) on house price (PRICE): The results show that 
SIZE positively affects PRICE. This indicates that houses in bigger size are usually 
priced more. This finding corroborates the reality and what has been reported by 
Kain and Quigley (1970), Selim (2009), Amenyah and Fletcher (2013), Shimizu 
(2014), Yayar and Demir (2014), Ndegwa (2018). Accordingly, big-size houses can 
meet demand of crowded families, bringing a wide and comfortable space. These 
houses can also specially be used for their business or offices.  
 
- The impact of the number of bedrooms (BEDROOMS) on house price (PRICE): 
The results show that BEDROOMS positively affects PRICE. This is consistent with 
the findings of Kain and Quigley (1970), Cebula (2009), Selim (2009), Baek et al. 
(2019). Accordingly, houses with more bedrooms suit families with many members. 
These houses allow their guests to stay over. As a result, they are more expensive. 
 
 
- The impact of house location (CENTER) on house price (PRICE): The study 
supports the negative influence of CENTER on PRICE. This shows that houses far 
from the center may sell for less than those with a close proximity to the center. This 
is because people living in the former find it difficult to commute to work or town 
center. Also,  their public amenities are more limited than the latter. This is in line 
with the results in the studies of Selim (2009), Aluko (2011), Amenyah and Fletcher 
(2013), Yayar and Demir (2014), Ndegwa (2018). 
 
 
- The impact of house type (TYPE) on house price (PRICE): The findings reveal 
that PRICE is positively affected by TYPE. In the metropolises with a high 
population, the presence of garden in a house can help improve people’s living 
standard. Consequently, this kind of house is higher priced. This finding is consistent 
with what has been stated by Limsombunchai et al. (2004).