1# RECORD baaee624057dd67c85185a4177e6d00e5b13c4a22# TITLE: The Hedonic Pricing Model Applied to the Housing Market3# YEAR: 2020.0 VENUE: International journal of economics and business administration DOI: 10.35808/ijeba/5264# AUTHORS: My-Linh Thi Nguyen5# PAGES: 13 CANDIDATE TABLE PAGES (1-based): [7, 8]6# 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.789===== PAGE 1 (score 8, numeric-paren 0, table-words 0, var-words 8) =====1011 12 13International Journal of Economics and Business Administration 14Volume VIII, Issue 3, 2020 15 16 pp. 416-428 17 18The Hedonic Pricing Model Applied to the Housing Market 19Submitted 11/04/20, 1st revision 19/05/20, 2nd revision 25/06/20, accepted 30/07/20 20 21 My-Linh Thi Nguyen1 22Abstract: 23 24Purpose: The article applies the hedonic pricing model to estimate house price in the 25housing market of Vietnam, which is a country with a fledgeling housing market, so the study 26is expected to bring interesting findings. 27Design/Methodology/Approach: By applying the hedonic pricing model, most of the 28previous studies have reported that house price is significantly influenced by the 29characteristics of the house itself, its location and surrounding facilities. Based on this, the 30article adopts the Ordinary Least Squares (OLS) regression in combination with robustness 31statistics in the model estimation, so the estimated results on house price are reliable and 32able to be widely applied. Data are collected through the survey into housing projects in Ho 33Chi Minh and Ha Noi city, which are the two largest cities in Vietnam. 34Findings: The findings reveal that the hedonic pricing model can be applied to estimate 35house price in Vietnam’s housing market. This can be said to be a big success in giving first 36empirical evidence in Vietnam on this matter. Specifically, house price is negatively affected 37by its proximity to the city center. Also, factors including house size, house type, house 38structure, number of bedrooms, amenities around the house exert a positive influence on the 39price. 40Practical Implications: The estimated results are typical and reliable, which can be applied 41universally. 42Originality/Value: The study confirms the hedonic pricing model can be well-applied to 43accurately estimate the price of houses in Vietnam. More than that, the findings are also 44essential for other countries, especially those with a nascent housing market like Vietnam. 45 46Keywords: Hedonic pricing model, house price, housing market, real estate, Vietnam. 47 48JEL: C53, E37, R32. 49 50Paper Type: Research article. 51 52 53 54 55 56 57 581Assoc. Prof. Faculty of Finance and Banking, University of Finance – Marketing (UFM), 59Ho Chi Minh city, Vietnam. E-mail: ntmylinh@ufm.edu.vn. 60ORCID ID: 0000-0001-7475-2502 616263===== PAGE 7 (score 32, numeric-paren 1, table-words 5, var-words 16) =====6465 The Hedonic Pricing Model Applied to the Housing Market 66 67 422 68 69 70PRICE = β0 + β1 SIZE + β2 BEDROOMS + β3 BATHROOMS + β4 CENTER + β5 71TYPE + β6 POOL + β7 PARK + β8 STRUCTURE + ε 72 73Where: Dependent variable (PRICE): House price (price per square meter). 74 75 76Independent variables: SIZE: House size (square meters); BEDROOMS: Number 77of bedrooms; BATHROOMS: Number of bathrooms; CENTER: Proximity to the 78city center (kilometers); TYPE: A binary variable indicating whether the house has a 79garden (= 1 if yes, = 0 otherwise); POOL: A binary variable indicating whether the 80house has its amenities such as a hot-tub and/or a swimming pool (= 1 if yes, = 0 81otherwise); PARK: A binary variable indicating whether the house has local 82amenities such as parks, squares, schools or shopping malls (= 1 if yes, = 0 83otherwise); STRUCTURE: A binary variable indicating whether the structure is new 84and modern (= 1 if yes, = 0 otherwise); ε: Error term; βi: Coefficients; β0: Constant. 85 864. Empirical Results 87 88First, the article analyses the correlation among the variables. The result is shown in 89Table 2 as follows: 90 91Table 2. Variable correlations 92 93PRICE 94SIZE 95BEDR96OOMS 97BATH98ROOM99S 100CENTE101R 102TYPE 103POOL 104PARK 105STRUC106TURE 107PRICE 1081.000 109 110 111 112 113 114 115 116 117SIZE 1180.128 1191.000 120 121 122 123 124 125 126 127BEDROOMS 1280.776 1290.146 1301.000 131 132 133 134 135 136 137BATHROOMS 1380.546 1390.167 1400.656 1411.000 142 143 144 145 146 147CENTER 148-0.590 1490.010 150-0.473 151-0.369 1521.000 153 154 155 156 157TYPE 1580.595 159-0.046 1600.429 1610.315 162-0.300 1631.000 164 165 166 167POOL 1680.011 1690.017 1700.025 1710.005 172-0.037 173-0.002 1741.000 175 176 177PARK 1780.776 1790.082 1800.471 1810.334 182-0.413 1830.615 184-0.044 1851.000 186 187STRUCTURE 1880.712 189-0.092 1900.536 1910.370 192-0.422 1930.435 1940.017 1950.661 1961.000 197Source: Own study. 198 199Table 2 reveals that the distance from the house to the center (CENTER) is 200negatively correlated to its price (PRICE) while the other independent variables are 201positively associated with the house price (PRICE). This is in line with what the 202article has been suggested previously. 203 204Table 3. Results of tests on multicollinearity, heteroscedasticity 205Multicollinearity test 206Heteroscedasticity test 207Variable 208VIF 209PARK 2102.52 211Chi2(42) = 89.01 212Prob>chi2 = 0.000*** 213BEDROOMS 2142.36 215STRUCTURE 2162.18 217BATHROOMS 2181.79 219220221===== PAGE 8 (score 28, numeric-paren 8, table-words 5, var-words 7) =====222223 My-Linh Thi Nguyen 224 225423 226TYPE 2271.73 228CENTER 2291.41 230SIZE 2311.14 232POOL 2331.01 234Mean VIF = 1.77 235Note: *** indicates significance at the 1% level. 236Source: Own study. 237 238The model is considered to experience no serious issues of multicollinearity (Mean 239VIF = 1.77). However, heteroscedasticity really exists at the 1% level of significance 240(Prob>chi2 = 0.000) (Table 3). This means that the estimation using the Ordinary 241Least Squares regression (OLS) does not provide a reliable result. Thus, robustness 242statistics is utilized for the estimation, following what have been suggested by White 243(1980) and recently performed by Bui (2020a). 244 245Table 4. Estimation results 246Dependent variable: House price (PRICE) 247Variables 248Coef. 249P>|z| 250Constant 25117.093*** 2520.000 253SIZE 2540.001** 2550.023 256BEDROOMS 2570.065*** 2580.000 259BATHROOMS 2600.003 2610.640 262CENTER 263-0.020*** 2640.000 265TYPE 2660.048** 2670.036 268POOL 2690.002 2700.709 271PARK 2720.187*** 2730.000 274STRUCTURE 2750.050*** 2760.000 277Number of obs 278205 279R-squared 28085.59% 281Significance level 282F(8, 196) = 276.08 283Prob>F = 0.000*** 284Note: ** and *** indicate significance at the 5% and 1% level, respectively. 285Source: own study. 286 287As can be seen from Table 4, the estimation results are significant at the 1% level 288(Prob>F = 0.000). Specifically, the independent variables can account for 85.59% 289variation in house price. It can be concluded that the model is appropriate and valid 290for the practical application. Alternatively, the hedonic pricing model is successfully 291applied in estimating house price in Vietnam. Furthermore, with a high explanation 292level of research model and highly-representative dataset, the results can be widely 293administered. 294 295 296Particularly, CENTER exerts the negative impact (β = -0.020) on PRICE at 1% level 297of significance whereas PRICE is positively influenced by SIZE (β = 0.001, at 5% 298level of significance), BEDROOMS (β = 0.065, at 1% level of significance), TYPE 299(β = 0.048, at 5% level of significance), PARK (β = 0.187, a 1% level of 300significance), STRUCTURE (β = 0.050, at 1% level of significance). Nevertheless, 301302303===== PAGE 9 (score 13, numeric-paren 0, table-words 1, var-words 11) =====304305 The Hedonic Pricing Model Applied to the Housing Market 306 307 424 308 309 310the study has not found the significant influence of BATHROOMS and POOL on 311PRICE. Hence, the estimated results of the model take the following equation: 312 313PRICE = 17.093 + 0.001 SIZE + 0.065 BEDROOMS - 0.020 CENTER + 0.048 314TYPE + 0.187 PARK + 0.050 STRUCTURE + ε 315 316Table 5. Results of hypothesis testing 317No 318Hypothesis 319Results 320H1 321SIZE ⟹ PRICE 322Accepted 323H2 324BEDROOMS ⟹ PRICE 325Accepted 326H3 327BATHROOMS ⟹ PRICE 328Rejected 329H4 330CENTER ⟹ PRICE 331Accepted 332H5 333TYPE ⟹ PRICE 334Accepted 335H6 336POOL ⟹ PRICE 337Rejected 338H7 339PARK ⟹ PRICE 340Accepted 341H8 342STRUCTURE ⟹ PRICE 343Accepted 344Source: Own study. 345 346- The impact of house size (SIZE) on house price (PRICE): The results show that 347SIZE positively affects PRICE. This indicates that houses in bigger size are usually 348priced more. This finding corroborates the reality and what has been reported by 349Kain and Quigley (1970), Selim (2009), Amenyah and Fletcher (2013), Shimizu 350(2014), Yayar and Demir (2014), Ndegwa (2018). Accordingly, big-size houses can 351meet demand of crowded families, bringing a wide and comfortable space. These 352houses can also specially be used for their business or offices. 353 354- The impact of the number of bedrooms (BEDROOMS) on house price (PRICE): 355The results show that BEDROOMS positively affects PRICE. This is consistent with 356the findings of Kain and Quigley (1970), Cebula (2009), Selim (2009), Baek et al. 357(2019). Accordingly, houses with more bedrooms suit families with many members. 358These houses allow their guests to stay over. As a result, they are more expensive. 359 360 361- The impact of house location (CENTER) on house price (PRICE): The study 362supports the negative influence of CENTER on PRICE. This shows that houses far 363from the center may sell for less than those with a close proximity to the center. This 364is because people living in the former find it difficult to commute to work or town 365center. Also, their public amenities are more limited than the latter. This is in line 366with the results in the studies of Selim (2009), Aluko (2011), Amenyah and Fletcher 367(2013), Yayar and Demir (2014), Ndegwa (2018). 368 369 370- The impact of house type (TYPE) on house price (PRICE): The findings reveal 371that PRICE is positively affected by TYPE. In the metropolises with a high 372population, the presence of garden in a house can help improve people’s living 373standard. Consequently, this kind of house is higher priced. This finding is consistent 374with what has been stated by Limsombunchai et al. (2004). 375