1# RECORD 69288e963b8055b19e0fd077000b14eb34270ae12# TITLE: THE EFFECTS OF TRANSACTION HOTSPOTS AND FLIPPING HOTSPOTS ON HOUSING PRICES3# YEAR: 2024.0 VENUE: International Journal of Strategic Property Management DOI: 10.3846/ijspm.2024.208994# AUTHORS: Bor-Ming Hsieh; Chih-Yuan Yang5# PAGES: 15 CANDIDATE TABLE PAGES (1-based): [8, 9, 10, 13, 14, 15]6# ABSTRACT: In contrast to much of the literature that focuses on the issue of spatial dependence in housing price research, this study addresses the spatial aggregation of housing transactions and analyzes the effects of transaction hotspots and short-term flipping hotspots on housing prices by using real housing transaction data in Taipei City, Taiwan. The empirical results show that after controlling for the effects of spatial dependence and individual housing attributes, the impact of transaction hotspot areas on housing prices is significantly negative, while the impact of flipping hotspot areas on housing prices is significantly positive. The results verify that the key to driving up housing prices lies in flipping activities. Furthermore, the results of the spatial quantile regression model show that low-priced residential properties are more sensitive to the spatial concentration of housing transactions and flipping transactions in the housing market. Our results have implications for the government’s policy intending to control hot trade volumes to cool skyrocketing housing prices in a booming housing market. It is suggested that the government should pay attention to restraining short-term flipping activities in the housing market rather than setting constrains on housing transactions.789===== PAGE 1 (score 11, numeric-paren 0, table-words 0, var-words 11) =====1011This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), 12which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.13Copyright © 2024 The Author(s). Published by Vilnius Gediminas Technical University14ISSN: 1648-715X / eISSN: 1648-917915INTERNATIONAL JOURNAL 16OF STRATEGIC PROPERTY 17MANAGEMENT18INTERNATIONAL JOURNAL 19OF STRATEGIC PROPERTY 20MANAGEMENT21202422Volume 2823Issue 124Pages 1–1525https://doi.org/10.3846/ijspm.2024.2089926THE EFFECTS OF TRANSACTION HOTSPOTS AND FLIPPING HOTSPOTS ON 27HOUSING PRICES28Bor-Ming HSIEH1, Chih-Yuan YANG2,*291 Department of Land Management and Development, Chang Jung Christian University, Tainan, Taiwan302 Department of Civic Education and Leadership, National Taiwan Normal University, Taipei, Taiwan31Article History: 32Abstract. In contrast to much of the literature that focuses on the issue of spatial dependence in housing 33price research, this study addresses the spatial aggregation of housing transactions and analyzes the effects 34of transaction hotspots and short-term flipping hotspots on housing prices by using real housing transac-35tion data in Taipei City, Taiwan. The empirical results show that after controlling for the effects of spatial 36dependence and individual housing attributes, the impact of transaction hotspot areas on housing prices is 37significantly negative, while the impact of flipping hotspot areas on housing prices is significantly positive. 38The results verify that the key to driving up housing prices lies in flipping activities. Furthermore, the results 39of the spatial quantile regression model show that low-priced residential properties are more sensitive to the 40spatial concentration of housing transactions and flipping transactions in the housing market. Our results have 41implications for the government’s policy intending to control hot trade volumes to cool skyrocketing housing 42prices in a booming housing market. It is suggested that the government should pay attention to restraining 43short-term flipping activities in the housing market rather than setting constrains on housing transactions.44 ■received 31 August 202345 ■accepted 26 December 202346Keywords: housing prices, flipping, spatial aggregation, housing volume, hotspot analysis.47 * Corresponding author. E-mail: cyy@ntnu.edu.tw481. Introduction49Housing transactions and housing prices have usually 50been the most important issues in housing market studies. 51Previous studies have indicated that housing transactions 52and housing prices interact (Rothenberg et al., 1991). In 53many cases, housing transactions can serve as an indicator 54of whether the housing market is hot or not. A traditional 55expectation based on this viewpoint suggests that the vol-56ume of housing transactions predicts housing prices; for 57example, a rapid increase in housing transactions induces 58a rise in housing prices (Hua et al., 2001; Yiu et al., 2009). 59A boom in housing transactions and housing prices in an 60area often indicates that the housing market is prospering; 61hence, more house buyers are attracted, including owner-62occupiers and investors, and as a result, this drives hous-63ing prices up. However, it is difficult to distinguish rigid 64residential demand from speculative investing demand 65from observing housing transactions. By analyzing the fre-66quency of housing transactions in a certain period of time, 67it can differentiate buyers’ demand for owner-occupation 68or short-term speculation in the housing market (Depken 69et al., 2009).70In the housing market, the behavior of short-term buy-71ing and reselling properties with a profit can be seen as 72speculation or called flipping (LaCour-Little & Yang, 2023). 73Bayer et al. (2020) pointed out that investors who repeat-74edly buy and resell properties within two years are seen 75as flippers. Most flippers choose the location of property 76and time of investment and resell it when the housing 77price rises promptly (Yilmaz, 2014). Previous studies have 78indicated that short-term flipping in the housing market 79causes a boom in housing transactions and thus rapidly 80driving housing prices up, which is not conducive to the 81sound development of the housing market (Leung & Tse, 822017; Li et al., 2023). However, fewer studies have con-83cerned the spatial content of housing flipping and exam-84ined its impact on housing prices. 85Regarding the spatial concept of housing, adjoining 86housing units share similar environmental conditions, 87such as neighborhoods, parks, and open space. Hous-88ing prices of adjoining housing units are affected by each 89other, which is explained as the spatial dependence of 90house prices (Basu & Thibodeau, 1998; Case et al., 2004). 91In addition to the spatial dependence of housing prices, 92the concentration of traded residential properties also 93has spatial impacts (ripple effects) on the prices of ad-94jacent properties. As stated earlier, previous studies have 95indicated that a rapid increase in housing transactions 96tends to drive up housing prices. In this sense, a boom in 979899===== PAGE 8 (score 54, numeric-paren 29, table-words 5, var-words 7) =====1001018102B.-M. Hsieh, C.-Y. Yang. The effects of transaction hotspots and flipping hotspots on housing prices103accounting for 50.3% of the total repeated-sale samples.9 104This figure illustrates the gravity of the issue of flipping 105in Taipei’s housing market. By using a hot spot analysis 106approach to calculate the Getis-Ord Gi* statistic for each 107flipping sample in 50 × 50 m grid areas in the city, the hot-108spot areas of housing flipping are identified as red areas 109in Figure 2.110As shown in Figure 2, the red areas account for ap-111proximately 50% of areas with repeated sale samples and 112account for 1% of total housing transaction areas in the 113city, representing the most significant and high concentra-114tion of flipping transactions in the housing market over 115the past decade. The transaction hotspot areas and flip-116ping hotspot areas do not exactly overlap. Compared to 117the hotspot areas of housing transactions (the blue ar-118eas), the hotspot areas of housing flipping are more likely 119to be concentrated in western areas of the city, such as 120Zhongshan District, Datong District, Zhongzeng District, 121and Wanhua District, where they are old city center areas 122in Taipei City. Furthermore, most flipping hotspot areas are 123located in relatively lower-priced areas. This implies that 124lower-priced properties are more likely to be flipped than 125high-priced properties in Taipei City.1269 Of the 62,140 transactions in our sample, 3,302 occurred within 1272 years. The proportion of 5.31% is larger than the correspond-128ing ratio (3.13% = 3,133/100,076) in the paper of Li et al. (2023) 129which using the dataset of residential property transactions in 130Clark County, Nevada in the USA for the period 2003–2013.1314.2. Results of OLS model and spatial lag 132model133Regarding the housing price model, this study employs 134the spatial lag model to improve the spatial dependence 135of housing prices.10 The results of the OLS and the spa-136tial autoregression models are shown in Table 2. The full 137model results are shown in the Appendix. The results show 138that there is a significant and positive coefficient of spatial 139autocorrelation, suggesting that spatial autocorrelations 140exist significantly among housing prices. Thus, using the 141spatial lag regression can alleviate the problem of spatial 142dependence among housing prices and therefore improve 143the model goodness-of-fit. As shown in Table 2, the good-144ness-of-fit of the spatial lag model is better than that of 145the OLS model, while the R-square of the spatial lag model 146is slightly higher than that of the OLS model.11147In both models, the directions and coefficients of the 148controlling variables, such as district dummy variables, are 149as expected and consistent with prior studies. An increase 150of one square meter in floor area increases housing prices 151by 0.8%, and an increase of one bathroom rises housing 152prices by approximately 6%. In contrast, an increase of one15310 The Moran’s I for housing transaction prices is 0.1741, with a 154z-score of 859.99, indicating a significant positive spatial cor-155relation among housing transaction prices.15611 As expected, the results show the housing prices in all 11 dis-157tricts are significantly higher than in the Wanhua District. In ad-158dition, housing prices decreased compared to 2021, indicating 159a steady increase of housing prices over the past decade.160Table 2. Results of OLS model and spatial autoregression models161Model162OLS model163Spatial lag model164Spatial error model165Variable166Coeff.167t value 168Coeff.169z-statistic 170Coeff.171z-statistic 172Constant17316.1491741394.93***17516.1841761426.16***17716.2241781350.06***179Area1800.008181315.66***1820.008183316.60***1840.007185179.92***186Floor187–0.0004188–1.05 189–0.0001190–0.37 191–0.0008192–2.33** 193TStory194–0.0003195–1.01 196–0.0004197–1.33 198–0.0009199–2.69***200Bathroom2010.05920224.31***2030.06020425.31***2050.06120625.42***207Age208–0.011209–118.36***210–0.011211–121.44***212–0.011213–114.10***214Dist. MRT215–0.011216–40.04***217–0.011218–39.90***219–0.011220–37.84***221Dist. CC222–0.004223–32.22***224–0.003225–27.49***226–0.003227–31.16***228THA229–0.102230–34.74***231–0.099232–34.56***233–0.101234–34.42***235FHA2360.0142371.08 2380.0222391.70* 2400.0202411.55 242THA × FHA2430.0422441.68* 2450.0502462.02** 2470.0322481.30 249Administrative districts250Yes251Yes252Yes253Trading years × quarters254Yes255Yes256Yes257Spatial autocorr. coeff.258 –259 2600.19626152.50***2620.05626322.32***264(Pseudo) R-square2650.8532660.8592670.855268F value2696215.69***270–271–272AIC27314781.5527414678.0827514737.21276Note: ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. The number of observations is 62,140. The results of controlling dummies 277of the administrative district and trading year and quarter are provided in the Appendix.11278279280===== PAGE 9 (score 88, numeric-paren 54, table-words 5, var-words 11) =====281282International Journal of Strategic Property Management, 2024, 28(1), 1–152839284year in dwelling age decreases housing prices by approxi-285mately 1.1%. The distance to the nearest MRT station and 286the distance to the city center all have negative effects on 287housing prices, which indicates that the closer to the MRT 288station and to the city center, the higher the housing prices.289The coefficient of THA is –0.099 and significant in the 290spatial lag model, indicating that housing prices in trans-291action hotspot areas decrease 9.9% compared to those 292not in hotspot areas. This is probably because of the price 293comparison effect; that is, when there are many properties 294for sale in the market, the competitive effect for on-sale 295properties usually makes transaction prices down. Such re-296sults are compatible with the findings of Deng et al. (2022). 297Properties in the areas of many transactions have more 298sufficient information and more competitors and thus re-299duce their transaction prices.300Furthermore, the coefficient of FHA in the spatial lag 301model is 0.022, denoting that housing prices increase up 302to 2.2% in hotspot areas of flipping activities. These re-303sults are in line with the viewpoints of short-term flipping 304speculation in the housing market (Leung & Tse, 2017; 305LaCour-Little & Yang, 2023). House flipping increases 306housing prices and harms sound markets.12 The sum-307marizing effect of both THA and FHA on housing prices 308is –0.027 (–0.099+0.022+0.050), denoting that housing 309prices decrease by 2.7% in both hotspot areas of housing 31012 See article titled “How Tales of ‘Flippers’ Led to a Housing Bub-311ble” by Robert Shiller, at Economist’s View on May 18, 2017, 312available from https://economistsview.typepad.com/economists-313view/2017/05/how-tales-of-flippers-led-to-a-housing-bubble.html314transactions and hotspot areas of flipping activities. Since 315properties in the FHA drive up neighboring prices, the 316negative effect of the interactive THA and FHA on hous-317ing prices is less than that on THA only. In short, the above 318results suggest that a high concentration of flipping activi-319ties in the housing market, rather than a high concentra-320tion of normal housing transactions, is an important driv-321ing force of rising housing prices in the housing market 322of Taipei City.3234.3. Results of the spatial quantile regression 324model325This subsection discusses the results of the spatial quantile 326regression model, which estimates the effects of transac-327tion hotspot areas and flipping hotspot areas on various 328quantiles of housing prices with the consideration of spatial 329autocorrelation. The results are summarized in Table 3. The 330full model results are also shown in the Appendix. The trad-331ing floor (Floor) and total story (TStory) have positive effects 332on properties in middle to high housing price levels, show-333ing that the influence of being located on a high floor and 334high-rise buildings on housing prices increases with a rise 335in housing price quantile levels. The number of bathrooms 336(Bathroom) has a greater influence on properties at lower 337housing price levels than at higher price levels.338Table 3. Results of the spatial quantile regression model339Model340Low341……………………………….. Housing prices ………………………………..342High3430.1 q3440.25 q3450.5 q3460.75 q3470.9 q348Variable349Coeff.350t value351Coeff.352t value353Coeff.354t value355Coeff.356t value357Coeff.358t value359Constant36015.916 582.66***36116.067 991.19***36216.155 1303.56***36316.236 1297.69***36416.334 3651150.88***366Area3670.008 368131.06***3690.008 370230.96***3710.008 372305.09***3730.008 374300.55***3750.008 376265.16***377Floor378–0.002 379–2.15** 3800.001 3811.51 3820.001 3833.91***3840.002 3855.064***3860.002 3873.27***388TStory389–0.004 390–5.40***391–0.001 392–3.01*** 3930.001 3943.31***3950.003 3967.32***3970.003 3986.37***399Bathroom4000.069 40112.16***4020.061 40317.98***4040.055 40521.54***4060.054 40720.79***4080.049 40916.63***410Age411–0.012 412–55.95***413–0.012 –88.20***414–0.011 –105.21***415–0.010 –100.58***416–0.010 417–86.66***418Dist. MRT419–0.012 420–17.80***421–0.013 –33.05***422–0.011 423–38.23***424–0.010 425–32.26***426–0.009 427–25.01***428Dist. CC429–0.002 430–9.48***431–0.003 –16.04***432–0.003 433–26.20***434–0.004 435–29.40***436–0.004 437–26.04***438THA439–0.122 440–17.62***441–0.124 –30.11***442–0.097 443–30.80***444–0.061 445–19.22***446–0.038 447–10.64***448FHA4490.069 4502.24** 4510.050 4522.74***4530.014 4541.69* 455–0.018 456–1.26 457–0.024 458–1.51 459THA × FHA460–0.023 461–0.40 4620.037 4631.07 4640.056 4652.10** 4660.089 4673.30***4680.045 4691.45 470Administrative districts471Yes 472Yes 473Yes 474Yes 475Yes 476Trading years × quarters477Yes 478Yes 479Yes 480Yes 481Yes 482Spatial autocorr. coeff.4830.16348418.14***4850.185 48634.70***4870.200 48848.77***4890.214 49051.82***4910.217 49246.49***493(Pseudo) R-square4940.5774950.6064960.6414970.6714980.693499Note: ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively. The number of observations is 62,140. The results of controlling dummies 500of administrative district and trading year and quarter are provided in the Appendix.501502503===== PAGE 10 (score 15, numeric-paren 0, table-words 5, var-words 3) =====50450510506B.-M. Hsieh, C.-Y. Yang. The effects of transaction hotspots and flipping hotspots on housing prices507The spatial quantile regression allows us to assess 508the effects of THA at specific quantiles of the depend-509ent variable. Thus, we estimate the effects of THA in both 510low-priced (10th quantile) and high-priced (90th quantile) 511areas. The results consistently show negative and signifi-512cant coefficients for THA across all quantiles, indicating 513that THA remains a significant negative factor not only in 514low-priced areas but also in high-priced areas. In Table 3, 515for example, in the 0.1 q column (representing the 10th 516percentile of Taipei City housing prices), the coefficient for 517THA is –0.122 (t = –17.62; p-value < 0.01). Similarly, in the 5180.25 q column (representing the 25th percentile of Tai-519pei City housing prices), the coefficient for THA is –0.124 520(t = –30.11; p-value < 0.01). These results demonstrate that 521transaction hotspots are not synonymous with low-priced 522areas, as THA remains significantly negative in low-priced 523regions even after controlling for administrative districts 524and distance-related spatial variables.525Since this study addresses the influences of transac-526tion hotspots and flipping hotspots on housing prices, 527the coefficients of THA and FHA in different quantiles 528are depicted in Figure 3. It is interesting to observe that 529the coefficients of THA are –0.122, –0.124, –0.097, –0.061, 530and –0.038 at quantiles of 0.1 q, 0.25 q, 0.5 q, 0.75 q, and 5310.9 q, respectively.13 That is, these transaction hotspot ar-532eas have greater negative impacts on properties at lower 533price levels than at middle to high price levels. In other 534words, prices in lower-priced properties are more likely to 535be reduced where they are located in areas with a high 536concentration of housing transactions.537Furthermore, the coefficients of FHA are 0.069, 0.050, 5380.014, –0.018, and –0.024 at quantiles of 0.1 q, 0.25 q, 5390.5 q, 0.75 q, and 0.9 q, respectively. This indicates that 540flipping activities have greater positive effects on prop-541erties at lower price levels than at middle to high price 542levels. The lower-priced properties are more likely to be 543flipped and thus raise their prices in Taipei’s housing 544market. In addition, the coefficients in the 0.75 and 0.9 545quantiles are insignificant, indicating that the influences 546of flipping activities are insignificant and weak in higher 547priced properties. The effects of properties in both THA 548and FHA are significant on housing prices in the 0.5 and 5490.75 quantiles, respectively. The interactive effect is greater 550in the 0.75 quantile than in the 0.5 quantile. These results 551indicate that the interactive effect of THA and FHA have 552greater and significant influences on middle- to high-553priced houses in Taipei’s housing market. Simply stated, 554with a rise in housing price level, the negative effects of 555transaction hotspots and the positive effects of flipping 556hotspots on housing prices decrease. Low-priced residen-557tial properties are more sensitive to the spatial concentra-558tion of housing transactions and flipping transactions in 559the housing market of Taipei City. 5605. Conclusions561In contrast to much of the literature that focuses on the 562issue of spatial dependence in housing price research, this 563study addresses the spatial aggregation of housing trans-564actions and analyzes the effects of transaction hotspots 565and short-term flipping hotspots on housing prices by us-566ing real housing transaction data in Taipei City, Taiwan. The 567empirical results show that after controlling for the effects 568of spatial dependence and individual housing attributes, 569the impact of transaction hotspot areas on housing prices 570is significantly negative, while the impact of flipping hot-571spot areas on housing prices is significantly positive. The 572results verify that the key to driving up housing prices lies 573in flipping activities. Furthermore, the results of the spatial 574quantile regression model show that low-priced residential 575properties are more sensitive to the spatial concentration 57613 As shown in Figure 3, the trend of the coefficients of THA is 577rising. However, the coefficients of THA are negative, so the ris-578ing trend means the negative influences of transaction hotspot 579areas on housing prices are decreased.580Figure 3. The coefficients of THA, FHA, and THA × FHA in 581different quantiles582583584===== PAGE 11 (score 9, numeric-paren 0, table-words 0, var-words 8) =====585586International Journal of Strategic Property Management, 2024, 28(1), 1–1558711588of housing transactions and flipping transactions in the 589housing market. 590Hot spot analysis has been widely used to analyze the 591spatial correlation of housing prices in the housing market, 592but less research has addressed the effect of “hotspot are-593as” of two types of housing transactions on housing prices. 594This study provides another perspective to analyze the re-595lationships between trade volumes and prices in housing 596market research. The results off two important implications 597for government. First, areas of hot transactions will not 598drive housing prices up, but hot flipping areas will. The 599government should not be overly concerned about wheth-600er the housing market is hot or not but should pay atten-601tion to short-term flipping (repeated transactions within 6022 years). In addition, the areas with low-priced houses are 603easily influenced by spatial aggregation of trading volume. 604Flipping detection and support measures should focus on 605areas with low housing prices, where they are more sen-606sitively influenced by flipping transactions. Corresponding 607policies such as selective credit controls should refer to 608transaction (flipping) hotspot areas and set the optimal 609applicable areas of administrative districts.610Future research could explore more about the poten-611tial limitations of this study. First, this paper considers the 612hotspot effect through quantitative aggregation. However, 613if there is a significant urban‒rural disparity in certain re-614gions, areas with fewer residences are less likely to ex-615perience the emergence of hotspots. For instance, spot 616A mainly comprises low-density mansions, so there are 617few transactions, while spot B has many high-rise build-618ings and many transactions. However, the turnover rate 619of spot A, i.e., the ratio between the number of transac-620tions relative to the number of housing units, is higher 621than that in spot B. It is an important issue to compare 622the nuanced difference between the quantity and the ra-623tio concept. Fortunately, the urban characteristics of Taipei 624City, which has a relatively small geographical area and 625more balanced development, may mitigate such concerns. 626Second, the endogeneity of flipping activities is also an im-627portant consideration. For example, in a search-theoretic 628model of house flipping, as proposed by Leung and Tse 629(2017), flipping activities are considered endogenous. The 630relationships among housing prices, transactions, and flip-631ping may exhibit multidirectional influences. Finally, it is 632worth noting that the transaction and flipping hotspots 633may change over time. Identifying the time-varying trans-634action and flipping hotspots could provide more practical 635implications. For future research, employing space-time 636permutation scan statistic methods (such as Kulldorff & 637Nagarwalla, 1995; Kulldorff et al., 2005) to analyze changes 638in clusters over time would be particularly suitable.639Acknowledgements 640We would like to thank the Editors and three anonymous 641referees for providing highly constructive comments. The 642authors acknowledge the funding support received from 643the Chin-Oh Chang’s Inheritance Award, Chinese Society of 644Housing Studies. Special appreciation is extended to Pro-645fessor Chin-Oh Chang for his valuable and insightful com-646ments on this paper. Chih-Yuan Yang extends thanks to the 647National Science and Technology Council of Taiwan for fi-648nancial support through grant (NSTC 112-2410-H-003-131).649References 650Anacker, K. B., & Schintler, L. A. (2015). Flip that house: Visualising 651and analysing potential real estate property flipping transac-652tions in a cold local housing market in the United States. Inter-653national Journal of Housing Policy, 15(3), 285–303. 654https://doi.org/10.1080/14616718.2015.1051401655Anselin, L. (1988). Spatial econometrics: Methods and models. Klu-656wer Academic. https://doi.org/10.1007/978-94-015-7799-1657Anselin, L. (1992). Spatial data analysis with GIS: An introduction 658to application in the social sciences. National Center for Geo-659graphic Information and Analysis.660Anselin, L., & Bera, A. K. (1998). Spatial dependence of linear re-661gression models with an introduction to spatial econometrics. 662In A. Ullah & D. E. A. Giles (Eds.), Handbook of applied economic 663statistics (pp. 237–289). Marcel Dekker.664Basu, S., & Thibodeau, T. (1998). Analysis of spatial autocorrelation 665in house prices. Journal of Real Estate Finance and Economics, 66617(1), 61–85. https://doi.org/10.1023/A:1007703229507667Bayer, P., Geissler, C., Mangum, K., & Roberts, J. W. (2020). Specu-668lators and middlemen: The strategy and performance of inves-669tors in the housing market. The Review of Financial Studies, 67033(11), 5212–5247. https://doi.org/10.1093/rfs/hhaa042671Bilal, A., & Rossi-Hansberg, E. (2021). Location as an asset. Econo-672metrica, 89(5), 2459–2495. https://doi.org/10.3982/ECTA16699 673Bø, E. E. (2018). Housing match quality and demand: What can we 674learn from comparing buyer characteristics? Journal of Housing 675Economics, 41, 184–199. 676https://doi.org/10.1016/j.jhe.2018.06.007677Bourassa, S., Cantoni, E., & Hoesli, M. (2007). Spatial dependence, 678housing submarkets, and house price prediction. Journal of 679Real Estate Finance and Economics, 35, 143–160. 680https://doi.org/10.1007/s11146-007-9036-8681Bourassa, S., Cantoni, E., & Hoesli, M. (2010). Predicting house 682prices with spatial dependence: A comparison of alternative 683methods. Journal of Real Estate Research, 32(2), 139–159. 684https://doi.org/10.1080/10835547.2010.12091276685Carrillo, P., & Pope, J. (2012). Are homes hot or cold potatoes? The 686distribution of marketing time in the housing market. Regional 687Science and Urban Economics, 42(1–2), 189–197. 688https://doi.org/10.1016/j.regsciurbeco.2011.08.010689Case, B., Clapp, J., Dubin, R., & Rodriguez, M. (2004). Modeling spa-690tial and temporal housing price patterns: A comparison of four 691models. Journal of Real Estate Finance and Economics, 29(2), 692167–191. https://doi.org/10.1023/B:REAL.0000035309.60607.53693Chakravorty, S. (1995). Identifying crime clusters: The spatial prin-694ciples. Middle States Geographer, 28, 53–58.695Chang, C. O., & Chen, S. M. (2018). Dilemma of housing demand 696in Taiwan. International Real Estate Review, 21(3), 397–418. 697https://doi.org/10.53383/100267698Chang, C. O., & Hsieh, B. M. (2018). Changes in housing policy, 699housing wellbeing and housing justice in Taiwan. In R. L. H. Chu 700& S.-K. Ha (Eds.), Housing policy, wellbeing and social develop-701ment in Asia (pp. 88–105). Routledge. 702https://doi.org/10.1201/9781315460055-6703704705===== PAGE 13 (score 163, numeric-paren 131, table-words 1, var-words 3) =====706707International Journal of Strategic Property Management, 2024, 28(1), 1–1570813709Appendix710Table A1. Complete results of OLS model and spatial autoregression models711Model712OLS model713Spatial lag model714Spatial error model715Variable716Coeff.717t value 718Coeff.719Variable720Coeff.721Variable722Constant72316.1497241394.93***72516.1847261426.16***72716.2247281350.06***729Area7300.008731315.66***7320.008733316.60***7340.007735179.92***736Floor737–0.0004738–1.05 739–0.0001740–0.37 741–0.0008742–2.33** 743TStory744–0.0003745–1.01 746–0.0004747–1.33 748–0.0009749–2.69***750Bathroom7510.05975224.31***7530.06075425.31***7550.06175625.42***757Age758–0.011759–118.36***760–0.011761–121.44***762–0.011763–114.10***764Dist. MRT765–0.011766–40.04***767–0.011768–39.90***769–0.011770–37.84***771Dist. CC772–0.004773–32.22***774–0.003775–27.49***776–0.003777–31.16***778THA779–0.102780–34.74***781–0.099782–34.56***783–0.101784–34.42***785FHA7860.0147871.08 7880.0227891.70* 7900.0207911.55792THA × FHA7930.0427941.68* 7950.0507962.02** 7970.0327981.30799Administrative districts800Zhongzheng8010.25080237.19***8030.17480425.94***8050.23280634.38***807Datong8080.0368095.26***8100.0258113.70***8120.0348134.83***814Zhongshan8150.11981622.63***8170.07781814.82***8190.10282019.26***821Songshan8220.30082346.97***8240.25382540.03***8260.28482744.32***828Da’an8290.39583058.62***8310.27583239.49***8330.36183452.55***835Xinyi8360.28783744.97***8380.20983932.59***8400.26584141.20***842Shilin8430.29884445.58***8450.23584636.03***8470.28384843.23***849Beitou8500.14985117.49***8520.10285312.14***8540.14485516.93***856Neihu8570.19385835.74***8590.13386024.65***8610.18786234.63***863Nangan8640.25186537.74***8660.16086723.70***8680.24986937.47***870Wensha8710.0328725.86***8730.0168743.06***8750.0368766.62***877Trading years × quarters8782012Q3879–0.259880–24.86***881–0.260882–25.51***883–0.247884–23.76***8852012Q4886–0.243887–25.68***888–0.249889–26.90***890–0.231891–24.45***8922013Q1893–0.186894–18.83***895–0.190896–19.73***897–0.176898–17.95***8992013Q2900–0.130901–14.11***902–0.135903–15.00***904–0.123905–13.36***9062013Q3907–0.112908–11.56***909–0.118910–12.50***911–0.106912–11.04***9132013Q4914–0.093915–9.87***916–0.096917–10.40***918–0.087919–9.25***9202014Q1921–0.075922–7.49***923–0.077924–7.87***925–0.071926–7.09***9272014Q2928–0.131929–13.41***930–0.137931–14.37***932–0.122933–12.55***9342014Q3935–0.087936–8.60***937–0.092938–9.26***939–0.080940–7.88***9412014Q4942–0.089943–9.10***944–0.090945–9.35***946–0.086947–8.83***9482015Q1949–0.073950–6.71***951–0.074952–6.91***953–0.070954–6.46***9552015Q2956–0.059957–5.68***958–0.066959–6.54***960–0.054961–5.24***9622015Q3963–0.066964–6.16***965–0.068966–6.48***967–0.058968–5.43***9692015Q4970–0.095971–9.33***972–0.095973–9.54***974–0.087975–8.64***9762016Q1977–0.107978–8.92***979–0.109980–9.37***981–0.099982–8.36***9832016Q2984–0.064985–6.42***986–0.068987–6.93***988–0.058989–5.79***9902016Q3991–0.102992–9.38***993–0.104994–9.76***995–0.096996–8.83***9972016Q4998–0.107999–9.75***1000–0.1061001–9.83***1002–0.1001003–9.07***10042017Q11005–0.1091006–9.48***1007–0.1071008–9.52***1009–0.1011010–8.81***10112017Q21012–0.0931013–8.47***1014–0.0911015–8.50***1016–0.0841017–7.70***10182017Q31019–0.1151020–10.39***1021–0.1161022–10.71***1023–0.1061024–9.59***10252017Q41026–0.1301027–11.95***1028–0.1301029–12.19***1030–0.1211031–11.16***10322018Q11033–0.1111034–9.63***1035–0.1081036–9.56***1037–0.1021038–8.93***10392018Q21040–0.1231041–11.44***1042–0.1211043–11.55***1044–0.1151045–10.72***104610471048===== PAGE 14 (score 178, numeric-paren 139, table-words 2, var-words 4) =====10491050141051B.-M. Hsieh, C.-Y. Yang. The effects of transaction hotspots and flipping hotspots on housing prices1052Model1053OLS model1054Spatial lag model1055Spatial error model10562018Q31057–0.1041058–9.39***1059–0.1031060–9.47***1061–0.0971062–8.79***10632018Q41064–0.0731065–6.65***1066–0.0751067–7.00***1068–0.0651069–5.98***10702019Q11071–0.0861072–7.64***1073–0.0901074–8.15***1075–0.0791076–7.01***10772019Q21078–0.0911079–8.62***1080–0.0921081–8.98***1082–0.0851083–8.10***10842019Q31085–0.0821086–7.62***1087–0.0851088–8.03***1089–0.0771090–7.22***10912019Q41092–0.0741093–7.03***1094–0.0751095–7.28***1096–0.0701097–6.71***10982020Q11099–0.0711100–6.28***1101–0.0721102–6.54***1103–0.0651104–5.80***11052020Q21106–0.0531107–5.06***1108–0.0541109–5.23***1110–0.0471111–4.50***11122020Q31113–0.0361114–3.48***1115–0.0411116–4.13***1117–0.0321118–3.05***11192020Q41120–0.0341121–3.28***1122–0.0391123–3.81***1124–0.0311125–2.99***11262021Q11127–0.0191128–1.82**1129–0.0221130–2.09**1131–0.0181132–1.71*11332021Q21134–0.0141135–1.301136–0.0181137–1.65*1138–0.0131139–1.2011402021Q31141–0.0141142–1.251143–0.0171144–1.611145–0.0121146–1.111147Spatial autocorr. coeff.1148–1149 11500.196115152.50***11520.056115322.32***1154(Pseudo) R-square11550.85211560.85911570.8551158F value11596215.69***1160–1161AIC 116214781.55116314678.08116414737.211165Robust lag1166–1167683.37***1168Note: ***, **, and * denote significance at the 1%, 5%, and 10% levels, respectively.1169Table A2. Complete results of the spatial quantile regression model1170Model1171Low1172……………………………….. Housing prices ………………………………..1173High11740.1 q11750.25 q11760.5 q11770.75 q11780.9 q1179Variable1180Coeff.1181t value1182Coeff.1183t value1184Coeff.1185t value1186Coeff.1187t value1188Coeff.1189t value1190Constant119115.916 1192582.66***119316.067 1194991.19***119516.155 1303.56***119616.236 1297.69***119716.334 1150.88***1198Area11990.008 1200131.06***12010.008 1202230.96***12030.008 1204305.09***12050.008 1206300.55***12070.008 1208265.16***1209Floor1210–0.002 1211–2.15** 12120.001 12131.51 12140.001 12153.91***12160.002 12175.064***12180.002 12193.27***1220TStory1221–0.004 1222–5.40***1223–0.001 1224–3.01*** 12250.001 12263.31***12270.003 12287.32***12290.003 12306.37***1231Bathroom12320.069 123312.16***12340.061 123517.98***12360.055 123721.54***12380.054 123920.79***12400.049 124116.63***1242Age1243–0.012 1244–55.95***1245–0.012 1246–88.20***1247–0.011 –105.21***1248–0.010 –100.58***1249–0.010 1250–86.66***1251Dist. MRT1252–0.012 1253–17.80***1254–0.013 1255–33.05***1256–0.011 1257–38.23***1258–0.010 1259–32.26***1260–0.009 1261–25.01***1262Dist. CC1263–0.002 1264–9.48***1265–0.003 1266–16.04***1267–0.003 1268–26.20***1269–0.004 1270–29.40***1271–0.004 1272–26.04***1273THA1274–0.122 1275–17.62***1276–0.124 1277–30.11***1278–0.097 1279–30.80***1280–0.061 1281–19.22***1282–0.038 1283–10.64***1284FHA12850.069 12862.24** 12870.050 12882.74***12890.014 12901.69* 1291–0.018 1292–1.26 1293–0.024 1294–1.51 1295THA × FHA1296–0.023 1297–0.40 12980.037 12991.07 13000.056 13012.10** 13020.089 13033.30***13040.045 13051.45 1306Administrative districts1307Zhongzheng13080.227130914.01***13100.113131111.76***13120.130131317.66***13140.190131525.65***13160.226131726.85***1318Datong13190.01713201.65* 1321–0.0201322–2.11** 13230.01513242.02** 13250.03813265.06***13270.05413286.34***1329Zhongshan13300.01013310.83 1332–0.0161333–2.18** 13340.078133513.67***13360.130133722.51***13380.170133926.03***1340Songshan13410.280134218.37***13430.209134423.20***13450.227134632.81***13470.251134836.07***13490.270135034.13***1351Da’an13520.282135316.76***13540.233135523.41***13560.253135733.15***13580.294135938.27***13600.326136137.37***1362Xinyi13630.233136415.08***13650.159136617.36***13670.189136826.95***13690.229137032.37***13710.252137231.44***1373Shilin13740.236137515.03***13760.170137718.28***13780.213137929.98***13800.272138137.88***13820.308138337.82***1384Beitou13850.05113862.52** 13870.01313881.08 13890.102139011.11***13910.172139218.66***13930.199139418.95***1395Neihu13960.186139714.31***13980.080139910.36***14000.101140117.17***14020.141140323.62***14040.163140524.13***1406Nangan14070.236140814.52***14090.129141013.31***14110.143141219.40***14130.141141418.97***14150.118141614.04***1417Wensha14180.07014195.46***1420–0.0271421–3.49***1422–0.0071423–1.20 14240.01714252.85***14260.02314273.39***1428End of Table A1142914301431===== PAGE 15 (score 220, numeric-paren 179, table-words 1, var-words 1) =====14321433International Journal of Strategic Property Management, 2024, 28(1), 1–151434151435Model1436Low1437……………………………….. Housing prices ………………………………..1438High14390.1 q14400.25 q14410.5 q14420.75 q14430.9 q1444Trading years × quarters14452012Q31446–0.3231447–13.15***1448–0.2691449–18.48***1450–0.2541451–22.80***1452–0.2201453–19.58***1454–0.2061455–16.19***14562012Q41457–0.2941458–13.19***1459–0.2691460–20.35***1461–0.2431462–24.05***1463–0.2061464–20.19***1465–0.2001466–17.32***14672013Q11468–0.2581469–11.09***1470–0.2201471–15.99***1472–0.1751473–16.64***1474–0.1551475–18.30***1476–0.1451477–12.03***14782013Q21479–0.1851480–8.54***1481–0.1441482–11.17***1483–0.1201484–12.20***1485–0.1081486–10.83***1487–0.1081488–9.56***14892013Q31490–0.1701491–7.46***1492–0.1331493–9.84***1494–0.1091495–10.61***1496–0.0871497–8.34***1498–0.0971499–8.19***15002013Q41501–0.1441502–6.53***1503–0.1091504–8.33***1505–0.0861506–8.60***1507–0.0731508–7.16***1509–0.0731510–6.33***15112014Q11512–0.1181513–5.01***1514–0.0861515–6.16***1516–0.0721517–6.75***1518–0.0611519–5.67***1520–0.0531521–4.30***15222014Q21523–0.3251524–14.12***1525–0.1361526–9.93***1527–0.0911528–8.66***1529–0.0651530–6.13***1531–0.0591532–4.90***15332014Q31534–0.1091535–4.55***1536–0.1031537–7.30***1538–0.0931539–8.63***1540–0.0621541–5.66***1542–0.0781543–6.29***15442014Q41545–0.1111546–4.79***1547–0.0881548–6.38***1549–0.0681550–6.45***1551–0.0661552–6.26***1553–0.0691554–5.78***15552015Q11556–0.1381557–5.37***1558–0.1071559–7.05***1560–0.0671561–11.89***1562–0.0411563–3.50***1564–0.0401565–3.06***15662015Q21567–0.0981568–4.00***1569–0.0831570–5.71***1571–0.0541572–4.70***1573–0.0411574–3.65***1575–0.0581576–4.60***15772015Q31578–0.0931579–3.68***1580–0.0761581–5.07***1582–0.0701583–6.36***1584–0.0661585–5.72***1586–0.0711587–5.39***15882015Q41589–0.1321590–5.53***1591–0.1121592–7.88***1593–0.1031594–9.50***1595–0.0821596–7.44***1597–0.0811598–6.52***15992016Q11600–0.1281601–4.54***1602–0.1171603–6.99***1604–0.0971605–7.63***1606–0.0961607–11.62***1608–0.1121609–7.68***16102016Q21611–0.0961612–4.09***1613–0.0841614–6.00***1615–0.0771616–7.23***1617–0.0571618–5.26***1619–0.0571620–4.69***16212016Q31622–0.1361623–5.29***1624–0.1251625–8.21***1626–0.1121627–9.65***1628–0.0821629–7.00***1630–0.0891631–6.68***16322016Q41633–0.1291634–4.97***1635–0.1241636–8.06***1637–0.1161638–9.83***1639–0.0851640–7.17***1641–0.0861642–6.37***16432017Q11644–0.1391645–5.12***1646–0.1221647–7.56***1648–0.1151649–9.31***1650–0.1051651–8.42***1652–0.1101653–7.82***16542017Q21655–0.1091656–4.24***1657–0.1071658–7.02***1659–0.1041660–8.91***1661–0.0941662–7.96***1663–0.0841664–6.31***16652017Q31666–0.1481667–5.66***1668–0.1201669–7.71***1670–0.1221671–10.32***1672–0.1151673–9.63***1674–0.1201675–8.86***16762017Q41677–0.1581678–6.17***1679–0.1381680–9.05***1681–0.1351682–11.61***1683–0.1441684–12.26***1685–0.1221686–9.20***16872018Q11688–0.1291689–4.78***1690–0.1091691–6.77***1692–0.1181693–9.60***1694–0.1111695–8.97***1696–0.1071697–7.58***16982018Q21699–0.1531700–6.06***1701–0.1161702–7.75***1703–0.1231704–10.71***1705–0.1191706–10.31***17070.1221708–9.32***17092018Q31710–0.1141711–4.37***1712–0.1161713–7.46***1714–0.1121715–9.43***1716–0.0991717–8.31***1718–0.0891719–6.65***17202018Q41721–0.1021722–3.96***1723–0.0931724–6.09***1725–0.0861726–7.34***1727–0.0591728–4.98***1729–0.0731730–5.45***17312019Q11732–0.0861733–3.23***1734–0.0951735–6.01***1736–0.0971737–8.02***1738–0.0951739–7.83***1740–0.1081741–7.83***17422019Q21743–0.0931744–3.73***1745–0.0771746–5.23***1747–0.0951748–8.42***1749–0.0981750–8.64***1751–0.1111752–8.66***17532019Q31754–0.1001755–3.93***1756–0.0961757–6.35***1758–0.0931759–8.09***1760–0.0701761–6.04***1762–0.0871763–6.57***17642019Q41765–0.0901766–3.63***1767–0.0701768–4.77***1769–0.0841770–7.50***1771–0.0671772–5.89***1773–0.0821774–6.34***17752020Q11776–0.0691777–2.61***1778–0.0761779–4.83***1780–0.0771781–6.35***1782–0.0731783–5.96***1784–0.0921785–6.67***17862020Q21787–0.0491788–1.97** 1789–0.0541790–3.72***1791–0.0651792–5.80***1793–0.0601794–5.28***1795–0.0701796–5.48***17972020Q31798–0.0391799–1.69* 1800–0.0361801–2.53** 1802–0.0521803–4.72***1804–0.0511805–4.60***1806–0.0671807–5.33***18082020Q41809–0.0361810–1.48 1811–0.0411812–2.88***1813–0.0351814–3.19***1815–0.0381816–3.46***1817–0.0471818–3.77***18192021Q11820–0.0331821–1.31 1822–0.0201823–1.35 1824–0.0231825–2.04** 1826–0.0151827–1.34 1828–0.0261829–1.99** 18302021Q21831–0.0291832–1.14 1833–0.0221834–1.43 1835–0.0221836–1.84* 1837–0.0141838–1.16 1839–0.0301840–2.21** 18412021Q31842–0.0171843–0.67 1844–0.0151845–0.97 1846–0.0181847–1.55 1848–0.0161849–1.39 1850–0.0361851–2.67***1852Spatial autocorr. coeff.18530.163185418.14***18550.185 185634.70***18570.200 185848.77***18590.214 186051.82***18610.217 186246.49***1863(Pseudo) R-square18640.57718650.60618660.64118670.67118680.6931869End of Table A21870