1# RECORD fe9b5a96e7007d52e9bebe9eaea418541aa3b0902# TITLE: An Automatic Tool for the Determination of Housing Rental Prices: An Analysis of the Italian Context3# YEAR: 2021.0 VENUE: Sustainability DOI: 10.3390/su140103094# AUTHORS: F. Tajani; F. Di Liddo; Rossana Ranieri; D. Anelli5# PAGES: 20 CANDIDATE TABLE PAGES (1-based): [10]6# ABSTRACT: In the last decades, some global events such as the economic crisis of 2008 and the COVID-19 emergency of 2020, have generated more attention towards the housing rental market and its capacity to meet several social needs. In order to study the existent demand for houses, then define the interventions on the residential assets to make them more affordable for the most fragile population groups, adequate evaluation tools are required. With reference to the residential property segment of five metropolitan cities located in the Italian territory, the present research is aimed at analyzing the contribution of the most influencing factors on rental prices. In particular, this research refers to the rented properties of the second half of 2019, with a set of variables that represent the intrinsic and extrinsic factors of the local market. The implementation of an automated valuation model allows the determination of the most significant factors and the functional relationships that they have with housing rental fees. The outputs obtained could support the improvement of equitable public housing policies or could guide private investment decisions, such as refurbishment interventions of certain significant factors that could increase the market rental value. This study is the first step in wider research that is currently in progress, which aims to investigate the effects of the existing COVID-19 pandemic on the residential rental market.789===== PAGE 1 (score 7, numeric-paren 0, table-words 0, var-words 7) =====1011 12 13 14 15Sustainability 2022, 14, 309. https://doi.org/10.3390/su14010309 16www.mdpi.com/journal/sustainability 17Article 18An Automatic Tool for the Determination of Housing Rental 19Prices: An Analysis of the Italian Context 20Francesco Tajani 1, Felicia Di Liddo 2, Rossana Ranieri 1 and Debora Anelli 2,* 211 Department of Architecture and Design, Sapienza University of Rome, Via Flaminia 359, 00196 Rome, Italy; 22francesco.tajani@uniroma1.it (F.T.); rossana.ranieri@uniroma1.it (R.R.) 232 Department of Civil, Environmental, Land, Building Engineering and Chemistry (DICATECh), Polytechnic 24University of Bari, Via Orabona 4, 70125 Bari, Italy; felicia.diliddo@poliba.it 25* Correspondence: debora.anelli@poliba.it 26Abstract: In the last decades, some global events such as the economic crisis of 2008 and the COVID‐2719 emergency of 2020, have generated more attention towards the housing rental market and its 28capacity to meet several social needs. In order to study the existent demand for houses, then define 29the interventions on the residential assets to make them more affordable for the most fragile 30population groups, adequate evaluation tools are required. With reference to the residential 31property segment of five metropolitan cities located in the Italian territory, the present research is 32aimed at analyzing the contribution of the most influencing factors on rental prices. In particular, 33this research refers to the rented properties of the second half of 2019, with a set of variables that 34represent the intrinsic and extrinsic factors of the local market. The implementation of an automated 35valuation model allows the determination of the most significant factors and the functional 36relationships that they have with housing rental fees. The outputs obtained could support the 37improvement of equitable public housing policies or could guide private investment decisions, such 38as refurbishment interventions of certain significant factors that could increase the market rental 39value. This study is the first step in wider research that is currently in progress, which aims to 40investigate the effects of the existing COVID‐19 pandemic on the residential rental market. 41Keywords: rent market; housing prices; automated valuation method; residential properties; Italian 42residential market 43 441. Introduction 45The recent rental market trend has constantly changed and is characterized by rapid 46bargaining, increasing rents, and high demand. Leased housing has gradually become an 47essential supplement to homeownership: worldwide, 1.2 billion people prefer this type of 48accommodation [1]. In Italy, leased housing has recently grown due to the various forms 49of leasing that ensure a tax benefit for the owners. The “short‐term” rental is the segment 50affected by a significant transformation, with a constant evolution in the supply of houses 51for temporary or tourist use. An analysis of the rental market, based on a sample of online 52ads just prior to the COVID‐19 pandemic, showed an increase in rents per m2 in the 53provincial capital municipalities. This has increased from 1.6% in 2017 to 3.3% in 2019, 54with the towing of smaller homes that are better furnished and in a good state of 55conservation [2]”. The situation has changed after the COVID‐19 emergency occurred, 56where the drop in rental fees was higher in the biggest cities (−9.5%) where a larger 57number of students and workers have been detected, and the supply of housing 58properties to be leased has increased more than in smaller towns (−3.8%). The number of 59regularly registered lease contracts have grown by +4.2% compared to 2018, and most of 60them (about 80% of the total number of properties) are houses [3]. It represents a relevant 61share of the market demand, where in 2018 the Bank of Italy demonstrated that about 20% 62Citation: Tajani, F.; Di Liddo, F.; 63Ranieri, R.; Anelli, D. An Automatic 64Tool for the Determination of 65Housing Rental Prices: An Analysis 66of the Italian Context. Sustainability 672022, 14, 309. https://doi.org/ 6810.3390/su14010309 69Academic Editors: Grazia Napoli 70Received: 5 December 2021 71Accepted: 24 December 2021 72Published: 28 December 2021 73Publisher’s 74Note: 75MDPI 76stays 77neutral with regard to jurisdictional 78claims in published maps and 79institutional affiliations. 80 81Copyright: © 2021 by the authors. 82Licensee MDPI, Basel, Switzerland. 83This article is an open access article 84distributed under the terms and 85conditions of the Creative Commons 86Attribution 87(CC 88BY) 89license 90(https://creativecommons.org/license91s/by/4.0/). 929394===== PAGE 10 (score 69, numeric-paren 57, table-words 4, var-words 0) =====9596Sustainability 2022, 14, 309 9710 of 20 98 99Application of the Methodology 100In the present analysis, the EPR methodology is iteratively implemented for each 101sample collected, by considering the structure of the generic expression identified in 102Equation (1) without the function (f) and with the dependent variable Rp (rental price) in 103logarithmic form (Y = ln(Rp)).Each additive monomial term of the mathematical 104expression is assumed to be a combination of the input variables raised to the proper 105numerical exponents. In particular, the candidate exponents selected from the research 106belong to the set (0; 0.5; 1; 2) and the maximum number l of additive terms in the final 107expressions are assumed to be eight. 108The implementation of the econometric technique for each Italian city considered has 109generated several models. For each study sample, the choice of the “best” model to be 110analyzed for the functional relationships between the explanatory variables and the rental 111prices has been carried out, taking into account (i) the statistical performance level, (ii) the 112complexity of the algebraic expressions, and (iii) the empirical coherence of the 113coefficient’s signs. The first aspects mentioned are immediately examined as the CoD is 114associated by EPR with each model and the mathematical form is visible in the quantity 115of the terms of the equation and in the combination of the variables in each term. The 116empirical coherence of the functional relationships between the input variables in each 117model and the rental prices, is a less immediate operation to be carried out by the user, 118especially if there are more variables combined in the terms of the equation and they occur 119repeated more times 120The five models chosen between those provided by the EPR technique for each study 121sample are reported in Table 2. 122Table 2. EPR models for each study sample. 123City 124Equation 125MILAN 126Y = +0.74039 * Fc + 0.65355 * W0.5 Fp0.5 + 0.85473 * W0.5 * Fsc2 + 11.8431 * Ae0.5 * 127Uc * Dp0.5 * Du * Fsc0.5 + 0.91529 * Ab * L0.5 * Ce0.5 + 3.1017 * Ai0.5 − 2.0858 * Ai0.5 128* L0.5 * Dp + 2.5784 * Ai * L0.5 * Ce0.5 * Eab2 + 5.2917 129TURIN 130Y = + 0.20484 * Fc + 0.20638 * Ce + 2.3667 * Ai0.5 − 0.90861 * Ai0.5 * Dp0.5 * Du0.5 + 1313.1746 * Ai0.5 * Eab0.5 * Du * C + 0.79503 * Ai0.5 * Ae0.5 Cg * Du + 4.2823 * Ai 132*Ag0.5 * W0.5 * Dm + 5.1051 133ROME Y = −0.95769 * Dp0.5 + 0.32292 * Ce2 * Dm0.5 + 0.72193 * W2 + 8.6191 * L * Ecde * 134Fsc + 0.58693 * Ab0.5 * Fc2 + 2.4715 * Ai0.5 + 6.3407 135NAPLES Y = −0.65252 * Dg + 0.79149 * W2 Fsc2 + 0.72862 * L * Uc + 1.5006 * Ai0.5 + 1360.96063* Ai0.5 * Ds0.5 * Fc0.5 + 5.7416 137CATANI138A 139Y = + 2.9701 * Eab0.5 * Ds * Dm2 + 0.11784 * Ce0.5 + 0.57677 * W2 + 4.2099 * L0.5 * 140Uc * Fc0.5 + 3.3906 * Ab0.5 * K * Du2 + 0.73513 * Ag2 + 1.5162 * Ai0.5 − 13.4711 * 141Ai0.5 * Uc0.5 * Ds * Dm + 5.5681 142First of all, it should be noted that the models selected are characterized by a different 143algebraic form complexity, and that they have an average CoD equal to about 80% (88.69% 144for the city of Milan, 79.31% for the city of Turin, 86.87% for the city of Rome, 74.92% for 145the city of Naples, and 67.06% for the city of Catania). 146The statistical performance indicator associated by EPR to each model shows a high 147statistical reliability in terms of the coherence of the detected data. This is confirmed by 148other calculated statistical indices, i.e., the root mean square error (RMSE), which refers 149to the square root of the mean of the squared errors between the prices of the original 150sample and the values estimated through the model the mean absolute percentage error 151(MAPE), which refers to the average percentage error between the detected prices and the 152values estimated; and the maximum absolute percentage error (MaxAPE), which refers to 153the maximum percentage error between the detected prices and the values estimated 154155156===== PAGE 11 (score 17, numeric-paren 0, table-words 2, var-words 13) =====157158Sustainability 2022, 14, 309 15911 of 20 160 161through the model. Furthermore, in order to analyze the stability of each functional model 162selected, and to flag problems like overfitting (or selection bias), a ten‐fold cross‐163validation [57] has been implemented on the starting database of each study sample, 164where the outputs obtained confirm the good performance prediction of the functional 165models. 166With reference to each sample detected for the Italian cities analyzed in the present 167research, in Table 3 the variables selected by EPR as the most influential on the residential 168rental prices are reported. Regarding the intrinsic variables selected by the model for the 169five cities, it should be pointed out that the internal area and the number of bathrooms are 170included in all models, whereas the bad quality of the property maintenance conditions 171are not in any models. The presence of the kitchen located in the same living room of the 172property is considered only in the case of the city of Catania, whereas the property EPC 173labels from C to E are relevant in the residential market of the city of Rome. With reference 174to the extrinsic variables, the property location in the central municipal trade area 175constitutes one of the most influencing factors considered by potential renters, as it is 176included in all models obtained for the study samples of the cities analyzed in the present 177research. 178Table 3. The variables selected by EPR for each city’s model. 179 180MILAN 181TURIN 182ROME 183NAPLES 184CATANIA 185Internal area 186◉ 187◉ 188◉ 189◉ 190◉ 191Surface area of 192balconies 193◉ 194 195◉ 196 197◉ 198Surface area of 199green areas 200 201◉ 202 203 204◉ 205Condominium 206areas 207◉ 208◉ 209 210 211 212Floor 213◉ 214 215◉ 216◉ 217◉ 218Bathroom 219◉ 220◉ 221◉ 222◉ 223◉ 224Kitchen 225 226 227 228 229◉ 230Bad quality of 231property 232maintenance 233conditions 234 235 236 237 238 239Good quality of 240property 241maintenance 242conditions 243 244◉ 245 246 247 248Excellent quality of 249property 250maintenance 251conditions 252◉ 253◉ 254◉ 255 256◉ 257EPC labels from A4 258to B 259◉ 260◉ 261 262 263◉ 264EPC labels from C 265to E 266 267 268◉ 269 270 271Age of the building 272◉ 273 274 275◉ 276◉ 277Subway 278 279◉ 280◉ 281 282◉ 283Central train 284station 285 286 287 288◉ 289◉ 290University center 291◉ 292◉ 293 294 295◉ 296