Analysis of positional aspects in the variation of real estate values in an Italian Southern Metropolitan area - Carmelo Maria Torre - Silvestro Montrone, Paola Perchinunno

International Conference on Computational

Science and Its Applications (ICCSA

2010)

2010 March, 23-26 Fukuoka

Analysis of positional aspects in the variation of real estate values in an Italian Southern Metropolitan

area

Carmelo Maria Torre

Department of Architecture and Urban Planning, Polytechnic of Bari,

Silvestro Montrone, Paola Perchinunno

Department of Statistical Science, University of Bari,

Analysis of positional aspects in the variation of real estate values in an Italian Southern Metropolitan area

Montrone, Torre, Perchinunno

Why this job?Why this job?

Property value appraisal and recaptureProperty value appraisal and recapture is the “ is the “chimerachimera” of all Italian ” of all Italian Researcher sin real estate marketResearcher sin real estate market

Main limitation are: Main limitation are: - - Cadastral system does not provide real and proper on Cadastral system does not provide real and proper on property property market of soils and buldings roperty. market of soils and buldings roperty.

- - Contracts of sales do not contain usually the true price of Contracts of sales do not contain usually the true price of propertyproperty

But: But: Property value can be considered as indirect effect of quality of Property value can be considered as indirect effect of quality of market and availability to residential servicesmarket and availability to residential services



2010)




Why this job?Why this job?

Property value can be identified indirectly Property value can be identified indirectly by the use of a fuzzy by the use of a fuzzy parameter, by the use of a measure of stability of a fuzzy function, parameter, by the use of a measure of stability of a fuzzy function, measuring the stability of property value-differential among different measuring the stability of property value-differential among different areas inside the cityareas inside the city

But: But: Property value can be considered as indirect effect of quality of Property value can be considered as indirect effect of quality of market and availability to residential services, therefore is possible to market and availability to residential services, therefore is possible to rank different market segmentation inside the cityrank different market segmentation inside the city



2010)




Analyzing the Analyzing the Stability of Stability of “Distance” of “Distance” of property value for property value for each quarter of the each quarter of the citycity

Preliminary analysis Preliminary analysis of stability of dataof stability of data

0

Palermo Palermo





0

NapoliNapoli





0

BariBari





0

X

jj

X

jjjjd dYdXYgXfYXdYdXXfYgXYygxfS )(),()(),())(),((

Fuzzy value of property - PalermoFuzzy value of property - Palermo





0

X

jj

X


Fuzzy value of property - NapoliFuzzy value of property - Napoli





0

X

jj

X






1 °

i

2

°

°

°

3°

°

°n

Espected Value Espected Value (euros/square (euros/square meters)meters)


0




1 °

i

2

°

°

°

3°

°

°n

Semantic Semantic Distance Distance Euros/(square Euros/(square meters)meters)


0




1

°

i

2

°

°

°

3 °

°° n

°°

°° °

°

°° °° °°

Stability index Stability index =expected value/semantic =expected value/semantic distance distance =1 ± =1 ± εε


0






0

X

jj

X


YXdYdX)Y(g),X(fYXdYdX)X(f),Y(gXY))y(g),x(f(SX

jj

X

jjjjd






0

Neighborhoods

Tor

re a

Mar

e

Sta

nic

S.S

pir

ito

S.P

asq

ual

e

S.P

aolo

S.N

icol

a

Pic

one

Pal

ese

Mu

rat

S.G

irol

amo

Mad

onn

ella

Los

eto

Lib

ertà

Jap

igia

Ceg

lie

Car

rass

i

Carbonara 20 479 -458 -491 154 466 -825 -541 -1554 21 -362 154 -379 -341 179 -512

Carrassi 533 992 54 21 666 979 -312 -29 -1041 533 150 667 133 170 691

Ceglie -158 300 -637 -670 -24 287 -1004 -720 -1733 -158 -541 -24 -558 -520

Japigia 362 820 -116 -150 495 -808 -483 -200 -1212 362 -20 495 -37

Libertà 400 858 -79 -112 533 845 -445 -162 -1175 400 16 533

Loseto -133 324 -612 -645 0 312 -979 -695 -1708 -133 -516

Madonnella 383 841 -95 -129 516 829 -462 -179 -1191 383

S.Girolamo 0 458 -479 -512 133 445 -845 -562 -1575

Murat 1575 2033 1095 1062 1708 2020 729 1012

Palese 562 1020 83 50 695 1008 -283

Picone 845 1304 366 333 979 1291

S.Nicola -445 12 -924 -958 -312

S.Paolo -133 324 -612 -645

S.Pasquale 512 970 33

S.Spirito 479 937

Stanic -458

Neighborhoods

Tor

re a

Mar

e

Sta

nic

S.S

pir

ito

S.P

asq

ual

e

S.P

aolo

S.N

icol

a

Pic

one

Pal

ese

Mu

rat

S.G

irol

amo

Mad

onn

ella

Los

eto

Lib

ertà

Jap

igia

Ceg

lie

Car

rass

i

Carbonara 328 523 579 540 320 493 809 677 1582 328 464 320 395 429 316 557

Carrassi 655 1050 476 448 694 1004 531 501 1052 655 458 694 383 396 744

Ceglie 348 326 742 695 270 329 1063 826 1762 348 595 270 558 583

Japigia 500 860 450 420 543 833 572 489 1237 500 396 543 313

Libertà 482 848 443 432 524 862 618 472 1224 482 404 524

Loseto 339 401 722 648 310 383 979 711 1705 339 550

Madonnella 533 866 534 486 569 871 660 558 1219 533

S.Girolamo 384 471 644 611 333 488 943 724 1620

Murat 1620 2080 1105 1132 1706 2035 785 977

Palese 731 1118 520 500 730 1049 514

Picone 952 1378 624 611 984 1319

S.Nicola 478 161 974 989 379

S.Paolo 339 401 722 648

S.Pasquale 610 986 529

S.Spirito 659 1023

Stanic 478






0

Neighbourhoods

Tor

re a

Mar

e

Sta

nic

S.S

pir

ito

S.P

asq

ual

e

S.P

aolo

S.N

icol

a

Pic

one

Pal

ese

Mu

rat

S.G

irol

amo

Mad

onn

ella

Los

eto

Lib

ertà

Jap

igia

Ceg

lie

Car

rass

i

Carbonara 0

,06 0

,92 0

,79 0

,91 0

,48 0

,95 1

,02 0

,80 0

,98 0

,06 0

,78 0

,48 0

,96 0

,79 0

,57 0

,92

Carrassi 0

,81 0

,94 0

,11 0

,05 0

,96 0

,98 0

,59 0

,06 0

,99 0

,81 0

,33 0

,96 0

,35 0

,43 0

,93

Ceglie 0

,45 0

,92 0

,86 0

,96 0

,09 0

,87 0

,94 0

,87 0

,98 0

,45 0

,91 0

,09 1

,00 0

,89

Japigia 0

,72 0

,95 0

,26 0

,36 0

,91 0

,97 0

,84 0

,41 0

,98 0

,72 0

,05 0

,91 0

,12

Libertà 0

,83 1

,01 0

,18 0

,26 1

,02 0

,98 0

,72 0

,34 0

,96 0

,83 0

,04 1

,02

Loseto 0

,39 0

,81 0

,85 1

,00 0

,00 0

,81 1

,00 0

,98 1

,00 0

,39 0

,94

Madonnella 0

,72 0

,97 0

,18 0

,27 0

,91 0

,95 0

,70 0

,32 0

,98 0

,72

S.Girolamo 0

,00 0

,97 0

,74 0

,84 0

,40 0

,91 0

,90 0

,78 0

,97

Murat 0

,97 0

,98 0

,99 0

,94 1

,00 0

,99 0

,93 1

,04

Palese 0

,77 0

,91 0

,16 0

,10 0

,95 0

,96 0

,55

Picone 0

,89 0

,95 0

,59 0

,55 0

,99 0

,98

S.Nicola 0

,93 0

,07 0

,95 0

,97 0

,82

S.Paolo 0

,39 0

,81 0

,85 1

,00

S.Pasquale 0

,84 0

,98 0

,06

S.Spirito 0

,73 0

,92

Stanic 0

,96






0

Quarters New property Good property Deprived property

Politeama 1,01 1,02 0,96

Libertà 0,95 0,95 0,96

Notarbartolo 0,95 0,95 0,96

Strasburgo 1,04 0,98 1,03

Unità d'Italia 1,04 1,04 0,95

San Lorenzo 0,85 0,96 0,85

Noce 0,98 0,96 0,99

Zisa 0,94 1,47 0,84

Oreto 0,85 1,47 0,84

Piazza Marina 0,97 1,00 0,93

Centro Storico 0,81 0,89 1,02

Calatafimi 0,98 0,97 0,93

Maqueda 0,93 0,81 0,83

Stability of differential value of Stability of differential value of property Palermoproperty Palermo







0


Chiaia 0,96 1,00 0,99

Posillipo 0,96 1,01 0,98

Vomero 0,97 0,98 0,99

Vomero alto 0,90 0,99 1,00

Arenella 0,91 1,00 0,96

Fuorigrotta 0,85 0,96 1,02

Cavalleggeri 0,98 0,96 0,94

Soccavo 1,85 0,94 0,90

Bagnoli 0,91 0,95 0,99

Centro storico/Antico 0,90 0,97 1,00

Materdei 1,02 0,90 0,91

C.V.E 0,94 0,99 0,96

Foria 0,96 0,92 0,93

G.B.Vico 0,97 0,92 0,91

Ferrovia 0,98 0,91 0,91

P.zza Carlo 3° 0,97 0,90 0,87

Centro direzionale 1,04 0,95 0,95

Capodimonte 0,91 0,97 0,97

S.Giovanni a Ted. 1,04 0,92 1,01

Secondigliano 0,96 0,94 0,95

Pozzuoli 0,93 0,98 0,96

Portici 1,04 0,95 0,90

Casoria 1,01 0,99 0,97

Casalnuovo 1,02 0,93 0,92

Sorrento e Penisola 0,96 1,03 0,99

Ischia 0,88 1,01 0,99

Capri 0,98 1,03 0,98

Quarto 1,05 0,96 0,95

Stability of differential value of Stability of differential value of propertypropertyNapoliNapoli






0


Madonella 1 0,96 0,96 0,91

Madonella 2 0,92 0,90 0,91

Madonella 3 0,92 0,86 0,91

Libertà 0,96 0,90 0,91

Murat 1 1,01 1,01 1,01

Murat 2 0,99 1,01 1,00

Città Vecchia1 0,96 1,00 0,96

Città Vecchia 2 0,99 1,24 0,97

Zona Faro

0,98 0,99 0,95

S.Girolamo-Fesca 0,90 1,04 0,99

Stanic 0,91 0,95 0,99

Poggiofranco 1 0,96 0,92 0,96

Poggiofranco 2 0,98 0,92 1,00

Carrassi 1

1,20 0,88 0,99

Carrassi 2

0,94 1,09 0,92

Picone 0,91 0,95 0,96

S.Pasquale 1 1,01 0,90 0,97

S.Pasquale 2 0,94 0,95 0,92

Japigia 1 1,27 0,97 0,96

Japigia 2 0,96 1,10 0,96

San Paolo

1,03 0,94 0,96

Carbonara-Ceglie 0,96 1,02 0,99

Loseto 0,98 0,95 1,00

Palese 0,92 0,95 0,97

S.Spirito 0,95 0,98 0,97

San Girgio

1,04 1,01 0,93

Torre a Mare 0,90 0,97 0,93

Stability of differential value of Stability of differential value of property Bariproperty Bari

PropertyProperty valuevalue Data (FIAIP)Data (FIAIP)

Value for each Value for each quarter of the city of quarter of the city of Bari, in the South of Bari, in the South of ItalyItaly

Quarters

New/refurbished(€/m2) Good state (€/m2) Degraded (€/m2)

min estate value

max estate value min estate value max estate valuemin estate

valuemax estate

value

Carbonara-Ceglie 2.200 2.500 1.900 2.200 1.300 1.700

Carrassi 2.600 3.800 2.200 3.300 1.500 2.600

Japigia 2.400 2.600 2.300 2.500 2.100 2.300

Libertà 2.700 3.500 2.200 2.800 1.800 2.200

Madonnella 2.400 3.500 2.350 3.200 2.300 2.600

S.Girolamo F. 2.200 4.500 2.000 3.000 1.800 2.400

Murat 3.700 5.500 3.200 4.500 3.000 3.500

Palese 2.000 2.600 1.400 1.750 800 1.000

Picone Poggiofr. 2.000 3.500 1.500 2.800 1.100 2.000

S.Nicola 1.500 4.200 1.100 3.000 1.000 2.200

S.Paolo 1.800 2.200 1.400 1.800 1.000 1.400

S.Pasquale 2.600 3.800 2.200 3.300 1.500 2.600

S.Spirito 1.900 2.400 1.300 1.600 800 1.000

Stanic 2.000 2.500 1.500 2.000 1.300 1.800

Torre M. S.Giorgio 1.400 2.500 1.050 1.500 810 1.250

1

Analysis of urban Analysis of urban difficulty difficulty distribution distribution

Identification of Identification of target areas for target areas for urban renewal urban renewal programmesprogrammes

Total fuzzy and Total fuzzy and relative relative approachapproach

2

Varied range of definitions on Urban-social exclusion – Housing Varied range of definitions on Urban-social exclusion – Housing

unavailability unavailability

Total Fuzzy and Relative Total Fuzzy and Relative

Use of Use of Fuzzy SetFuzzy Set to obtain a measurement of the to obtain a measurement of the incidence of relative incidence of relative

difficulty difficulty within a populationwithin a populationAnalysis of positional aspects in the variation of real estate values in an Italian Southern Metropolitan area


The fuzzy TFR approach consists in the definition of the The fuzzy TFR approach consists in the definition of the

measurement of a measurement of a degree of membership degree of membership of an individual to the of an individual to the

fuzzy totality of the difficulty conditions, in the range (0,1)fuzzy totality of the difficulty conditions, in the range (0,1)

Mathematically such a method consists of the construction of a Mathematically such a method consists of the construction of a

function of membership to "the fuzzy totality of difficulty" function of membership to "the fuzzy totality of difficulty"

The function subdividing in clusters of difficulty indicates The function subdividing in clusters of difficulty indicates a a

distribution rule of urban difficulty inside different urban realitiesdistribution rule of urban difficulty inside different urban realities



Supposing the observation of k indicators of poverty for every family, Supposing the observation of k indicators of poverty for every family,

the function of membership of i-th family to the fuzzy subset of the the function of membership of i-th family to the fuzzy subset of the

poor may be defined thus: poor may be defined thus:

niw

wxg

xfk

jj

k

jjji

i ,.....,1

).(

)(

1

1.

The values wj in the function of membership are just a weighting The values wj in the function of membership are just a weighting

system, whose explication is: system, whose explication is:

)(/1log jj xgw



Positional social

Quality of housing

Positional Economic quality of Housing

Low educational level

Unemployment rate

Housing Overcrowding

Rented dwellings

lack of functional services (landline telephone)

lack of functional services (heating systems, parking space)

Total difficulty



Distribution of intrinsic and positional quality of housing

Neighbourhoods

0,0 - 0,2property non-

decay

0,2 - 0,4slightly

property decay

0,4 - 0,6almost

property decay

0,6 - 0,8property decay

0,8 - 1,0Unquestionably

property decay

Carbonara 49 40 52 7 4

Carrassi 76 9 6 10 0

Ceglie 13 12 18 2 4

Japigia 51 16 6 3 2

Libertà 23 14 30 25 16

Roseto 5 3 4 1 0

Madonnella 20 6 8 4 11

S.Girolamo F. 21 28 15 6 2

Murat 106 15 5 2 0

Palese 28 16 15 0 1

Picone - Poggiofranco 106 16 13 4 3

S.Nicola 6 8 9 24 27

S.Paolo 12 32 27 11 3

S.Pasquale 40 14 5 7 2

S.Spirito 28 10 16 3 4

Stanic 5 8 6 1 1

Torre a Mare 17 18 14 2 0

Bari 606 265 249 112 80

After indicators-based clustering, by the imposition of spatial constraints (Patil and Taillie 2004), the approach suggested by

Kulldorff (1997Kulldorff (1997), has been used:

““Sat Scan” Sat Scan”

locating a circular window of arbitrary radius, and calculating the probability of poverty (risk) p1, inside the circle, or the probability of

poverty (risk) p2, outside the circle.

Of all windows, the minimum p-value (probability of critical region referring to the test) corresponds to the most important cluster.

The identification of a special area can be based on the intensity of a statistical attribute, instead of the number of attribute-characterised

elements.

Comparison among Comparison among different census different census sections difficulty sections difficulty distribution distribution

Graduation of Graduation of target areas for target areas for real estate fundingreal estate funding

Sat scan approachSat scan approach3

)(

)(

rand

NNdNNI

A

Nrand

N

dNNd

K

Iij 5.0

)(;)min(

)( 1

Nearest Neighbor Index Nearest Neighbor Index (Chou et al., 2003) use the analysis of

average among close and similar conditions:

if average is higher than the teorethical one: scattered

conditions

if average is lower than the teorethical one: concentrated

conditions



SAT SCAN – BARI Almost CriticalSAT SCAN – BARI Almost Critical

ClusterNumber of

casesinside Mean

outside Mean

Standard deviation p-value

Nearest Neighbour Index

1 59 1,900 2,880 1.04 0.0010 0.41555

2 212 1,970 3,000 0.99 0.0010 0.46437

3 210 2,070 2,980 1.01 0.0010 0.40666

4 114 5,020 2,630 0.82 0.0010 0.15294

Total 595





Final RemarksFinal Remarks

High instability of real estate dataHigh instability of real estate data

Good Possibility of predict data indirectly:Good Possibility of predict data indirectly:

Some open questions:Some open questions:

How to predict change in investor preference, when data How to predict change in investor preference, when data about property character do not change, but market about property character do not change, but market

trends are shocking varying?trends are shocking varying?



GRAZIEGRAZIE

Analysis of positional aspects in the variation of real estate values in an Italian Southern Metropolitan area - Carmelo Maria Torre - Silvestro Montrone, Paola Perchinunno

Documents

stability of distance

fuzzy value of property

city stability index

measure of stability

property value appraisal

n stability index

italian researcher

silvestro montrone