loans in light of the new support system the financial map: a graphical data-mining analysis (r...

33
LOANS IN LIGHT OF THE NEW SUPPORT SYSTEM: THE FINANCIAL MAP A GRAPHICAL DATA-MINING ANALYSIS (R SOFTWARE APPLICATIONS) FATMA ÇINAR MBA, CAPITAL MARKETS BOARD OF TURKEY Assoc. Prof. Dr. C. COŞKUN KÜÇÜKÖZMEN, İZMİR UNIVERSITY OF ECONOMICS Istanbul Conference Of Economics And Finance ICEF’14 08-09 September

Upload: fatma-cinar

Post on 12-Jul-2015

1.190 views

Category:

Data & Analytics


0 download

TRANSCRIPT

Page 1: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

LOANS IN LIGHT OF THE NEW SUPPORT SYSTEM: THE FINANCIAL MAP

A GRAPHICAL DATA-MINING ANALYSIS (R SOFTWARE APPLICATIONS)

FATMA ÇINAR MBA, CAPITAL MARKETS BOARD OF TURKEY

Assoc. Prof. Dr. C. COŞKUN KÜÇÜKÖZMEN, İZMİR UNIVERSITY OF ECONOMICS

Istanbul Conference Of Economics And Finance ICEF’14 08-09 September

Page 2: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Real Time Interactive Data Management for the Effect and Response Analysis

Technique; Graphical Datamining with Lattice and ggplot2 Graphical Packages of R Software

Investment Promotion

Dataset Graphical

Datamining Analysis

Page 3: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

AgendaData: Ministry of Economy and Foreign Investment Promotion Practice General Administration and BRSA*

Dataset: 6 Region Investment Promotion and 6 account period Graphical Datamining Analysis

Period: 2008-2013 Accounts

Dataset are factorized according to city and year factors.

Graphical Datamining applied on this factorized data.

Friday, December 26, 2014

*BRSA: Banking Regulations and Supervisison Agency

Page 4: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Purpose of the study

Friday, December 26, 2014

To Analyse various credit and financial situation of loans and loans defaults of some of the cities.Relationships and correlations were analyzed by R-based Graphic Data Mining program developed by us.

Page 5: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

The new incentive system was enacted by the Council of Ministers 5th June 2012 date and No. 2012/3305.

In this context, taking into account the level of development Turkey is divided into six regions .

The most developed provinces, are located on the first level of development while the provinces lowest level of development are classified as the sixth of the province .

07:55:26

Purpose of the study

Page 6: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

In this study, based on the entry into force of the

subsidies in question various types of

development and change in bank lending

analyzed by R based Graphical Datamining

Analysis Software we developed for this purpose

07:55:26

Purpose of the study

Page 7: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

In this study the data set is transformed into a

factor analysis based on the values of time and

space factors .

Visualization of the data contains valuable

findings for incentive system which differs

according to the terms of ratings criteria of

practitioners and banks.

07:55:26

Purpose of the study

Page 8: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

1st. Region 2nd. Region 3rd. Region 4th. Region 5th. Region 6th. Region

Ankara Adana Balıkesir Afyonkarahisar Adıyaman Ağrı

Antalya Aydın Bilecik Amasya Aksaray Ardahan

Bursa Bolu Burdur Artvin Bayburt Batman

Eskişehir Çanakkale Gaziantep Bartın Çankırı Bingöl

İstanbul Denizli Karabük Çorum Erzurum Bitlis

İzmir Edirne Karaman Düzce Giresun Diyarbakır

Kocaeli Isparta Manisa Elazığ Gümüşhane Hakkari

Muğla Kayseri Mersin Erzincan K.maraş Iğdır

Kırklareli Samsun Hatay Kilis Kars

Konya Trabzon Kastamonu Niğde Mardin Sakarya Uşak Kırıkkale Ordu Muş

Tekirdağ Zonguldak Kırşehir Osmaniye Siirt

Yalova Kütahya Sinop Şanlıurfa

  Malatya Tokat Şırnak

  Nevşehir Tunceli Van

Rize Yozgat

Sivas

8 City 13 City 12 City 17 City 16 City 15 City

Page 9: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Page 10: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

summary(Dataset)

ILKOD SEHIR SYIL NYIL AY

Min. : 1.00 ADANA : 22 Y2008:60 Min. :2008 Min. : 3.000

1st Qu.:16.00 ANKARA : 22 Y2009:60 1st Qu.:2009 1st Qu.: 3.000

Median :33.00 ANTALYA : 22 Y2010:60 Median :2010 Median : 6.000

Mean :29.67 BURSA : 22 Y2011:60 Mean :2010 Mean : 7.227

3rd Qu.:42.00 DENİZLİ : 22 Y2012:60 3rd Qu.:2012 3rd Qu.: 9.000

Max. :55.00 GAZİANTEP: 22 Y2013:30 Max. :2013 Max. :12.000

(Other) :198

SDONEM NDONEM TOPNAKDIKREDI NAKDIKREDI

D200803: 15 Min. :200803 Min. : 2301180 Min. : 2249452

D200806: 15 1st Qu.:200906 1st Qu.: 5006994 1st Qu.: 4775867

D200809: 15 Median :201011 Median : 9001388 Median : 8623443

D200812: 15 Mean :201035 Mean : 14542560 Mean : 14003463

D200903: 15 3rd Qu.:201203 3rd Qu.: 15756949 3rd Qu.: 15263775

D200906: 15 Max. :201306 Max. :113564461 Max. :110692193

(Other):240

TAKIPALACAK GNAKDIKREDI TASIT KONUT

Min. : 39600 Min. : 215400 Min. : 34377 Min. : 313429

1st Qu.: 251686 1st Qu.: 971274 1st Qu.: 70138 1st Qu.: 625852

Median : 339949 Median : 1923710 Median :106403 Median : 944120

Mean : 539097 Mean : 4654118 Mean :168232 Mean : 1740547

3rd Qu.: 599559 3rd Qu.: 2933005 3rd Qu.:213790 3rd Qu.: 1812319

Max. :2872268 Max. :62782383 Max. :789062 Max. :13037891

KMH DIGERTUKETICI KREDIKARTI TAKIPTASIT

Min. : 13457 Min. : 329107 Min. : 2929 Min. : 1454

1st Qu.: 37365 1st Qu.: 728700 1st Qu.: 503812 1st Qu.: 3398

Median : 56261 Median : 1215660 Median : 828166 Median : 5354

Mean : 88532 Mean : 1815939 Mean :1197707 Mean : 8673

Summary of the dataset

Page 11: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

In this section we investigate the effects of

various factors by the aid of gridplot programme

based on ggplot2 package of R software

Each grid represents 6 graphs describing the

cross effects and profiles of the variables

according to factors

07:55:26

DESCRIPTION OF GRID PANELS

Page 12: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Overall Promotion Regions ILKOD Vs

(Log10 scale) Default Energy

According to Region

Factorize Grid Graphics

Page 13: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

1.st Region ILKOD Vs (Log10

scale) Default Ebergy

According to the Year Factor

Grid Graphics

Page 14: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

1st Region ILKOD Vs

(Log10 scale)

Default Energy

According to Year

Factor Grid Graphics

Page 15: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Overall Regions Log10

Default Loans Vs

Log10 Default

Credit Cards According to

Year and Region

Factor Grid

Graphics

Page 16: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Overall Regions

Log10 Default Loans Vs

Log10 Default Credit Cards According to Year Factor Density and

Violin Graphs

Page 17: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Overall Region Log10 Default Loans

Vs Log10 Default

Mortgages According to

Year and Region

Factors Grid Graphics

Page 18: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Overall Region Log10 Default

Loans Vs Log10 Default

Mortgages According

to Year Factor

Density and Violin

Graphics

Page 19: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Baloon graphs of ggplot2 package can show us

3-dimensional relations distributed according 1-3

factors in scatterplot form.

With this type 2-dimensional numerical relations

can be represented under effect of 3rd numerical

value.

07:55:25

DESCRIPTION OF BALOON

GRAPHS

Page 20: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

1st Region Log10 Default

Loans Vs Log10 Default Energy against

Noncash Loans

According to Year and City

Factors Baloon

Graphics

Page 21: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

2nd. Region Log10

Default Loans Vs Log10

Default Energy against

Noncash Loans

According to Year and City

Factors Baloon

Graphics

Page 22: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

3rd. Region Log10 Default

Loans Vs Log10 Default Energy against

Noncash Loans

According to Year and City

Factors Baloon

Graphics

Page 23: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

4th. Region Log10 Default

Loans Vs Log10 Default Energy against

Noncash Loans

According to Year and City

Factors Baloon

Graphics

Page 24: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

5th. Region Log10 Default

Loans Vs Log10 Default Energy against

Noncash Loans

According to Year and City

Factors Baloon

Graphics

Page 25: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

6th. Region Log10 Default

Loans Vs Log10 Default Energy against

Noncash Loans

According to Year and City

Factors Baloon

Graphics

Page 26: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Facet graphs of ggplot2 package can show us 3-dimensional graphs distributed according 3 factors in matrix form.

In which we can see the anomalies occurs on which year and which region and which period.

Here we investigate default energy versus default loans bloonad by total loans according to region, year and period factors.

Colors period, balloons Total Cash loans.

07:55:26

DESCRIPTION OF FACET GRAPHS OF

GGPLOT2

Page 27: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

Overall Regions

Log 10 Default Loans

Vs. Log10 Default Energy

According to Year and Region

Factors Facet Graph

Page 28: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

With this graph we can see which region

represents anomalic behavior on which year and

which period under the effect of Total Cash

Credits.

3rd period of 4th region represents very anomalic

behaviour on the year 2008.

07:55:26

Overall Regions

Log 10 Default Loans

Vs. Log10 Default Energy

According to Year and Region

Factors Facet Graph

Page 29: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

07:55:26

With this study we investigate 6 Regions Investment Promotion and 6 account period by Graphical Datamining Analysis technique developed by us.

Period: 2008-2013 accounts.

Dataset are factorized according to city and year factors.

Graphical Datamining applied on this factorized data and financial anomalies dedected acording to time and space factors.

Page 30: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Concerning the energy investments 1st region. Promoting an increase in the proportion of the supports it received in the Energy field by years.

2.region non-performing loans in energy in year2009 is more risky comparing with other risky assets. On the other hand in 2013 the proportion of debt collection prone to decrease in non-performing loans of energy while the energy investments in a decrease.

For İzmir and Manisa; Manisa energy investments are ahead of İzmir.

07:55:26

Page 31: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

I would like to express my deep gratitude to;

Dr. Kutlu MERİH,

Dr. C. Coşkun KÜÇÜKÖZMEN

for their valuable contibutions,

Fatma ÇINAR

Page 32: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Friday, December 26, 2014

[email protected]

http://www.ieu.edu.tr/tr

[email protected]

http://www.coskunkucukozmen.com

[email protected]

http://www.spk.gov.tr/

http://www.riskonomi.com

@fatma_cinar_ftm

@ckucukozmen

@Riskonometri

@Riskonomi

@RiskLab Turkey

@datanalitik

@Riskanaltigi

tr.linkedin.com/in/fatmacinar/

tr.linkedin.com/in/coskunkucukozmen

Contact

Page 33: Loans In Light of the New Support System The Financial Map: A Graphical Data-Mining Analysis  (R Software Applications)

Küçüközmen, C. C. and Çınar F., (2014). “Modelling of Corporate Performance In Multi-Dimensional Complex Structured Organizations “CBBC” Management”, Submitted to the “2nd International Symposium on Chaos, Complexity and Leadership (ICCLS), December 17-19 at Middle East Technical University (METU), Ankara, Turkey.

Küçüközmen, C. C. ve Çınar F., (2014). “Finansal Karar Süreçlerinde Grafik-Datamining Analizi”, TROUGBI/DW SIG, Nisan 2014 İstanbul, http://www.troug.org/?p=684

Küçüközmen, C. C. ve Çınar F., (2014). “Görsel Veri Analizinde Devrim” Söyleşi, Ekonomik Çözüm, Temmuz 2014, http://ekonomik-cozum.com.tr/gorsel-veri-analizinde-devrim-mi.html.

Küçüközmen, C. C. ve Merih K., (2014). “Görsel Teknikler Çağı" Söyleşi, Ekonomik Çözüm, Temmuz 2014, http://ekonomik-cozum.com.tr/gorsel-teknikler-cagi.html

Küçüközmen, C. C. and Çınar F., (2014). “Banking Sector Analysis of Izmir Province: A Graphical Data Mining Approach”, Submitted to the 34th National Conference for Operations Research and Industrial Engineering (YAEM 2014), Görükle Campus of Uludağ University in Bursa, Turkey on 25-27 June 2014.

Merih, K. ve Çınar, F., (2013). “Modelling of Corporate Performance In Multi-Dimensional Complex Structured Organizations: “Cbbc” Approach”, Submitted to the EconAnadolu 2013: Anadolu International Conference in Economics III June 19-21, 2013, Eskişehir. http://www.econanadolu.org/en/index.php/articles2013/3683

Küçüközmen, C. C. and Çınar F., (2014). “New Sectoral Incentive System and Credit Defaults: Graphic-Data Mining Analysis”, Submitted to the ICEF 2014 Conference, Yıldız Technical University in İstanbul, Turkey on 08-09 Sep. 2014.

Pedroni M., and Bertrand Meyer (2009). “Object-oriented modeling of Object-Oriented Concepts”, ‘A Case Study in Structuring an Educational Domain’, Chair of Software Engineering, ETH Zurich, Switzerland. fmichela.pedroni|[email protected]

Merih, K. ve Çınar, F., (2013). “Modelling of Corporate Performance In Multi-Dimensional Complex Structured Organizations: “Cbbc” Approach”, Submitted to the EconAnadolu 2013: Anadolu International Conference in Economics III June 19-21, 2013, Eskişehir. http://www.econanadolu.org/en/index.php/articles2013/3683

07:55:26