Mining Big Data in Statistical systems of Monetary Financial...
Transcript of Mining Big Data in Statistical systems of Monetary Financial...
![Page 1: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/1.jpg)
Mining Big Data in Statistical systems of Monetary Financial Institutions
Afshin Ashofteh
www.linkedin.com/in/statas
![Page 2: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/2.jpg)
Section 1
Information Management and Statistics in Banking
![Page 3: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/3.jpg)
![Page 4: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/4.jpg)
Board
Specialised Committee of Information
and Technology Management
Strategic Management and Control
StatisticsDepartment
IT DepartmentDepartments
Information Management IT Management
Level 1
Level 2
Level 3
Decision
Strategic
Coordination
Operational
Coordination
Integrated Information Management in MFI’s
![Page 5: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/5.jpg)
Important points:
1. Statistics and IT departments are able to make big improvements and even a big bang.
![Page 6: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/6.jpg)
Section 2
Mining Big Data: in statistical systems of the monetary financial institutions (MFIs)
Past – Present - Future
![Page 7: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/7.jpg)
INTRODUCTION
An Insight into Banking System’s Big Data
![Page 8: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/8.jpg)
![Page 9: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/9.jpg)
![Page 10: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/10.jpg)
Main problems with Big Data
0 10 20 30 40 50 60
Data Protection & Security
Budgetin/setting priorities
Technical challenges of data…
Expertise
Lack of awareness of big data…
% of respondents
Capgemini Irving Fisher Committee
![Page 11: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/11.jpg)
Irving Fisher Committee on Central Bank StatisticsReport on Central banks’ use of and interest in “big data”(October 2015 – 69 Central Banks were participated)Europe, IFC members, Turkey and Banco Central del Paraguay
![Page 12: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/12.jpg)
Irving Fisher Committee on Central Bank StatisticsReport on Central banks’ use of and interest in “big data”(October 2015 – 69 Central Banks were participated)Europe, IFC members, Turkey and Banco Central del Paraguay
![Page 13: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/13.jpg)
Irving Fisher Committee on Central Bank StatisticsReport on Central banks’ use of and interest in “big data”(October 2015 – 69 Central Banks were participated)
• Conclusion 1: There is strong interest in big data in the central banking community, in particular at senior policy level.
• Conclusion 2: Central banks actual involvement in the use of big data is currently limited.
• Conclusion 3: Big data can be useful for conducting central bank policies.
• Conclusion 4: Big data are perceived as a potentially effective tool in supporting macroeconomic and financial stability analyses.
![Page 14: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/14.jpg)
Irving Fisher Committee on Central Bank StatisticsReport on Central banks’ use of and interest in “big data”(October 2015 – 69 Central Banks were participated)
• Conclusion 5: Big data may also create new information/research needs.
• Conclusion 6: International cooperation can add value.
• Conclusion 7: Exploring big data is a complex, multifaceted task.
• Conclusion 8: Regular production of big data-based information will take time, especially because of resource issues.
![Page 15: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/15.jpg)
Big Data does not replace banks’ current analytical infrastructure
but simply extends its scope
It has now become conceivable to conduct analyses based on the whole spectrum of data
available, not just a limited sample.
![Page 16: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/16.jpg)
Important points:
1. Statistics and IT departments mutually are able to make big improvement and even a big bang.
2. new algorithms and technologies are helping banks continually, they are also interested, but they need a platform to have personal data of clients and at the same time take care of TRUST and SATISFACTION of them.
![Page 17: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/17.jpg)
PAST…
Extracting information from official information of each clients.
![Page 18: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/18.jpg)
![Page 19: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/19.jpg)
HSBC found the primary barriers for internet banking were:
• Customer habit
• Security concerns
• And a lack of confidence.
They now have an active migration strategy to address
these concerns.
Part of the HSBC migration strategy is to enable customers
to undertake increasingly more complicated banking
activates via internet.
![Page 20: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/20.jpg)
Banque de France
![Page 21: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/21.jpg)
Banque de France
![Page 22: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/22.jpg)
NOW …
Extracting information from registered data by each clients.
![Page 23: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/23.jpg)
Banque de France
![Page 24: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/24.jpg)
Banque de France
![Page 25: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/25.jpg)
Important points:
1. Statistics and IT departments mutually are able to make big improvement and even a big bang.
2. new algorithms and technologies are helping banks continually, they are also interested, but they need a platform to have personal data of clients with taking care of TRUST and SATISFACTION of them.
3. Till now, registered data and official information of clients were used for different purposes.
![Page 26: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/26.jpg)
Now let’s talk about the Big Bang!
The Big Bang is Future!
The big bang is applying each client’s desires, decisions, risk appetite, wishes, profile, characteristics and personality as a data source into decision making and continuous improvement process of banks in benefit of both sides. Banks and Customers!
![Page 27: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/27.jpg)
FUTURE …
Extracting information from behaviors and desires of each clients.
![Page 28: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/28.jpg)
BigData
![Page 29: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/29.jpg)
We want to offer a solution to:
• Maximize Capital Adequacy Ratio of Bank to have more free
capital.
• Minimize the Risks.
• Maximize the Benefits of Clients.
• Maximize the Satisfaction of Clients.
• Minimize the requisite Client’s Trust to the bank’s activities.
ALL TOGETHER!
![Page 30: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/30.jpg)
MoneyMakers(M&M)
Imagine M&M mobile app based on Imbalanced Big Data mining and AI that recommend the investment opportunities with different risk levels to each of clients, help them to play a real investment game & give them the tips on their own risk appetite one by one!
Business field: Banking, Finance & Insurance.
Technology area: Big Data & Machine Learning
Business model: B2B
Try making Money by Investing and enjoy it like playing a GAME!
![Page 31: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/31.jpg)
Lack of blindness trust
Lack of Knowledge
Lack of Experience
Lack of Skills
Lack of Customization
Investm
ent
is R
isky
and U
np
leasan
tactivity
MoneyMaker(M&M)----------------------
A Mobile Application
Platform: Imbalanced
BigData mining + Machine
Learning + Text Mining +
Structural equation modeling
(SEM) + Enterprise risk
management (ERM)
Main
Pro
ble
m o
f C
usto
mers
![Page 32: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/32.jpg)
M&M Advantage 1
Adding pleasure into INVESTING activities like playing a
GAME
![Page 33: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/33.jpg)
M&M Advantage 2
A platform for the customers to play in Stretch zone
Bingo!
Customers will
ENJOY investment
and TAKE the RISKbut in an almost
SAFE zone out of panic zone
![Page 34: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/34.jpg)
Customers requirements to play in Stretch zone
![Page 35: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/35.jpg)
Internal sources (incl. Micro-databases)MoneyMakers Mobile App.
CBSDB Customer InfoCCRChat Experts recom
Text Mining Risk Analysis
Data Mart(Inf. Domain 1)
Big Data Mining
Re
fere
nce
Tab
les,
Me
tad
ata
e C
atal
ogs Structural equation modeling
Investment Activities
MoneyMaker Mobile App Risk Management Data Store
Acquisition Data Bases
...
Layer 1
Layer 3
Layer 4
Layer 5
Payments Operations
Operational Data StoreLayer 2
Recipients
Exploration of Information
Data Mart(Inf. domain N)Data Warehouse
Machine Learning
The
M&
M d
igit
al p
latf
orm
Mo
de
l
![Page 36: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/36.jpg)
Req
uir
em
en
ts
![Page 37: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/37.jpg)
bigger and better data might enhance the Bank’s
analytical toolkit and improve its operational
efficiency to maintain monetary and financial
stability -> Granular data & Quality of data
Req
uir
em
en
ts
![Page 38: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/38.jpg)
The MapReduce execution environment is the most common framework used in the scenario of Big Data.
Apache Spark is clearly
emerging as a more
commonly embraced
platform for implementing
Machine Learning
solutions that scale with Big
Data.
Req
uir
em
en
ts
![Page 39: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/39.jpg)
![Page 40: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/40.jpg)
DATA SCIENCE
Req
uir
em
en
ts
![Page 41: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/41.jpg)
Req
uir
em
en
ts
Structural Equation Modeling
CBSEM or PLS
![Page 42: Mining Big Data in Statistical systems of Monetary Financial Institutionscarmaconf.org/carma2018/wp-content/uploads/ppts/8570.pdf · 2019-08-07 · Irving Fisher Committee on Central](https://reader033.fdocuments.us/reader033/viewer/2022053009/5f0c74517e708231d4357b94/html5/thumbnails/42.jpg)
Right Information at the Right Time for the Right Client
Imbalanced Big Data Machine Learning Algorithm
Based on Structural Equation Modeling
M&M STARTUP