Others

Thesis Woramanot Yomjinda

Description
Thesis Woramanot Yomjinda
Categories
Published
of 88
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  The Optimizer’s Guide to MoneyLaundering: Navigating DetectionAlgorithms with Na¨ıve Beam Search Woramanot YomjindaAdvisor: Professor Amir Ali AhmadiSubmitted in partial fulfillmentof the requirements for the degree ofBachelor of Science in EngineeringDepartment of Operations Research and Financial EngineeringPrinceton UniversityApril 2019  I hereby declare that I am the sole author of this thesis.I authorize Princeton University to lend this thesis to other institutions or individualsfor the purpose of scholarly research.Woramanot YomjindaI further authorize Princeton University to reproduce this thesis by photocopying orby other means, in total or in part, at the request of other institutions or individualsfor the purpose of scholarly research.Woramanot Yomjinda  Abstract In this paper, our goal is to optimize the total amount of monetary value we canlaunder within given constraints. We cast the problem into an undirected graph,with various financial institutions represented as nodes. To this end, we assumethat the problem is a complete graph and the more we move or “layer” the money,the harder it is for law enforcement to trace back to its illegal srcin, reducing riskof detection. As we progress, we also include other prominent components of themoney laundering process, such as shell companies. Because each transfer providesdifferent level of layering, for instance, international is harder to trace than national,the expected payoff of each route is a function of total layers minus total cost of moving money.The paper focuses on exploring the rich algorithms and constraints of the moneylaundering process against detection algorithms. To this end, we review extensivelystate-of-the-art techniques used by money launderers and state-of-the-art detectionalgorithms used by financial institutions. We progressively generate more complicatedmodels while retaining as much realism as possible.We first solve the optimization problem for the single best route, assuming thateach node can only be travelled through once. We then compare the practicalityand calculation time of each method against the CVX calculated upper bound andfind that the greedy algorithm family performs exceptionally well. After which, weexpand the question to include multiple routes without using each node more thanonce or less than a given cap when shell companies are incorporated. Naive beamsearch algorithms prove to be the best performing across all data. Adding a timeconstraint, we conclude that naive beam search performs over 98.7% of the upperbound and is ideal for path selection due to its low computational complexity.iii  Acknowledgements Many thanks to Professor Amir Ali Ahmadi for being an amazing advisor. I am verygrateful for his acceptance of the unusual topic of this paper and the overwhelmingsupport that he has provided me, as well as for giving me textbooks, a great amountof advice, and countless meetings. I am very happy that I have had the chance toexplore the topic of money laundering from a quantitative perspective and hopefullycontribute to a better understanding of this topic for future researchers and regulators.I also love and am beyond grateful for my family! I thank my mother, father,and older brother from the bottom of my heart for encouraging and cheering me onthrough Princeton and outside of Princeton.For the creation of this thesis, I am grateful for David Vegvari for his expertise onmoney laundering and corruption in Russia, and Dr. Sutthi Suntharanurak for adviceand details on money laundering regulators. I am also grateful for Cemil Dibek forhis support. Shout out to Andrew Inoue and Sherry Bai for their help in readingthrough and commenting on my thesis tirelessly.I am also happy that I have gotten to spent time at Princeton alongside myfriends at Colonial, ORFE, and Thaigers. You guys have made my college experiencean amazing memory that I will cherish, and I will be a life-long friend with all of youeven after Princeton!iv  Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iiiAcknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv 1 Introduction 1 1.1 Defining Money Laundering . . . . . . . . . . . . . . . . . . . . . . . 2 2 Literature Review 4 2.1 Quantifying Money Laundering . . . . . . . . . . . . . . . . . . . . . 42.1.1 Mathematical Model Review . . . . . . . . . . . . . . . . . . . 52.1.2 State-of-the-art Money Laundering Schemes . . . . . . . . . . 72.2 State-of-the-art Anti-money Laundering . . . . . . . . . . . . . . . . 102.2.1 Traditional Method and Anti-money Laundering Organizations 112.2.2 Money Laundering Detection Algorithms . . . . . . . . . . . . 122.2.3 Limitations and Loopholes . . . . . . . . . . . . . . . . . . . . 16 3 Single Path Optimization 18 3.1 Single Loop Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183.1.1 Base Model Assumptions . . . . . . . . . . . . . . . . . . . . . 183.1.2 Linear Programming Model . . . . . . . . . . . . . . . . . . . 203.2 Defining Upper bound . . . . . . . . . . . . . . . . . . . . . . . . . . 253.3 Approximation Algorithms . . . . . . . . . . . . . . . . . . . . . . . . 263.3.1 Generating L and C . . . . . . . . . . . . . . . . . . . . . . . 27v

Book Details

Oct 16, 2019

NOM035 MV

Oct 16, 2019
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x