Detektering och Visualisering av Långsiktiga Avvikande Handelsbeteenden
(2022) In Master's Theses in Mathematical Sciences FMSM01 20222Mathematical Statistics
- Abstract (Swedish)
- Olagliga transaktioner i form av bland annat penningtvätt, insiderhandel och marknadsmanipulation är stora problem som ofta glöms bort. Bara under 2020 tvättades 1.6 biljoner dollar världen över, motsvarande 2.7\% av världens totala BNP. Insiderhandel och marknadsmanipulation bidrar till misstro för marknaden vilket gör att den till lägre grad kan allokera pengar till företag från investerare då färre väljer att investera. Att detektera dessa problem är mycket viktigt, och är något som banker och institutioner, speciellt under senare år, har lagt stora resurser på. Att detektera dessa problem på ett statistiskt sätt är naturligt av stor vikt och är vad denna rapport fokuserar på. Att använda sig av aktör- och marknadsdata för att kunna... (More)
- Olagliga transaktioner i form av bland annat penningtvätt, insiderhandel och marknadsmanipulation är stora problem som ofta glöms bort. Bara under 2020 tvättades 1.6 biljoner dollar världen över, motsvarande 2.7\% av världens totala BNP. Insiderhandel och marknadsmanipulation bidrar till misstro för marknaden vilket gör att den till lägre grad kan allokera pengar till företag från investerare då färre väljer att investera. Att detektera dessa problem är mycket viktigt, och är något som banker och institutioner, speciellt under senare år, har lagt stora resurser på. Att detektera dessa problem på ett statistiskt sätt är naturligt av stor vikt och är vad denna rapport fokuserar på. Att använda sig av aktör- och marknadsdata för att kunna hitta avvikande handels- och transaktionsmönster är ett mycket användbart sätt för att kunna stoppa dessa olagligheter som ofta finansierar mycket stora delar av världens korruption, terrorism och annan illegala aktivitet. I projektet tas olika mått fram som sedan används för att hitta olika sorters avvikelser i marknadsaktörers (simulerade) handelsdata. Resultatet visualiseras i ett GUI som gör det smidigt för användaren att applicera måtten på individuella aktörer. Slutligen ges en kort sammanfattning av hur detta projekt och dess resultat kan utvecklas för att på ett ännu bättre sätt hitta avvikelser samt appliceras på verkliga scenarion. (Less)
- Abstract
- Illegal transactions in the form of money laundering, insider trading and market manipulation is a major problem that is often forgotten. In the year of 2020, 1.6 trillion dollars were successfully laundered, corresponding to 2.7\% of the world's total GDP. Insider trading and market manipulation contribute to mistrust of the market, resulting in worse capital distribution from investors to companies as fewer choose to invest. Detecting these problems is very important, and is something that banks and institutions, especially in recent years, have put a lot of resources into. Detecting these problems using statistics is naturally of great importance and is what this report focuses on. Using user and market data to find anomalous trade and... (More)
- Illegal transactions in the form of money laundering, insider trading and market manipulation is a major problem that is often forgotten. In the year of 2020, 1.6 trillion dollars were successfully laundered, corresponding to 2.7\% of the world's total GDP. Insider trading and market manipulation contribute to mistrust of the market, resulting in worse capital distribution from investors to companies as fewer choose to invest. Detecting these problems is very important, and is something that banks and institutions, especially in recent years, have put a lot of resources into. Detecting these problems using statistics is naturally of great importance and is what this report focuses on. Using user and market data to find anomalous trade and transaction patterns is a very useful way to stop these illegalities that often finance very large parts of the world's corruption, terrorism and other illegal activity. In the project, various "measures" are developed and then used to find various types of deviations in individuals' (simulated) trading data. These are visualized in a GUI that makes it easy for the user to apply the measurements to individual participants. Finally, a brief summary is given of how this project and its results can be further developed to find deviations in an even better way and be applied to real scenarios. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9103449
- author
- Pettersson, Mattias LU
- supervisor
- organization
- alternative title
- Detection and Visualization of Long-Term Anomalous Trading Behaviors
- course
- FMSM01 20222
- year
- 2022
- type
- H2 - Master's Degree (Two Years)
- subject
- keywords
- Money laundering, insider trading, market manipulation, statistics, outliers, visualization Penningtvätt, insiderhandel, marknadsmanipulation, statistik, extremvärden, visualisering
- publication/series
- Master's Theses in Mathematical Sciences
- report number
- LUTFMS-3461-2022
- ISSN
- 1404-6342
- other publication id
- 2022:E73
- language
- Swedish
- id
- 9103449
- date added to LUP
- 2022-11-24 11:17:13
- date last changed
- 2022-12-12 14:08:17
@misc{9103449, abstract = {{Illegal transactions in the form of money laundering, insider trading and market manipulation is a major problem that is often forgotten. In the year of 2020, 1.6 trillion dollars were successfully laundered, corresponding to 2.7\% of the world's total GDP. Insider trading and market manipulation contribute to mistrust of the market, resulting in worse capital distribution from investors to companies as fewer choose to invest. Detecting these problems is very important, and is something that banks and institutions, especially in recent years, have put a lot of resources into. Detecting these problems using statistics is naturally of great importance and is what this report focuses on. Using user and market data to find anomalous trade and transaction patterns is a very useful way to stop these illegalities that often finance very large parts of the world's corruption, terrorism and other illegal activity. In the project, various "measures" are developed and then used to find various types of deviations in individuals' (simulated) trading data. These are visualized in a GUI that makes it easy for the user to apply the measurements to individual participants. Finally, a brief summary is given of how this project and its results can be further developed to find deviations in an even better way and be applied to real scenarios.}}, author = {{Pettersson, Mattias}}, issn = {{1404-6342}}, language = {{swe}}, note = {{Student Paper}}, series = {{Master's Theses in Mathematical Sciences}}, title = {{Detektering och Visualisering av Långsiktiga Avvikande Handelsbeteenden}}, year = {{2022}}, }