東華大學圖書館 |

Macroeconomic and Financial Data Framework for Information Discovery and Predictive Analytics.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	Macroeconomic and Financial Data Framework for Information Discovery and Predictive Analytics./
作者:	Seth, Taruna.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2020,
面頁冊數:	166 p.
附註:	Source: Dissertations Abstracts International, Volume: 82-04, Section: B.
Contained By:	Dissertations Abstracts International82-04B.
標題:	Computer science. -
電子資源:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28092959
ISBN:	9798672198071

Macroeconomic and Financial Data Framework for Information Discovery and Predictive Analytics.
Seth, Taruna.

Macroeconomic and Financial Data Framework for Information Discovery and Predictive Analytics. - Ann Arbor : ProQuest Dissertations & Theses, 2020 - 166 p.

Source: Dissertations Abstracts International, Volume: 82-04, Section: B.

Thesis (Ph.D.)--State University of New York at Buffalo, 2020.

This item must not be sold to any third party vendors.

The U.S. financial markets play a critical role in contributing to the growth and development of the economy through various channels ranging from injecting liquidity for market participants through trading activity to facilitating international flow of funds between countries. Financial markets are driven by complex dynamics and interplay, often stemming from convoluted investor interactions, asset and inter-market complexities. Recent financial market events such as the sub-prime mortgage crisis of 2008, have necessitated the need for the development of strategies to deal with the acute stresses of renewed economic uncertainties, monitor systemic activities, and generate actionable intelligence. Despite several advancements, the modeling of financial markets events remains elusive due to complex interactions among the market constituents. In this dissertation, we contribute to this domain and address three key financial problems through the development of novel solutions to: (i) unravel linkages among institutional investments and equity market co-movements, (ii) forecast high impact economic events, and (iii) mine illegal trading activities driven by material non-public information.To address the first research problem , we develop a solution to uncover linkages between the common asset holdings by institutional investors and market co-movements. The current institutional investment landscape is very complex as the interactions among different market participants are not trivial or direct and therefore, their market impact may or may not be reflective of a simple aggregation of their individual interactions. In this study, we analyze whether the co-ownerships of stocks among institutional investors correlate with the market data movements and whether the interactions among these entities show discernible patterns capturing market fragility. We further evaluate the extent of these interactions and their prevalence among different investor preferences through the construction of distinct investor level portfolios that capture their trading inclinations for institutional holdings. Our results show that market co-movements are driven by institutional investor ownerships. The results also confirm that the portfolio selections made by the institutional investors influence the degree of equity market co-movements.We address the second research problem through the development and exploration of different techniques to forecast high impact economic events such as economic recessions and market corrections. Forecasting of widespread economic events, often characterized by significant decline in the economic activity extended over a period of few months, can enable the economists and policymakers to prepare and react pro-actively to such events thereby minimizing the risk or damage to the economy. To this end, we use a diverse set of financial and macro-economic parameters and propose multiple predictive models to forecast economic recessions over multiple horizon periods. Furthermore, we show the applicability of our predictive models in the area of market corrections forecasting. Specifically, we propose five different multi-variate models capable of producing accurate multi-horizon forecasts including a fusion model technique with multi-headed and multi-variate characteristics. The proposed fusion model approach yields the best performance results.We address the third research problem in this dissertation, through the development of a novel solution to mine illegal trading activities driven by material non-public information. We accomplish this task by deploying a multi-stage methodology that includes a predictive modeling approach without the added constraint of having data with the events of interest for training, an event mining methodology that utilizes unstructured and structured data, a classification, and an evaluation approach to identify the illegal events with good confidence. Our results confirm the efficacy of the proposed solution to identify illegal events in equity markets.

ISBN: 9798672198071Subjects--Topical Terms:

523869
Computer science.
Subjects--Index Terms:

US financial market

Macroeconomic and Financial Data Framework for Information Discovery and Predictive Analytics.
LDR:05585nmm a2200505 4500 001 2277531
005 20210521102017.5
008 220723s2020 ||||||||||||||||| ||eng d
020 $a 9798672198071
035 $a (MiAaPQ)AAI28092959
035 $a AAI28092959
040 $a MiAaPQ $c MiAaPQ
100 1 $a Seth, Taruna. $3 3555848
245 1 0 $a Macroeconomic and Financial Data Framework for Information Discovery and Predictive Analytics.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2020
300 $a 166 p.
500 $a Source: Dissertations Abstracts International, Volume: 82-04, Section: B.
500 $a Advisor: Chaudhary, Vipin.
502 $a Thesis (Ph.D.)--State University of New York at Buffalo, 2020.
506 $a This item must not be sold to any third party vendors.
520 $a The U.S. financial markets play a critical role in contributing to the growth and development of the economy through various channels ranging from injecting liquidity for market participants through trading activity to facilitating international flow of funds between countries. Financial markets are driven by complex dynamics and interplay, often stemming from convoluted investor interactions, asset and inter-market complexities. Recent financial market events such as the sub-prime mortgage crisis of 2008, have necessitated the need for the development of strategies to deal with the acute stresses of renewed economic uncertainties, monitor systemic activities, and generate actionable intelligence. Despite several advancements, the modeling of financial markets events remains elusive due to complex interactions among the market constituents. In this dissertation, we contribute to this domain and address three key financial problems through the development of novel solutions to: (i) unravel linkages among institutional investments and equity market co-movements, (ii) forecast high impact economic events, and (iii) mine illegal trading activities driven by material non-public information.To address the first research problem , we develop a solution to uncover linkages between the common asset holdings by institutional investors and market co-movements. The current institutional investment landscape is very complex as the interactions among different market participants are not trivial or direct and therefore, their market impact may or may not be reflective of a simple aggregation of their individual interactions. In this study, we analyze whether the co-ownerships of stocks among institutional investors correlate with the market data movements and whether the interactions among these entities show discernible patterns capturing market fragility. We further evaluate the extent of these interactions and their prevalence among different investor preferences through the construction of distinct investor level portfolios that capture their trading inclinations for institutional holdings. Our results show that market co-movements are driven by institutional investor ownerships. The results also confirm that the portfolio selections made by the institutional investors influence the degree of equity market co-movements.We address the second research problem through the development and exploration of different techniques to forecast high impact economic events such as economic recessions and market corrections. Forecasting of widespread economic events, often characterized by significant decline in the economic activity extended over a period of few months, can enable the economists and policymakers to prepare and react pro-actively to such events thereby minimizing the risk or damage to the economy. To this end, we use a diverse set of financial and macro-economic parameters and propose multiple predictive models to forecast economic recessions over multiple horizon periods. Furthermore, we show the applicability of our predictive models in the area of market corrections forecasting. Specifically, we propose five different multi-variate models capable of producing accurate multi-horizon forecasts including a fusion model technique with multi-headed and multi-variate characteristics. The proposed fusion model approach yields the best performance results.We address the third research problem in this dissertation, through the development of a novel solution to mine illegal trading activities driven by material non-public information. We accomplish this task by deploying a multi-stage methodology that includes a predictive modeling approach without the added constraint of having data with the events of interest for training, an event mining methodology that utilizes unstructured and structured data, a classification, and an evaluation approach to identify the illegal events with good confidence. Our results confirm the efficacy of the proposed solution to identify illegal events in equity markets.
590 $a School code: 0656.
650 4 $a Computer science. $3 523869
650 4 $a Finance. $3 542899
650 4 $a Computer engineering. $3 621879
650 4 $a International law. $3 560784
650 4 $a Criminology. $3 533274
650 4 $a American studies. $3 2122720
650 4 $a Public policy. $3 532803
653 $a US financial market
653 $a United States
653 $a Economic development
653 $a Market liquidity
653 $a International funds
653 $a Investor interactions
653 $a Institutional investments
653 $a Illegal trading activity
653 $a Non-public information
653 $a Investor holdings
690 $a 0984
690 $a 0508
690 $a 0464
690 $a 0323
690 $a 0630
690 $a 0501
690 $a 0601
690 $a 0616
690 $a 0627
710 2 $a State University of New York at Buffalo. $b Computer Science and Engineering. $3 1035503
773 0 $t Dissertations Abstracts International $g 82-04B.
790 $a 0656
791 $a Ph.D.
792 $a 2020
793 $a English
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28092959