Value-at-risk.
Item
-
Title
-
Value-at-risk.
-
Identifier
-
AAI9820560
-
identifier
-
9820560
-
Creator
-
Louisy-Louis, Valerie.
-
Contributor
-
Adviser: Salih Neftci
-
Date
-
1998
-
Language
-
English
-
Publisher
-
City University of New York.
-
Subject
-
Economics, Finance | Economics, General | Business Administration, Banking
-
Abstract
-
This paper develops a comprehensive framework of the new emerging market risk measurement model known as Value-at-Risk, VaR. It analyses, summarizes and integrates the existing literature which it supplements in two important respects. First, it presents, assesses and compares the three basic approaches to computing VaR--historical, analytical, and Monte Carlo simulation. Then, it explores proposed adjustments and improvements to the currently used models. Finally, it addresses the difficulties of model validity and accuracy testing techniques. Second, parametric and nonparametric forms of VaR models are applied to seventeen years of historical flow Jones Industrial Average data to appraise the approaches' respective ability to estimate market risk exposures against the history of profits and losses (P&L) incurred during the Great Depression on portfolio of blue chip stocks.;Statistical treatment of the data generally supports the theory. It confirmed that not only the methodologies can provide substantially different estimates but also that these VaRs are dependent on the choice of time horizon, reference period and confidence level. The application also confirmed that VaR estimates are more accurate in assessing market risk exposures under normal conditions and that they may exhibit greater failure rates than expected from their confidence level. Finally, it corroborated that shorter-time VaRs, parametric VaRs and higher confidence level VaRs are generally more accurate than others. Overall, even in a context of financial markets turmoil and economic depression, the VaR approaches did perform surprisingly well and provided accurate and reliable estimates of potential losses. This VaR application confirmed that the methodology is indeed a valuable market risk measurement technique that would apprise investors of blue chip stocks of their risk exposures and potential losses in a satisfying manner. In the end, the most interesting finding lies in the realization, that, had the VaR framework been discovered seventy years ago, the Great Depression might not have had such a tragic impact on investors in high quality stocks (and even maybe on the US financial industry)!
-
Type
-
dissertation
-
Source
-
PQT Legacy CUNY.xlsx
-
degree
-
Ph.D.
