Težki repi in tveganja v financah

magistrsko delo

Timotej Oražem (Author), Tomaž Košir (Mentor)

Abstract

Pri upravljanju s tveganji je pomembna porazdelitev vrednosti finančnih instrumentov. Precej časa se je domnevalo, da so te porazdeljene normalno. Skozi čas so s pomočjo empiričnih dokazov zavrnili to hipotezo in se nagibajo k porazdelitvam s težkimi repi. V primerjavi z normalno porazdelitvijo so pri težkorepih porazdelitvah večja odstopanja od povprečne vrednosti. Se pravi, da je pri normalni porazdelitvi manjša verjetnost, da se bo zgodil ekstremni dogodek. V obdobju po finančni krizi leta 2009 so finančne inštitucije postale še bolj pozorne na tveganja, ki jih prinesejo ekstremni dogodki. Te ekstremne izgube, ki so se zgodile v krizi, so sprožile vprašanja o ustreznosti in pravilnosti modelov za upravljanje s tveganji, ki večinoma temeljijo na normalni porazdelitvi. Zaradi vedno več dokazov, da so finančni instrumenti porazdeljeni s porazdelitvami s težkimi repi, je prišlo do poudarka na modeliranju težkih repov in do izboljšav modelov za modeliranje ekstremnih dogodkov. Cilj mojega dela je opisati, kako težki repi vplivajo na upravljanje s tveganji, in predstaviti nekaj metod, ki so pogosto uporabljene pri delu s težkimi repi. Za ilustracijo bom različne metode uporabil na primeru indeksa NASDAQ.

Keywords

težkorepe porazdelitve;normalna porazdelitev;VaR;tvegana vrednost;pričakovan izpad;teorija ekstremne vrednosti;EVT;testiranje modelov;

Data

Language:	Slovenian
Year of publishing:	2018
Typology:	2.09 - Master's Thesis
Organization:	UL FMF - Faculty of Mathematics and Physics
Publisher:	[T. Oražem]
UDC:	519.2
COBISS:	18451801
Views:	829
Downloads:	295
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Financial risk and heavy tails
Secondary abstract:	In risk management it is very important to assess how financial instruments are distributed. For a long time, it was assumed that the distribution of financial instruments is normal. Over time they have rejected this hypothesis through empirical evidence and are now leaning toward distributions with heavy tails. Compared with normal distribution, deviations from average value are higher for heavy-tailed distributions. This means that in a normal distribution, it is less likely that an extreme event will occur. In the period after the financial crisis in 2009, financial institutions have become even more attentive to the risks brought about by extreme events. These extreme losses that occured during the crisis raised questions about the appropriateness and correctness of risk management models, which are mostly based on normal distribution. Due to the growing evidence that financial instruments are distributed with heavy-tailed distributions, emphasis has been placed on modeling heavy tails and on upgrading models for modeling extreme events. The goal of my thesis is to describe how heavy tails affect risk management and to present some of the methods that are often used in handling heavy tails. To illustrate, I will use different methods on the example of the NASDAQ index.
Secondary keywords:	heavy-tailed distributions;normal distribution;VaR;value at risk;expected shortfall;EVT;extreme value theory;backtesting;
Type (COBISS):	Master's thesis/paper
Study programme:	0
Thesis comment:	Univ. v Ljubljani, Fak. za matematiko in fiziko, Oddelek za matematiko, Finančna matematika - 2. stopnja
Pages:	59 str.
ID:	10962319