Wyniki wyszukiwania - Biblioteka Nauki

1

Modelling and Forecasting WIG20 Daily Returns

100%

Amado C. , Silvennoinen A. , Teräsvirta T.

|

nr nr 3

173-200

EN

The purpose of this paper is to model daily returns of the WIG20 index. The idea is to consider a model that explicitly takes changes in the amplitude of the clusters of volatility into account. This variation is modelled by a positive-valued deterministic component. A novelty in specification of the model is that the deterministic component is specified before estimating the multiplicative conditional variance component. The resulting model is subjected to misspecification tests and its forecasting performance is compared with that of commonly applied models of conditional heteroskedasticity. (original abstract)

2

Online Testing of Switching Volatility

100%

Bock D.

|

2006

|

nr nr 2 Financial markets : principles of modeling forecasting and decision-making

99-120

EN

Chapter 6 presents the methods for online detection of a change in the unconditional volatility. A test based on a moving sum is proposed and evaluated. The test has a controlled asymptotic size, i.e. the false alarm probability during an infinitely long monitoring period is fixed. The effects of autoregression and conditional heteroscedasticity have been also addressed and a test that allows for heteroscedasticity has been proposed. The testing procedures are exemplified with the Hang Seng Index to see whether a shift during the Asian crises period could have been detected online (fragment of text)

3

Bayesian Analysis of Stochastic Volatility Model and Portfolio Allocation

84%

Pajor A.

|

tom 192 Issues in Modeling, Forecasting and Decision-making in Financial Markets

229-249

XX

W artykule przedstawiono model zmienności stochastycznej, oparty na dekompozycji Choleskiego. Następnie model SV oraz podejście Bayesowskie zostało wykorzystane do modelowania zmienności dwuwymiarowych finansowych szeregów czasowych oraz budowy optymalnego portfela walutowego. Rozważono hipotetyczny portfel, w skład którego wchodzą złotówkowe kursy dwóch walut: dolara amerykańskiego i marki niemieckiej. W procesie optymalizacji portfela wykorzystano predyktywny rozkład stóp zwrotu oraz predyktywny rozkład macierzy warunkowych kowariancji, uzyskany w rozważanym modelu MSV za pomocą metod Monte Carlo (MCMC). (abstrakt oryginalny)

EN

In this paper we present the multivariate stochastic volatility model based on the Cholesky decomposition. This model and the Bayesian approach is used to model bivariate daily financial time series and construct an optimal portfolio. We consider the hypothetical portfolios consisted of two currencies that were most important for the Polish economy: the US dollar and the German mark. In the optimization process we used the predictive distributions of future returns and the predictive covariance matrix obtained from the MSV model. (original abstract)

4

Volatile ARMA Modelling of GARCH Squares

84%

Lawrance A. J.

|

2010

|

tom 2

|

nr nr 3

195-203

EN

This paper points out that the ARMA models followed by GARCH squares are volatile and gives explicit and general forms of their dependent and volatile innovations. The volatility function of the ARMA innovations is shown to be the square of the corresponding GARCH volatility function. The prediction of GARCH squares is facilitated by the ARMA structure and predictive intervals are considered. Further, the developments suggest families of volatile ARMA processes. (original abstract)

5

Bayesian Value-at-Risk for a Portfolio : Multi- and Univariate Approaches Using MSF-SBEKK Models

84%

Osiewalski J. , Pajor A.

|

nr nr 4

253-277

EN

The s-period ahead Value-at-Risk (VaR) for a portfolio of dimension n is considered and its Bayesian analysis is discussed. The VaR assessment can be based either on the n-variate predictive distribution of future returns on individual assets, or on the univariate Bayesian model for the portfolio value (or the return on portfolio). In both cases Bayesian VaR takes into account parameter uncertainty and non-linear relationship between ordinary and logarithmic returns. In the case of a large portfolio, the applicability of the n-variate approach to Bayesian VaR depends on the form of the statistical model for asset prices. We use the n-variate type I MSF-SBEKK(1,1) volatility model proposed specially to cope with large n. We compare empirical results obtained using this multivariate approach and the much simpler univariate approach based on modelling volatility of the value of a given portfolio. (original abstract)

6

Markov Switching In-Mean Effect : Bayesian Analysis in Stochastic Volatility Framework

84%

Kwiatkowski Ł.

|

nr nr 1

59-94

EN

In the study we introduce an extension to a stochastic volatility in mean model (SV-M), allowing for discrete regime switches in the risk premium parameter. The logic behind the idea is that neglecting a possibly regime- changing nature of the relation between the current volatility (conditional standard deviation) and asset return within an ordinary SV-M specification may lead to spurious insignificance of the risk premium parameter (as being 'averaged out' over the regimes). Therefore, we allow the volatility-in-mean effect to switch over different regimes according to a discrete homogeneous two-state Markov chain. We treat the new specification within the Bayesian framework, which allows to fully account for the uncertainty of model parameters, latent conditional variances and hidden Markov chain state variables. Standard Markov Chain Monte Carlo methods, including the Gibbs sampler and the Metropolis-Hastings algorithm, are adapted to estimate the model and to obtain predictive densities of selected quantities. Presented methodology is applied to analyse series of the Warsaw Stock Exchange index (WIG) and its sectoral subindices. Although rare, once spotted the switching in-mean effect substantially enhances the model fit to the data, as measured by the value of the marginal data density. (original abstract)

7

wpływ wartości ekstremalnych na zmienność stochastyczną

84%

Majewska J.

|

nr nr 162 Wielowymiarowe modelowanie i analiza ryzyka

131-143

XX

Celem pracy jest zbadanie zasadności stosowania modeli stochastycznych uwzględniających obserwacje ekstremalne na polskim rynku kapitałowym. Oprócz podstawowego modelu SV w analizie porównawczej zostały uwzględnione modele, których konstrukcja pozwala lepiej opisywać pojawianie się obserwacji ekstremalnych. (fragment tekstu)

EN

This article takes up validity of the use (on the Polish capital market) of stochastic models which take into account extreme observations. In the comparative analysis aside from the SV been considered models whose structure can better describe the appearance of extreme observations (original abstract)

8

Bayesian SVLEDEJ Model for Detecting Jumps in Logarithmic Growth Rates of One Month Forward Gas Contract Prices

84%

Kostrzewski M.

|

tom 8

|

nr nr 3

161-179

EN

The aim of this paper is to analyse the welfare consequences of the processes of liberalisation of trade between asymmetric states in terms of the various sizes and effectiveness of their economies and the type of international exchange. These characteristics ultimately define the distribution of benefits from the liberalisation of international trade. When it is inter-industry or vertical intra-industry and barriers in trade are smaller than the difference in the effectiveness of the economies, the trade liberalisation undoubtedly contributes to improved social welfare, regardless of the level of effectiveness and the size of the economy. In the situation, however, of horizontal intra-industry trade, changes in the welfares of asymmetric countries, caused by their progressing trade liberalisation, depend on the sizes and effectiveness of their economies. The welfare of society in either a very big and ineffective or in a small and very ineffective country could even decrease in such a situation. This is the case when the increase in consumers' surplus is not sufficient to compensate for the decreasing profits of firms. (original abstract)

9

Application of Iterated Filtering to Stochastic Volatility Models Based on Non-Gaussian Ornstein-Uhlenbeck Process

84%

Szczepocki P.

|

tom 21

|

nr nr 2

173-187

EN

Barndorff-Nielsen and Shephard (2001) proposed a class of stochastic volatility models in which the volatility follows the Ornstein-Uhlenbeck process driven by a positive Levy process without the Gaussian component. The parameter estimation of these models is challenging because the likelihood function is not available in a closed-form expression. A large number of estimation techniques have been proposed, mainly based on Bayesian inference. The main aim of the paper is to present an application of iterated filtering for parameter estimation of such models. Iterated filtering is a method for maximum likelihood inference based on a series of filtering operations, which provide a sequence of parameter estimates that converges to the maximum likelihood estimate. An application to S&P500 index data shows the model perform well and diagnostic plots for iterated filtering ensure convergence iterated filtering to maximum likelihood estimates. Empirical application is accompanied by a simulation study that confirms the validity of the approach in the case of Barndorff-Nielsen and Shephard's stochastic volatility models. (original abstract)

10

Stochastyczne problemy przywozu

84%

Sikora W.

|

nr nr 184 Prace Naukowe Instytutu Cybernetyki Ekonomicznej

81-91

XX

Zagadnienie przywozu podobnie jak symetryczne zagadnienie rozwozu można traktować, jako pewne uogólnienie zagadnienia m - komiwojażerów. Zagadnienie to ma szereg sformułowań i istnieją różne metody jego rozwiązywania. Nas w pracy będą interesowały stochastyczne wersje zagadnienia przywozu. Uzyskujemy je przyjmując, że pewne parametry problemu deterministycznego stają się obecnie zmiennymi losowymi. Dwa podstawowe podejścia do rozwiązywania liniowych zadań programowania stochastycznego to metoda Charnesa-Coopera i metoda Dantziga-Madansky'ego. Obydwa podejścia możemy także stosować do rozwiązywania stochastycznych problemów przywozu. Pokażemy, jakie stochastyczne problemy powstają, jeżeli przyjmujemy losowość pewnych parametrów, oraz przy jakich dodatkowych założeniach możemy w sposób naturalny stosować te dwa podejścia do sformułowania liniowych i deterministycznych odpowiedników poszczególnych problemów stochastycznych. (fragment tekstu)

11

Dynamic Linkages in the Pairs (GBP/EUR, USD/EUR) and (GBP/USD, EUR/USD): How Do They Change During a Day?

84%

Doman M. , Doman R.

|

nr nr 1

33-56

EN

In the paper, we document how conditional dependencies observed in the FOREX market change during a trading day. The analysis is performed for the pairs (GBP/EUR, USD/EUR) and (GBP/USD, EUR/USD) of exchange rates. We consider daily returns calculated using the exchange rates quoted at different hours of a day. The dynamics of the dependencies is modeled by means of 3-regime Markov regime switching copula models, and the strength of the linkages is described using dynamic Spearman's rho and the dynamic coefficients of tail dependence. The established approach allows us to monitor the changes in the dependence structure. (original abstract)

12

Hybrid MSV-MGARCH Models - General Remarks and the GMSF-SBEKK Specification

84%

Osiewalski J. , Osiewalski K.

|

nr nr 4

241-271

EN

The first so-called hybrid MSV-MGARCH models were characterized by the conditional covariance matrix that was a product of a univariate latent process and a matrix with a simple MGARCH structure (Engle's DCC or scalar BEKK). The aim was to parsimoniously describe volatility of a large group of assets. The proposed hybrid models, similarly as pure MSV specifications(and other models based on latent processes), required the Bayesian approach equipped with efficient MCMC simulation tools. The numerical effort has payed - the hybrid models seem particularly useful due to their good fit and ability to jointly cope with large portfolios. In particular, the simplest hybrid, now called the MSF-SBEKK model, has been successfully used in many applications. However, one latent process may be insufficient in the case of a highly heterogeneous portfolio. Thus, in this study we discuss a general hybrid MSV-MGARCH model structure, showing its basic characteristics that explain greater flexibility of such hybrid structure with respect to the corresponding MGARCH class. From the empirical perspective, we advocate the GMSF-SBEKK specification, which uses as many latent processes as there are relatively homogeneous groups of assets. We present full Bayesian inference for such models, with the use of an efficient MCMC simulation strategy. The approach is used to jointly model volatility on very different markets. Joint modelling is formally compared to individual modelling of volatility on each market. (original abstract)

13

Bayesowska wycena opcji na index WIG20 : procesy Itô a dyskretne procesy SV

84%

Kostrzewski M. , Pajor A.

|

2005-2006

|

tom 46-47

65-85

XX

Autorzy prezentują rezultaty zastosowania ciągłego uogólnionego procesu Ornsteina-Uhlenbecka (Extended Ornstein-Uhlenbeck process, w skrócie EOU) będącego rozwiązaniem niejednorodnego stochastycznego równania różniczkowego Itô oraz dyskretnego procesu zmienności (wariancji) stochastycznej z korelacją i grubymi ogonami (Fat-tailed and Correlated SV, w skrócie FCSV) do wyceny europejskich opcji kupna na indeks WIG20. Wykorzystywane modele różnią się nie tylko założeniem dotyczącym procesu generowania danych finansowych, ale również paradygmatem związanym z ciągłością procesu powstawania obserwacji. Autorzy podjęli (po raz pierwszy w Polsce) problem bayesowskiej wyceny opcji z wykorzystaniem ciągłego procesu EOU oraz porównania uzyskanych w ten sposób wyników z wynikami otrzymanymi w oparciu o model dyskretny FCSV.

EN

In this paper the Extended Ornstein-Uhlenbeck process (EOU) and the Fat-tailed and Correlated Stochastic Volatility process (FCSV) are used in Bayesian pricing of European call options. The main difference between EOU and FCSV models is that the first of them is a continuous process whilst the second one is a discrete process. The Bayesian option pricing is based on the distribution of the payoff function given by the predictive density of future observables. The basic instrument was WIG20 index. The most surprising conclusion of our research is a similarity of forecast and option pricing based on different Bayesian models. The results presented in the paper are obtained by Monte Carlo Markov chain. The Metropolis-Hastings algorithm and the acceptance-rejection sampling are used within the Gibbs sampler. (original abstract)

14

Modeling the Realized Volatility with ARFIMA and Unobserved Component Models: Results from the Polish Financial Market

84%

Doman M.

|

2006

|

nr nr 2 Financial markets : principles of modeling forecasting and decision-making

121-135

EN

Chapter 7 is focused on the notion of realized volatility in financial econometrics. The chapter presents an approach to the estimation of the daily realized volatility based on intraday returns. It also takes into account effects of market microstructure. The volatility has been modeled and predicted for stock index WIG20 and exchange rate USD/PLN using ARFIMA and unobserved component models. The findings are that modeling realized volatility with UC and ARFIMA models provides comparable volatility forecasts. (fragment of text)

15

VECM-TSV Models for Two Polish Official Exchange Rates

84%

Pajor A.

|

2006

|

nr nr 2 Financial markets : principles of modeling forecasting and decision-making

49-66

EN

Chapter 3 presents a modeling of EUR/USD PLN/USD and PLN/EUR exchange rates under cointegration and error correction in VAR-TSV model. It has been shown that EUR/USD exchange rate and ECM term have effects on the inference on the conditional variances and correlation coefficient. (fragment of text)

16

Bayesowska estymacja i prognozowanie w modelu Stochastycznej zmienności z normalnym błędem obserwacji

84%

Pajor A.

|

tom Postępy ekonometrii

239-254

XX

W tej pracy wykorzystamy podejście bayesowskie do estymacji i prognozy procesu SV, zaproponowane przez Jacquier, Polson i Rossi. Aby wyznaczyć podstawowe charakterystyki rozkładu a posteriori parametrów modelu, wykorzystamy metodę MCMC (Markov - chain Monte Carlo). Dalej przypominamy podstawowe definicje i oznaczenia związane z procesem SV z normalnym błędem obserwacji. Następnie omówimy bayesowską estymację i predykcję naszego procesu i wykorzystamy przedstawione podejście do modelowania logarytmicznych stóp zmian indeksu WIG. (fragment tekstu)

17

Bayesian Comparison of Bivariate SV Models for Two Related Time Series

67%

Pajor A.

|

2005

|

tom 190 Macromodels 2004 : Problems of Building and Estimation of Econometic Models

177-196

XX

W pracy przedstawiono bayesowskie porównanie czterech różnych specyfikacji modeli zmienności stochastycznej (stochastic volatility, SV) dla dwóch kursów walutowych (PLN/USD i PLN/DEM, 6.02.1996-31.12.2001). Rozważono cztery modele o warunkowym dwuwymiarowym rozkładzie normalnym: SDF, BSV, JSV i TSV. Specyfikacje te różnią się liczbą procesów ukrytych opisujących trzy różne elementy macierzy warunkowych kowariancji oraz założeniami dotyczącymi warunkowego współczynnika korelacji. W modelu SDF występuje jeden proces ukryty, przez co warunkowy współczynnik korelacji jest stały. W model BSV dynamika wariancji warunkowych opisana jest za pomocą dwóch procesów ukrytych, ale warunkowe kowariancje są równe zero. Natomiast w modelach JSV i TSV warunkowy współczynnik korelacji jest zmienny. W modelu TSV macierz warunkowych kowariancji opisana jest za pomocą trzech procesów ukrytych, podczas gdy w modelu JSV występująjedynie dwa procesy ukryte. Porównanie modeli oparte jest na prawdopodobieństwach a posteriori rozważanych modeli, a w szczególności na czynnikach Bayesa (Bayes factors). Wyniki empiryczne wskazują na zdecydowaną przewagę modelu TSV, w którym do opisu dynamiki zarówno warunkowych wariancji jak i kowariancji wykorzystuje się trzy procesy ukryte. Natomiast modele zakładające stały lub zerowy warunkowy współczynnik korelacji (tj. SDF i BSV) są silnie odrzucane przez dane (prawdopodobieństwa a posteriori tych modeli są znacznie niższe niż modeli o zmiennym warunkowym współczynniku korelacji). (abstrakt oryginalny)

EN

In the paper two series of daily growth rates (of PLN/USD and PLN/DEM, 6.02.1996 - 31.12.2001) are used to compare four different specifications of multivariate stochastic volatility models. We consider the following bivariate SV models: the Basic Stochastic Volatility model (BSV), the Stochastic Discount Factor model (SDF), the SV model with the Cholesky decomposition of the conditional covariance matrix (TSV), and the SV model with the spectral decomposition (JSV). The Stochastic Volatility models differ in assumption on conditional correlation and in the number of latent processes. The BSV model uses two separate univariate SV processes to describe the conditional variances, assuming zero conditional covariance. The SDF assumes that the conditional covariance is stochastic, but the conditional correlation among the series is constant over time - the dynamics of the conditional variances and covariance is described by one latent process. The TSV and JSV models assume that the conditional correlation is time-varying and stochastic. In the TSV model the conditional covariance matrix is modelled using three separate stochastic processes, while in the JSV model there are only two latent processes. (original abstract)