Tesis y Trabajos de Investigación PUCP
URI permanente para esta comunidadhttp://54.81.141.168/handle/123456789/6
El Repositorio Digital de Tesis y Trabajos de Investigación PUCP aporta al Repositorio Institucional con todos sus registros, organizados por grado: Doctorado, Maestría, Licenciatura y Bachillerato. Se actualiza permanentemente con las nuevas tesis y trabajos de investigación sustentados y autorizados, así como también con los que que fueron sustentados años atrás.
Ingresa a su web: Repositorio Digital de Tesis y Trabajos de Investigación PUCP
Explorar
5 resultados
Resultados de búsqueda
Ítem Texto completo enlazado Empirical modelling of latin american stock markets returns and volatility using Markov - Switching garch models(Pontificia Universidad Católica del Perú, 2017-03-09) Ataurima Arellano, Miguel; Rodríguez, GabrielUsing a sample of weekly frequency of the stock markets returns series, we estimate a set of Markov-Switching-Generalized Autoregressive Conditional Heterocedastic- ity (MS-GARCH) models to a set of Latin American countries (Argentina, Brazil, Chile, Colombia, Mexico and Peru) with an approach based on both the Monte Carlo Expectation-Maximization (MCEM) and Monte Carlo Maximum Likelihood (MCML) algorithms suggested by Augustyniak (2014). The estimates are compared with a stan- dard GARCH, MS and other models. The results show that the volatility persistence is captured di¤erently in the MS and MS-GARCH models. The estimated parameters with a standard GARCH model exacerbates the volatility in almost double compared to MS-GARCH model. There is di¤erent behavior of the coe¢ cients and the variance according the two regimes (high and low volatility) by each model in the Latin Amer- ican stock markets. There are common episodes related to global international crises and also domestic events producing the di¤erent behavior in the volatility of each time series.Ítem Texto completo enlazado Estimation of the sovereign yield curve of Peru : the role of macroeconomic and latent factors(Pontificia Universidad Católica del Perú, 2017-03-04) Olivares Ríos, Alejandra; Rodríguez, GabrielThe study of the yield curve has been a topic that interested economists for a long time since the term structure of interest rates is an important transmission channel of monetary policy to inflation and real activity. In this paper, following Ang and Piazzesi (2003), we study the relevance of macroeconomic factors on Peruvian sovereign yield curve through an Affine Term Structure model for the period from November 2005 to December 2015. We estimate a Gaussian model to understand the joint dynamics of macro variables -inflation and real activity factors- and Peruvian bond yields in a multifactor model of the term structure. Risk premia are modeled as time varying and depend on both observable and unobservable factors. A Vector Autoregressive (VAR) model is estimated considering no-arbitrage assumptions, which let us to derive Impulse Response Functions and Variance Decompositions. We find evidence that macro factors help to improve the fit of the model and explain a substantial amount of variation in bond yields. Variance decompositions show that macro factors explain a significant amount of the movements in the short and middle segments of the yield curve (up to 50%) while unobservable factors are the main drivers for most of the movements at the long end of the yield curve (up to 80%). Furthermore, we find that setting no-arbitrage restrictions improve the forecasting performance of a VAR and that models that include macro factors forecast better than models with only unobservable components.Ítem Texto completo enlazado Aplicación de un modelo de corta memoria con cambios de nivel aleatorios a la volatilidad del mercado bursátil de Latinoamérica(Pontificia Universidad Católica del Perú, 2017-03-04) Tramontana Tocto, Roxana; Rodríguez, GabrielLa evidencia empírica indica que la volatilidad de la serie de los retornos bursátiles tiene larga memoria. Sin embargo, se ha demostrado que los procesos de corta memoria contaminados con cambios de nivel aleatorios pueden ser confundidos con larga memoria. Frecuentemente esta característica es conocida como larga memoria espúrea. Este trabajo presenta un estudio empírico del modelo de cambios de nivel aleatorios, Random Level Shift (RLS) usando la aproximación de Lu y Perron (2010) y Li y Perron (2013) para la volatilidad de los retornos bursátiles diarios para cinco países latinoamericanos. El modelo RLS consiste en la suma de un componente de corta memoria y un componente de cambio de nivel, donde este último es gobernado por un proceso Bernoulli con probabilidad de cambio a. Los resultados obtenidos sugieren que los cambios de nivel de la volatilidad de los retornos diarios son poco frecuentes pero una vez que se toman en cuenta, la característica de larga memoria y los efectos GARCH desaparecen. También se realizó un ejercicio de predicción fuera de la muestra.Ítem Texto completo enlazado Explaining the determinants of the frequency rate interventions in Peru using count models(Pontificia Universidad Católica del Perú, 2017-03-04) Ventura Neyra, Edgar; Rodríguez, GabrielLa presente investigación analiza los determinantes de la frecuencia de las intervenciones de tipo de cambio por parte del Banco Central de Reserva del Perú (BCRP). Esto a partir de información semanal entre enero de 2001 y diciembre de 2010, usando modelos de conteo como Poisson, Negativo Binomial y Zero Inflated. Los resultados muestran que las desviaciones del logaritmo del tipo de cambio respecto de su tendencia de largo plazo, las intervenciones del periodo anterior (persistencia), el riesgo país medido por el EMBIG, el spread entre las tasas de interés bancarias, y el interés entre las tasas de interés doméstica y foránea son importantes determinantes.Ítem Texto completo enlazado Duration models and value at risk using high-frequency data for the peruvian stock market(Pontificia Universidad Católica del Perú, 2017-02-20) Téllez De Vettori, Giannio; Najarro Chuchón, Ricardo; Rodríguez, GabrielMost empirical studies in nance use data on a daily basis which is obtained by retaining the last observation of the day and ignoring all intraday records. However, as a result of the increased automatization of nancial markets and the evolution of computational trading systems, intraday data bases that record every transaction along with their characteristics have been stablished. These data sets prompted the development of a new area of research ( nance with high frequency data), and in 1980 a literature based on the mechanisms of trading began (forms of trading, rules on securities trading, market structure, etc.), originating the Theory of Market Microstructure for the valuation of nancial assets, whose models advocate that timing transmits information. Then the literature proposed an extension to risk management by calculating the implied volatility, which is estimated by the realized volatility on an intraday level, and its applications for a ner value at risk (VaR).