ResumenEn este trabajo se hace un examen del comportamiento de la proporción de rechazos equivocados de la hipótesis nula (error tipo I) en condiciones plenas de aplicabilidad de la distribución t de Student, es decir, con variables independientes cuya distribución es normal tanto bajo el supuesto de homogeneidad de varianzas como en las condiciones de heterocedasticidad.Palabras clave: estadística t, prueba t, monotonía, estabilidad, prueba de WelchSatterwhaite.
AbstractIn this paper we review the behavior of the type I error rate of Student's t-test and Welch-Sattetthwaite test for comparing two means with independent samples from normal populations under the assumption of homogeneity of variances and under conditions of heteroscedasticity. The results, obtained by the Monte Carlo method show the Welch-Satterthwaite well behaved in all cases.
<p>Clusters of large values are observed in sample paths of certain open-loop threshold autoregressive (TAR) stochastic processes. In order to characterize the stochastic mechanism that generates this empirical stylized fact, three types of marginal conditional distributions of the underlying stochastic process are analyzed in this paper. One allows us to find the conditional variance function that explains the aforementioned stylized fact. As a by-product, we are able to derive a sufficient condition to have asymptotic weak stationarity in an open-loop TAR stochastic process.</p>
ResumenEste artículo pone en perspectiva algunos de los resultados más importantes acerca de un nuevo método de muestreo y estimación para poblaciones donde no existe un marco muestral. Tal método es conocido como RDS, por sus siglas en inglés ((Respondent-Driven Sampling)). Un análisis teórico basado en cadenas de Markov permite mostrar que este método reduce los sesgos generalmente asociados a las muestras por cadenas referenciales, además de producir estimadores asintóticamente insesgados. Se comprobó el potencial del método por medio de una simulación empírica.
Palabras clave: estimador asintóticamente insesgado, población oculta, RDS.
AbstractThis article reviews some of the most important results about a new method of sampling and estimation for populations where there is no sampling frame, this is known as Respondent-Driven Sampling (RDS). A theoretical analysis based on Markov chains shows that this method reduces the bias generally associated with chain-reference samples; in addition the method produces asymptotically unbiased estimators. We demonstrated the potential of the method through an empirical simulation.
ResumenEn este artículo, se evalúa el desempeño de un modelo autorregresivo de umbrales (TAR),en el análisis de series de tiempo financieras. Se utilizan datos del mercado accionario brasilero y norteamericano a fin de ajustar un modelo; además se realiza una comparación con los modelos GARCH vía los momentos condicionales.Palabras clave: heterocedasticidad condicional, modelo TAR, series temporales rinancieras.
AbstractThe performance of TAR models to analyse financial time series is evaluated. Empirically, and using data from the Brasilian stock market, the TAR model is compared with GARCH models via conditional moments.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.