Bayesian Models to Enhance Parameter Estimation of Financial Assets – Proposal and Evaluation of Probabilistic Methods
نویسندگان
چکیده
Assets are often classified according to their risk and expected return. The estimates of these parameters based on classical methods are considered to lead to unnecessary uncertainty since they are based on a limited amount of information. With few observations they generally lead to poor inferences for assets with less data. With a Bayesian methodology other sources of information are allowed to be included in the estimation process. Probability distributions of risk and expected return are derived by pooling information contained in data with information from analogue data sources or even subjective belief. The current state of belief is updated in a way that take individual precisions into account. This thesis present models to update the probability distributions of assets’ parameters by incorporating direct information (through a non-hierarchical model) or information derived through analogue data sources by a hierarchical model. Methods for robust implementation of these models are suggested. It is also discussed how the models can be extended to be better fit outliers. Referat Bayesianska metoder för att förbättra parameterskattningar av finansiella tillgångar – utvärdering och förslag på Bayesianska modeller Finansiella tillgångar klassificeras ofta med deras risk och förväntad avkastning. Det klassiska sättet att skatta dessa parametrar leder ofta till onödigt stor osäkerhet eftersom de baseras på en begränsad mängd information. Med få observationer kan dessa skattningar leda till dåliga slutsatser. Med ett Bayesianskt synsätt kan fler informationkällor inkluderas i skattningsförfarandet. Sannolikhetsfördelningar av risk och förväntad avkastning kan härledas genom att ta hänsyn till likvärdiga datakällor eller kunskap baserad på subjektiv tro. Den aktuella kunskapen om en parameter uppdateras på ett sätt som tar hänsyn till de precisionerna. Den här uppsatsen presenterar modeller för att uppdatera sannolikhetsfördelningarna för parameterar hos tillgångar genom att ta tillvara direkt information (genom en icke-hierarkisk modell) eller information som härleds ur analoga datakällor genom en hierarkisk modell. Metoder för implementation av dessa modeller presenteras. Dessutom diskuteras hur modellerna kan utvecklas för att bättre ta hänsyn till outliers.
منابع مشابه
A Surface Water Evaporation Estimation Model Using Bayesian Belief Networks with an Application to the Persian Gulf
Evaporation phenomena is a effective climate component on water resources management and has special importance in agriculture. In this paper, Bayesian belief networks (BBNs) as a non-linear modeling technique provide an evaporation estimation method under uncertainty. As a case study, we estimated the surface water evaporation of the Persian Gulf and worked with a dataset of observations ...
متن کاملA Surface Water Evaporation Estimation Model Using Bayesian Belief Networks with an Application to the Persian Gulf
Evaporation phenomena is a effective climate component on water resources management and has special importance in agriculture. In this paper, Bayesian belief networks (BBNs) as a non-linear modeling technique provide an evaporation estimation method under uncertainty. As a case study, we estimated the surface water evaporation of the Persian Gulf and worked with a dataset of observations ...
متن کاملBayesian Inference for Spatial Beta Generalized Linear Mixed Models
In some applications, the response variable assumes values in the unit interval. The standard linear regression model is not appropriate for modelling this type of data because the normality assumption is not met. Alternatively, the beta regression model has been introduced to analyze such observations. A beta distribution represents a flexible density family on (0, 1) interval that covers symm...
متن کاملSpeech Enhancement Using Gaussian Mixture Models, Explicit Bayesian Estimation and Wiener Filtering
Gaussian Mixture Models (GMMs) of power spectral densities of speech and noise are used with explicit Bayesian estimations in Wiener filtering of noisy speech. No assumption is made on the nature or stationarity of the noise. No voice activity detection (VAD) or any other means is employed to estimate the input SNR. The GMM mean vectors are used to form sets of over-determined system of equatio...
متن کاملBayesian change point estimation in Poisson-based control charts
Precise identification of the time when a process has changed enables process engineers to search for a potential special cause more effectively. In this paper, we develop change point estimation methods for a Poisson process in a Bayesian framework. We apply Bayesian hierarchical models to formulate the change point where there exists a step < /div> change, a linear trend and a known multip...
متن کامل