Results 11 - 20
of
22
Publications
"... Copyright © b y Georgios N. B a n a \ ^ 2000 This copy of the thesis has been supplied on condition that anyone who consults it is understood to recognise that its copyright rests with its author and that no quotation from the thesis and no information derived from it may be published without the a ..."
Abstract
- Add to MetaCart
(Show Context)
Copyright © b y Georgios N. B a n a \ ^ 2000 This copy of the thesis has been supplied on condition that anyone who consults it is understood to recognise that its copyright rests with its author and that no quotation from the thesis and no information derived from it may be published without the author's prior consent. To my parents, Maria and Nikolaos Banavas Author's declaration At no time during the registration for the degree of Doctor of Philosophy has the author been registered for any other University award. This study was financed by the Higher Education Funding Council of England ( H E F C E Qr). A programme of advanced study was undertaken, including a postgraduate course in fi-nancial modelling (MSc Computational Intelligence- C0IN511). Relevant scientific seminars and conferences were regularly attended at which work was often presented; external institutions were visited for consultation purposes and several papers prepared for publication.
Generalized Neutral Portfolio
, 2009
"... An asset allocation framework decomposes the universe of asset returns into factors either by Fundamental Analysis, Factor Analysis or Principal Component Analysis. Asset allocation methods then attempt to neutralize the sensitivity of the portfolio to a select factor. However, Factor Analysis or PC ..."
Abstract
- Add to MetaCart
(Show Context)
An asset allocation framework decomposes the universe of asset returns into factors either by Fundamental Analysis, Factor Analysis or Principal Component Analysis. Asset allocation methods then attempt to neutralize the sensitivity of the portfolio to a select factor. However, Factor Analysis or PCA fail to take into account moments higher than the second moment. This paper outlines a method based on Independent Component analysis to effectively neutralize a portfolio to a component. This method is illustrated for building a market neutral portfolio in the Mean variance framework, and can be extended to encompass any factor(s). 1
Evolutionary Multiobjective Optimization for Selecting Members of an Ensemble Streamflow Forecasting Model
"... Chaire de recherche EDS en prévisions et actions hydrologiques, Dép. de génie civil et de génie des eaux ..."
Abstract
- Add to MetaCart
(Show Context)
Chaire de recherche EDS en prévisions et actions hydrologiques, Dép. de génie civil et de génie des eaux
(wileyonlinelibrary.com) DOI: 10.1002/jae.1112 DEFAULT PRIORS AND PREDICTIVE PERFORMANCE IN BAYESIAN MODEL AVERAGING, WITH APPLICATION TO GROWTH DETERMINANTS †
"... Bayesian model averaging (BMA) has become widely accepted as a way of accounting for model uncertainty, notably in regression models for identifying the determinants of economic growth. To implement BMA the user must specify a prior distribution in two parts: a prior for the regression parameters an ..."
Abstract
- Add to MetaCart
Bayesian model averaging (BMA) has become widely accepted as a way of accounting for model uncertainty, notably in regression models for identifying the determinants of economic growth. To implement BMA the user must specify a prior distribution in two parts: a prior for the regression parameters and a prior over the model space. Here we address the issue of which default prior to use for BMA in linear regression. We compare 12 candidate parameter priors: the unit information prior (UIP) corresponding to the BIC or Schwarz approximation to the integrated likelihood, a proper data-dependent prior, and 10 priors considered by Fernández et al. (Journal of Econometrics 2001; 100: 381–427). We also compare two model priors: the uniform model prior and a prior with prior expected model size 7. We compare them on the basis of crossvalidated predictive performance on a well-known growth dataset and on two simulated examples from the literature. We found that the UIP with uniform model prior generally outperformed the other priors considered. It also identified the largest set of growth determinants. Copyright © 2009 John Wiley & Sons, Ltd. 1.
Journal of Forecasting J. Forecast. 23, 541–557 (2004) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/for.930 Comparing the Accuracy of Density Forecasts from Competing Models
"... A rapidly growing literature emphasizes the importance of evaluating the forecast accuracy of empirical models on the basis of density (as opposed to point) forecasting performance. We propose a test statistic for the null hypothesis that two competing models have equal density forecast accuracy. Mo ..."
Abstract
- Add to MetaCart
A rapidly growing literature emphasizes the importance of evaluating the forecast accuracy of empirical models on the basis of density (as opposed to point) forecasting performance. We propose a test statistic for the null hypothesis that two competing models have equal density forecast accuracy. Monte Carlo simulations suggest that the test, which has a known limiting distribution, displays satisfactory size and power properties. The use of the test is illustrated with an application to exchange rate forecasting. Copyright © 2004 John Wiley & Sons, Ltd. key words forecasting; forecast evaluation; density forecast; exchange rates
Acknowledgements
, 1994
"... I had the great pleasure of working with Dr. Anamitra Makur during my stay at the Institute. But for him, this thesis would not have seen light. I would like to thank him for the help he has rendered during this work. I was fortunate to have had Gaggan, Mouli, Suba, Selvi, Doss and Vinay as labmates ..."
Abstract
- Add to MetaCart
I had the great pleasure of working with Dr. Anamitra Makur during my stay at the Institute. But for him, this thesis would not have seen light. I would like to thank him for the help he has rendered during this work. I was fortunate to have had Gaggan, Mouli, Suba, Selvi, Doss and Vinay as labmates who had been excellent people to work with. I would like to thank Sitaram, Vasuki, Ganesh and Suryan for their help. I would also like to thank Venn, Gopal, Snajay, Kathy, Delay, Pattar, Babu, Manoj, Jyothish and Jemlin for making my stay at the Institute a happy and memorable one. I would also like to thank Ramu, Pozhan, Eapen, Uncle, Vappan, Vaman, Anil and Sunil, Prahh and Mala who had been always helpful to me. I dso remember many a afternoon (Saturday special) sessions with Babu, Kannan, Padi-yar, Madhu, Jeby, Joy and Ajith. Finally I would like to thank my parents for their constant encouragement throughout the course.
excellent software programming, Tilmann Gneiting for kindly sharing his CPRS code for BMA applications, and
, 2007
"... France, for hospitality during the preparation of this paper. The views expressed in this study are the sole responsibility of the authors and should not be attributed to the International Monetary Fund, its Executive Board, or its management. Economic growth has been a showcase of model uncertainty ..."
Abstract
- Add to MetaCart
(Show Context)
France, for hospitality during the preparation of this paper. The views expressed in this study are the sole responsibility of the authors and should not be attributed to the International Monetary Fund, its Executive Board, or its management. Economic growth has been a showcase of model uncertainty, given the many competing theories and candidate regressors that have been proposed to explain growth. Bayesian Model Averaging (BMA) addresses model uncertainty as part of the empirical strategy, but its implementation is subject to the choice of priors: the priors for the parameters in each model, and the prior over the model space. For a well-known growth dataset, we show that model choice can be sensitive to the
A Comparison of Bayesian and Conditional Density Models in Probabilistic Ozone Forecasting
"... Abstract — Probabilistic models were developed to provide predictive distributions of daily maximum surface level ozone concentrations. Five forecast models were compared at two stations (Chilliwack and Surrey) in the Lower Fraser Valley of British Columbia, Canada, with local meteorological variabl ..."
Abstract
- Add to MetaCart
(Show Context)
Abstract — Probabilistic models were developed to provide predictive distributions of daily maximum surface level ozone concentrations. Five forecast models were compared at two stations (Chilliwack and Surrey) in the Lower Fraser Valley of British Columbia, Canada, with local meteorological variables used as predictors. The models were of two types, conditional density models and Bayesian models. The Bayesian models (especially the Gaussian Processes) gave better forecasts for extreme events, namely poor air quality events defined as having ozone concentration ≥ 82 ppb. I.
