Forecast combination has become popular in central banks as a means to improve forecasts and to alleviate the risk of selecting poor models. However, if a model suite is populated with many similar models, then the weight attached to other independent models may be lower than warranted by their...

Forecast combination has become popular in central banks as a means to improve forecasts and to alleviate the risk of selecting poor models. However, if a model suite is populated with many similar models, then the weight attached to other independent models may be lower than warranted by their...

One of the main problems in modelling multivariate conditional covariance time series is the parameterization of the correlation structure because, if no constraints are imposed, it implies a large number of unknown coefficients. The most popular models propose parsimonious representations,...

In this paper, a new algorithm for RBF centers initialization for functional approximation problems is proposed. The design of the algorithm is inspired in previous clustering algorithms but adding some features to adapt the algorithm to the characteristics of our problem. We compare the results...

This document provides supplementary results to the analyses of Rose (2011), “Geographic Variation in Subprime Loan Features, Foreclosures and Prepayments,” which examines the geographic variation in the effects of prepayment penalties, balloon loans, and reduced documentation on the...

In this paper I present a time series analysis based on a metrical approach. I use a definition of distance which depends on the sample correlation coefficient among bonds. The dataset consists on daily returns of US treasury bonds. By mean of a Linkage-Algorithm bonds are classified according...

We present cross and time series analysis of price fluctuations in the U.S. Treasury fixed income market. By means of techniques borrowed from statistical physics we show that the correlation among bonds depends strongly on the maturity and bonds' price increments do not fulfill the random walk...

We present cross and time series analysis of price fluctuations in the U.S. Treasury fixed income market. Bonds have been classified according to a suitable metric based on the correlation among them. The classification shows how the correlation among fixed income securities depends strongly on...

Nowcasting regards the inference on the present realization of random variables, on the basis of information available until a recent past. This paper proposes a modelling strategy aimed at a best use of the data for nowcasting based on panel data with severe deficiencies, namely short times...