# Dr Federico Martellosio

### Biography

### Biography

Federico Martellosio graduated in Management, Economics and Industrial Engineering from the Polytechnic of Milan in 1999. He completed his MSc degree in Economics at the University of Southampton in 2001, and received his PhD in Economics from the same institution in 2006. From 2004 to 2005 he was a teaching fellow at the University of Southampton. He then spent a year at the University of Amsterdam as a PostDoctoral Fellow. In 2006 he moved to the University of Reading as a Lecturer in Economics, and in 2012 he joined the School of Economics at Surrey.

Personal website: https://sites.google.com/site/federicomartellosio/

### Research interests

Theoretical and applied econometrics, spatial econometrics, multivariate statistics.

### Departmental duties

PhD Programme Director.

### My teaching

ECOD022 Topics in Econometrics (PhD)

ECO2047 Introductory Econometrics (UG)

ECO1005 Mathematics for Economics (UG)

### My publications

### Publications

focussing on estimation and inference for the spatial autoregressive parameter ». The

quasi-maximum likelihood estimator for » usually cannot be written in closed form, but

using an exact result obtained earlier by the authors for its distribution function, we are

able to provide a complete analysis of the properties of the estimator, and exact inference

that can be based on it, in models that are balanced. This is presented first for the so-called

pure model, with no regression component, but is also extended to some special cases of

the more general model. We then study the much more difficult case of unbalanced models,

giving analogues of some, but by no means all, of the results obtained for the balanced

case earlier. In both balanced and unbalanced models, results obtained for the pure model

generalize immediately to the model with group-specific regression components.

autoregressive models, with applications to inference.,Journal of Econometrics 205 (2) pp. 402-422 Elsevier

autoregression usually cannot be written explicitly in terms of the data. A rigorous analysis

of the first-order asymptotic properties of the estimator, under some assumptions

on the evolution of the spatial design matrix, is available in Lee (2004), but very little

is known about its exact or higher-order properties. In this paper we first show that the

exact cumulative distribution function of the estimator can, under mild assumptions,

be written in terms of that of a particular quadratic form. Simple examples are used

to illustrate important exact properties of the estimator that follow from this representation.

In general models a complete exact analysis is not possible, but a higher-order

(saddlepoint) approximation is made available by the main result. We use this approximation

to construct confidence intervals for the autoregressive parameter. Coverage

properties of the proposed confidence intervals are studied by Monte Carlo simulation,

and are found to be excellent in a variety of circumstances

on maximum likelihood estimation of a parameter of interest is to recenter the profile score

for that parameter. We apply this general principle to the quasi-maximum likelihood estimator (QMLE) of the autoregressive parameter » in a spatial autoregression. The resulting estimator for » has better finite sample properties compared to the QMLE for », especially in the presence of a large number of covariates. It can also solve the incidental parameterproblem that arises, for example, in social interaction models with network fixed effects, or infspatial panel models with individual or time fixed effects. However, spatial autoregressions present specific challenges for this type of adjustment, because recentering the profile score may cause the adjusted estimate to be outside the usual parameter space for ». Conditions for this to happen are given, and implications are discussed. For inference, we propose confidence intervals based on a Lugannani{Rice approximation to the distribution of the adjusted QMLE of ». Based on our simulations, the coverage properties of these intervals are excellent even in models with a large number of covariates.

The first chapter, investigates the forecast ability of Random Forest, Functional Non-parametric and Dynamic Nelson-siegel models. Results of this study indicate the superiority of the Random Forest model in forecasting short end of the yield curve, and Dynamic Nelson-Siegel model in predicting yields of bonds with long term to maturity. In line with the literature, results recommend the employment of external source of information such as macroeconomic variables.

The second chapter examines the relative ability of Generalized Autoregressive Conditional Heteroskedastic (GARCH) models in volatility prediction. The existing literature produce conflicting results regarding this topic, which is mainly attributed to the choice of loss functions in forecast evaluation in this field. This chapter exploits the recent Robust Forecast Comparison (RFC) test of Jin et al. (2017) to evaluate the out-of-sample forecast performance of 11 GARCH type models. The advantage of using the test of Jin et al. (2017) is that no predetermined form of loss function is required. This study concludes that, on average, all competing models over-predict the volatility, and that the simple GARCH model is outperformed in predicting the volatility.