WP

"Detecting Big Structural Breaks in Large Factor Model" (with Liang Cheng and Juan Jose Dolado) (PDF + Extra Material) (December 2011)

A Short Informal Abstract: Time invariance of factor loadings is a standard assumption in the analysis
of large factor models. Yet, this assumption may be restrictive unless
parameter shifts are mild (i.e., local to zero). In this paper we develop a
new testing procedure to detect \textit{big} breaks in these loadings at
either known or unknown dates. It relies upon testing for parameter breaks
in a regression of the first of the $\bar{r}$ \ factors estimated by PCA
on the remaining $\bar{r}-1$ factors, where $\bar{r}$
is chosen according to Bai and Ng's (2002) information criteria. The test
fares well in terms of power relative to other recently proposed tests on
this issue, and can be easily implemented to avoid forecasting failures in
standard factor-augmented (FAR, FAVAR) models where the number of factors is
a priori imposed on the basis of theoretical considerations.

"Summability of Stochastic Processes (A Generalization of Integration and Co-integration valid for
Non-linear Processes)" (with Vanessa Berenguer-Rico) (PDF) (June 2011)

A Short Informal Abstract: The order of integration is valid to characterize linear processes;
but it is not appropriate for non-linear worlds. We propose the concept of summability (a re-scaled partial sum of the
process being Op(1)) to handle non-linearities. The paper shows that this new concept, S (delta): (i)
generalizes I (delta); (ii) measures the degree of persistence as well as of the evolution of the variance;
(iii) controls the balancedness of non-linear relationships; (iv) opens the door to the concept of
co-summability which represents a generalization of co-integration for non-linear processes. To
make this concept empirically applicable, an estimator for delta and its asymptotic properties are
provided. The …nite sample performance of subsampling con…dence intervals is analyzed via a
Monte Carlo experiment. The paper …finishes with the estimation of the degree of summability of
the macroeconomic variables in an extended version of the Nelson-Plosser database.

"Conditional Stochastic Dominance Tests in Dynamic Settings" (with Jose Olmo) (PDF) (October 2010)

A Short Informal Abstract: Variance is a nice measure of uncertainty but is not designed to measure risk. For the latter, better to use
lower partial moments (LPM) and test for Stochastic Dominance. In order for the concept of stochastic dominance to be fully
operational it needs to be exploited dynamically. While there are many influential methods to test the hypothesis of stochastic
dominance in an unconditional or marginal setting, there are just a few methods that aim to do this dynamically or conditionally on an
information set. Moreover, these conditional stochastic dominance tests rely heavily on assuming an appropriate parametric structure
for the dependence between the variables and hence are subject to misspecification issues.
This paper presents a nonparametric test for conditional stochastic dominance that accommodates very easily the presence of dynamics in
the variables without having to impose strong assumptions on the specific form of these dynamics.

"Regime Specific Predictability in Predictive Regressions" (with Jean-Yves Pitarakis)(PDF) (Revised December 2010)

A Short Informal Abstract: Until now the literature has considered only LINEAR Predictive Regression Models. We introduce (modeling and testing) non-linearities of threshold type in these models.
The big advantage is to be able to consider predictability by regimes. In this way we could have predictability in one regime and no predictability in the other.

"SHOCKS (Can we identify them? YES, WE CAN)" (with Oscar Martinez)
A Short Informal Abstract: At time "t" there is a shock e_t. This shock can be big or small, positive or negative, blue or red, may have been produced when the economic is
recession or expansion, when inflation is high or small, when Barcelona soccer team wins or looses, etc. In this paper, via a new Threshold Moving Average Model,
we show how to identify and test if these type of shocks are transitory or permanent.