Showing papers by "Serena Ng published in 2017"

Shock Restricted Structural Vector-Autoregressions

Abstract: It is well known that the covariance structure of the data alone is not enough to identify an SVAR, and the conventional approach is to impose restrictions on the parameters of the model based on a priori theoretical considerations. This paper suggests that much can be gained by requiring the properties of the identified shocks to agree with major economic events that have been realized. We first show that even without additional restrictions, the data alone are often quite informative about the quantitatively important shocks that have occurred in the sample. We propose shrinking the set of solutions by imposing two types of inequality constraints on the shocks. The first restricts the sign and possibly magnitude of the shocks during unusual episodes in history. The second restricts the correlation between the shocks and variables external to the SVAR. The methodology provides a way to assess the validity of assumptions imposed as equality constraints. The effectiveness and limitations of this approach are exemplified with three applications.

Opportunities and Challenges: Lessons from Analyzing Terabytes of Scanner Data

Abstract: This paper seeks to better understand what makes big data analysis different, what we can and cannot do with existing econometric tools, and what issues need to be dealt with in order to work with the data efficiently. As a case study, I set out to extract any business cycle information that might exist in four terabytes of weekly scanner data. The main challenge is to handle the volume, variety, and characteristics of the data within the constraints of our computing environment. Scalable and efficient algorithms are available to ease the computation burden, but they often have unknown statistical properties and are not designed for the purpose of efficient estimation or optimal inference. As well, economic data have unique characteristics that generic algorithms may not accommodate. There is a need for computationally efficient econometric methods as big data is likely here to stay. INTRODUCTION The goal of a researcher is often to extract signals from the data, and without data, no theory can be validated or falsified. Fortunately, we live in a digital age that has an abundance of data. According to the website Wikibon (www.wikibon.org), there are some 2.7 zetabytes of data in the digital universe. The US Library of Congress collected 235 terabytes of data as of 2011. Facebook alone stores and analyzes over 30 petabytes of user-generated data. Google processed 20 petabytes of data daily back in 2008, and undoubtedly much more are being processed now. Walmart handles more than one million customer transactions per hour. Data from financial markets are available at ticks of a second. We now have biometrics data on finger prints, handwriting, medical images, and last but not least, genes. The 1000 Genomes project stored 464 terabytes of data in 2013 and the size of the database is still growing. Even if these numbers are a bit off, there is lot of information out there to be mined. The data can potentially lead economists to a better understanding of consumer and firm behavior, as well as the design and functioning of markets.

Principal Components and Regularized Estimation of Factor Models

Abstract: It is known that the common factors in a large panel of data can be consistently estimated by the method of principal components, and principal components can be constructed by iterative least squares regressions. Replacing least squares with ridge regressions turns out to have the effect of shrinking the singular values of the common component and possibly reducing its rank. The method is used in the machine learning literature to recover low-rank matrices. We study the procedure from the perspective of estimating a minimum-rank approximate factor model. We show that the constrained factor estimates are biased but can be more efficient in terms of mean-squared errors. Rank consideration suggests a data-dependent penalty for selecting the number of factors. The new criterion is more conservative in cases when the nominal number of factors is inflated by the presence of weak factors or large measurement noise. The framework is extended to incorporate a priori linear constraints on the loadings. We provide asymptotic results that can be used to test economic hypotheses.

Opportunities and Challenges: Lessons from Analyzing Terabytes of Scanner Data

Abstract: This paper seeks to better understand what makes big data analysis different, what we can and cannot do with existing econometric tools, and what issues need to be dealt with in order to work with the data efficiently. As a case study, I set out to extract any business cycle information that might exist in four terabytes of weekly scanner data. The main challenge is to handle the volume, variety, and characteristics of the data within the constraints of our computing environment. Scalable and efficient algorithms are available to ease the computation burden, but they often have unknown statistical properties and are not designed for the purpose of efficient estimation or optimal inference. As well, economic data have unique characteristics that generic algorithms may not accommodate. There is a need for computationally efficient econometric methods as big data is likely here to stay.

Shock Restricted Structural Vector-Autoregressions

Abstract: It is well known that the covariance structure of the data alone is not enough to identify an SVAR, and the conventional approach is to impose restrictions on the parameters of the model based on a priori theoretical considerations. This paper suggests that much can be gained by requiring the properties of the identified shocks to agree with major economic events that have been realized. We first show that even without additional restrictions, the data alone are often quite informative about the quantitatively important shocks that have occurred in the sample. We propose shrinking the set of solutions by imposing two types of inequality constraints on the shocks. The first restricts the sign and possibly magnitude of the shocks during unusual episodes in history. The second restricts the correlation between the shocks and variables external to the SVAR. The methodology provides a way to assess the validity of assumptions imposed as equality constraints. The effectiveness and limitations of this approach are exemplified with three applications.

Shock Restricted Structural Vector-Autoregressions

Abstract: Identifying assumptions need to be imposed on autoregressive models before they can be used to analyze the dynamic effects of economically interesting shocks. Often, the assumptions are only rich enough to identify a set of solutions. This paper considers two types of restrictions on the structural shocks that can help reduce the number of plausible solutions. The first is imposed on the sign and magnitude of the shocks during unusual episodes in history. The second restricts the correlation between the shocks and components of variables external to the autoregressive model. These non-linear inequality constraints can be used in conjunction with zero and sign restrictions that are already widely used in the literature. The effectiveness of our constraints are illustrated using two applications of the oil market and Monte Carlo experiments calibrated to study the role of uncertainty shocks in economic fluctuations.