This paper develops a new methodology that makes use of the factor structure of large dimensional panels to understand the nature of non-stationarity in the data. We refer to it as PANIC – a ‘Panel Analysis of Non-stationarity in Idiosyncratic and Common components’. PANIC consists of univariate and panel tests with a number of novel features. It can detect whether the nonstationarity is pervasive, or variable-specific, or both. It tests the components of the data instead of the observed series. Inference is therefore more accurate when the components have different orders of integration. PANIC also permits the construction of valid panel tests even when cross-section correlation invalidates pooling of statistics constructed using the observed data. The key to PANIC is consistent estimation of the components even when the regressions are individually spurious. We provide a rigorous theory for estimation and inference. In Monte Carlo simulations, the tests have very good size and power. PANIC is applied to a panel of inflation series.

This paper builds on the structural equations, treatment effect, and machine learning literatures to provide a causal framework that permits the identification and estimation of causal effects from observational studies. We begin by providing a causal interpretation for standard exogenous regressors and standard “valid” and “relevant” instrumental variables. We then build on this interpretation to characterize extended instrumental variables (EIV) methods, that is methods that make use of variables that need not be valid instruments in the standard sense, but that are nevertheless instrumental in the recovery of causal effects of interest. After examining special cases of single and double EIV methods, we provide necessary and sufficient conditions for the identification of causal effects by means of EIV and provide consistent and asymptotically normal estimators for the effects of interest.

In 2008 all ECB publications feature a motif taken from the 10 banknote. This paper can be downloaded without charge from

This paper tests whether macroeconomic news is transmitted to exchange rates via the transactions process and if so, what share occurs via transactions versus the traditional direct channel. We identify the link between order flow and macro news using a heteroskedasticity-based approach, a la Rigobon and Sack (2002). In both daily and intra-daily data, order flow varies considerably with macro news flow. At least half of the effect of macro news on exchange rates is transmitted via order flow.

Abstract. This paper demonstrates that “commodity currency ” exchange rates have remarkably robust power in predicting future global commodity prices, both in-sample and out-of-sample. A critical element of our in-sample approach is to allow for structural breaks, endemic to empirical exchange rate models, by implementing the approach of Rossi (2005b). Aside from its practical implications, our forecasting results provide perhaps the most convincing evidence to date that the exchange rate depends on the present value of identifiable exogenous fundamentals. We also find that the reverse relationship holds; that is, that commodity prices Granger-cause exchange rates. However, consistent with the vast post-Meese-Rogoff (1983a,b) literature on forecasting exchange rates, we find that the reverse forecasting regression does not survive out-of-sample testing. We argue, however, that it is quite plausible that exchange rates will be better predictors of exogenous commodity prices than vice-versa, because the exchange rate is fundamentally forward looking. Therefore, following Campbell and Shiller (1987) and Engel and West (2005), the exchange rate is likely to embody important information about future commodity price movements well beyond what econometricians can capture with simple time series models. In contrast, prices for most commodities are extremely sensitive to small shocks to current demand and supply, and are therefore likely to be less forward looking.

We consider issues related to the order of an autoregression selected using information criteria. We study the sensitivity of the estimated order to i) whether the effective number of observations is held fixed when estimating models of different order, ii) whether the estimate of the variance is adjusted for degrees of freedom, and iii) how the penalty for overfitting is defined in relation to the total sample size. Simulations show that the lag length selected by both the Akaike and the Schwarz information criteria are sensitive to these parameters in finite samples. The methods that give the most precise estimates are those that hold the effective sample size fixed across models to be compared. Theoretical considerations reveal that this is indeed necessary for valid model comparisons. Guides to robust model selection are provided.

Abstract not found

The objective of this paper is to identify which parameters of a model are stable over time. Existing procedures can only be used to test whether a given subset of parameters is stable, and cannot be used to find which subset of parameters is stable. We propose a new procedure that is informative on the nature of instabilities affecting economic models, and sheds light on the economic interpretation and causes of such instabilities. Furthermore, our procedure provides clear guidelines on which parts of the model are reliable for policy analysis and which are possibly mis-specified. Our empirical findings suggest that instabilities during the Great Moderation were mainly concentrated in Euler and IS equations as well as in monetary policy. Such results offer important insights to guide the future theoretical development of macroeconomic models.

Value-at-Risk (VaR) is widely used as a tool for measuring the market risk of asset portfolios. However, alternative VaR implementations are known to yield fairly di#erent VaR forecasts. Hence, every use of VaR requires choosing amongst alternative forecasting models.

We consider the estimation of the coefficients of a linear structural equation in a simultaneous equation system when there are many instrumental variables. We derive some asymptotic properties of the limited information maximum likelihood (LIML) estimator when the number of instruments is large; some of these results are new as well as old, and we relate them to results in some recent studies. We have found that the variance of the LIML estimator and its modifications often attain the asymptotic lower bound when the number of instruments is large and the disturbance terms are not necessarily normally distributed, that is, for the microeconometric models with many instruments.