Associate professor, Department of Business and Management, Aalborg University.

PhD in Finance, Finance Research Group, Aarhus School of Business, Aarhus University. MSc Economics.

- A large-dimensional factor analysis of the Federal Reserve's large-scale asset purchases
- Do Exchange Rates Really Help Forecasting Commodity Prices?
- Forecasting House Prices in the 50 States Using Dynamic Model Averaging and Dynamic Model Selection
- Housing Price Forecastability: A Factor Analysis
- Estimating US Monetary Poicy Shocks Using a Factor-Augmented Vector Autoregression: An EM Algorithm Approach
- Identification of Macroeconomic Factors in Large Panels
- Macro Factors, Monetary Policy Analysis and Affine Term Structure Models

..............................................................................................................

We examine U.S. housing price forecastability using principal component analysis (PCA), partial least squares (PLS), and sparse PLS (SPLS). We incorporate information from a large panel of 121 economic time series and show that macroeconomic fundamentals have strong predictive power for future movements in housing prices. We find that (S)PLS models systematically dominate PCA models. (S)PLS models also generate signi.cant out-of-sample predictive

power over and above the predictive power contained by the price-rent ratio and autoregressive benchmarks.

..............................................................................................................

We propose a new measure for housing sentiment and show that it accurately tracks expectations about future house price growth rates. We construct the housing sentiment index using partial least squares on household survey responses to questions about buying conditions for houses. We find that housing sentiment explains a large share of the time-variation in house prices during both boom and bust cycles and it strongly outperforms several macroeconomic variables typically used to forecast house prices.

Presented at BoE/ECB/CEPR/CFM in London 2015, Science Po (OFCE) in Paris 2015, IAAE in Thessaloniki, June 2015. SSRN

..............................................................................................................

This paper assesses the economy-wide effects of US unconventional monetary policy shocks. An unconventional expansionary monetary policy shock is identified as a shock that increases the Federal Reserve's market share of US treasuries and mortgage-backed securities, and leads to an improvement in the real economy and improved credit conditions.

I find that an unconventional monetary policy shock significantly drives down the long-term interest rate spread and the credit spread, and improves both the financial market conditions and the commercial and industrial loans activity. Moreover, the impact on the real economy is significant.

The roughly $2 trillion purchases of mortgage backed securities by the Federal Reserve Bank avoided a severe downturn according to estimates from a counterfactual analysis. [Read more under PAPERS .....]

Chen et al. (2010) report that for ‘commodity currencies’, the exchange rate predicts the country’s commodity index but not vice versa. The commodity currency hypothesis is consistent with the Engel and West (2005) exchange rate model if the fundamental is chosen to be the country’s key export prices and if the latter are exogenous to the exchange rate dynamics. In our view, however, commodity prices are essentially financial asset prices that are set in a forward-looking way, exactly like exchange rates. If both the exchange rate and the commodity prices are based on discounted future expectations, one should mostly observe contemporaneous correlations, not one-directional cross-predictability from one variable toward the other.

Using three different data sets and various econometric techniques, we do find the contemporaneous correlations as predicted by the financial asset view of commodity prices. Cross-predictability, in contrast, seems to be only minor at best, not robust to plausible variations in the test design, and bi-directional rather than one-directional. The difference between Chen et al’s empirical findings and ours is to a large extent traceable to the presence of time-averaged prices in the commodity index data that they use. Price averaging induces spurious autocorrelation and predictability that disappears if one uses e.g. month’s-end prices. Some slip-ups in their test design seem to play an additional role too.

Forecasting house prices in the 50 states using Dynamic Model Averaging and Dynamic Model Selection.

We examine house price forecastability across the 50 states using Dynamic Model Averaging and Dynamic Model Selection, which allow for model change and parameter shifts. By allowing the entire forecasting model to change over time and across locations, the forecasting accuracy improves substantially. The states in which housing markets have been the most volatile are the states where model change and parameter shifts have been the most needed.

(under revision).

Economy-wide effects of shocks to the US federal funds rate are estimated in a state space model with 120 US macroeconomic and financial time series driven by the dynamics of the federal funds rate and a few dynamic factors. This state space system is denoted a factor-augmented VAR (FAVAR) by Bernanke et al. (2005). I estimate the FAVAR by the fully parametric one-step EM algorithm as an alternative to the two-step principal component method and the one-step Bayesian method in Bernanke et al. (2005). The EM algorithm which is an iterative maximum likelihood method estimates all the parameters and the dynamic factors simultaneously and allows for classical inference. I demonstrate empirically that the same impulse responses but better fit emerge robustly from a low order FAVAR with eight correlated factors compared to a high order FAVAR with fewer correlated factors, for instance four factors. This empirical result accords with one of the theoretical results from Bai & Ng (2007) in which it is shown that the information in complicated factor dynamics may be substituted by panel information.

This paper presents a dynamic factor model in which the extracted factors and shocks are given a clear economic interpretation. The economic interpretation of the factors is obtained by means of a set of over-identifying loading restrictions, while the structural shocks are estimated following standard practices in the SVAR literature. Estimators based on the EM algorithm are developed. We apply this framework to a large panel of US monthly macroeconomic series. In particular, we identify nine macroeconomic factors and discuss the economic impact of monetary policy stocks. The results are theoretically plausible and in line with other findings in the literature.

Defended February 26, 2010.

The thesis consists of three self-contained chapters. Chapter 1 is about estimation of the Factor-Augmented VAR (FAVAR) by the EM algorithm. Chapter 2 is about well-defined economic interpretation of the dynamic factors which is achieved by means of over-identifying loading restrictions. Chapter 3 is about macroeconomic sources of variation in the expected excess bond returns as implied by an affine term structure model driven by data-rich macroeconomic state variables.