Knut Are Aastveit

Journal of business & economic statistics Doi: https://doi.org/10.1080/07350015.2022.2039159

We propose a novel and numerically efficient quantification approach to forecast uncertainty of the real price of oil using a combination of probabilistic individual model forecasts. Our combination method extends earlier approaches that have been applied to oil price forecasting, by allowing for sequentially updating of time-varying combination weights, estimation of time-varying forecast biases and facets of miscalibration of individual forecast densities and time-varying inter-dependencies among models. To illustrate the usefulness of the method, we present an extensive set of empirical results about time- varying forecast uncertainty and risk for the real price of oil over the period 1974–2018. We show that the combination approach systematically outperforms commonly used benchmark models and combination approaches, both in terms of point and density forecasts. The dynamic patterns of the estimated individual model weights are highly time-varying, reflecting a large time variation in the relative performance of the various individual models. The combination approach has built-in diagnostic information measures about forecast inaccuracy and/or model set incompleteness, which provide clear signals of model incompleteness during three crisis periods. To highlight that our approach also can be useful for policy analysis, we present a basic analysis of profit-loss and hedging against price risk.

American Economic Journal: Macroeconomics 14(4) s. 499-529 Doi: https://doi.org/10.1257/mac.20190011

The responsiveness of house prices to monetary policy shocks depends on the nature of the shock—expansionary versus contractionary—and on local housing supply elasticities. These findings are established using a panel of 263 US metropolitan areas. Expansionary monetary policy shocks have a larger impact on house prices in supply inelastic areas. Contractionary shocks are orthogonal to housing supply elasticities. In supply elastic areas, contractionary shocks have a greater impact on house prices than expansionary shocks. The opposite holds true in supply inelastic areas. We attribute this to asymmetric housing supply adjustments.

Review of Economics and Statistics Doi: https://doi.org/10.1162/rest_a_01120

We investigate whether the Federal Reserve has responded systematically to house and stock prices and whether this response has changed over time using a Bayesian structural VAR model with time-varying parameters and stochastic volatility. To recover the systematic component of monetary policy, we interpret the interest rate equation in the VAR as an extended monetary policy rule responding to ination, the output gap, house prices and stock prices. Our results indicate that the systematic component of monetary policy in the U.S. responded to real stock price growth significantly but episodically, mainly around recessions and periods of financial instability, and took real house price growth into account only in the years preceding the Great Recession. Around half of the estimated response captures the predictor role of asset prices for future ination and real economic activity, while the remaining component reects a direct response to stock prices and house prices.

Journal of the American Statistical Association Doi: https://doi.org/10.1080/01621459.2019.1660171

We present new methodology and a case study in use of a class of Bayesian predictive synthesis (BPS) models for multivariate time series forecasting. This extends the foundational BPS framework to the multivariate setting, with detailed application in the topical and challenging context of multi-step macroeconomic forecasting in a monetary policy setting. BPS evaluates- sequentially and adaptively over time- varying forecast biases and facets of miscalibration of individual forecast densities for multiple time series, and- critically- their time-varying inter-dependencies. We define BPS methodology for a new class of dynamic multivariate latent factor models implied by BPS theory. Structured dynamic latent factor BPS is here motivated by the application context- sequential forecasting of multiple US macroeconomic time series with forecasts generated from several traditional econometric time series models. The case study highlights the potential of BPS to improve of forecasts of multiple series at multiple forecast horizons, and its use in learning dynamic relationships among forecasting models or agents.

International Journal of Forecasting Doi: https://doi.org/10.1016/j.ijforecast.2018.09.008

We assess the importance of residential investment for the prediction of economic recessions for an unbalanced panel of 12 OECD countries over the period 1960Q1–2014Q4. Our approach is to estimate various probit models with different leading indicators and evaluate their relative prediction accuracies using the area under the receiver operating characteristic curve as our forecasting performance metric. We document that residential investment contains information that is useful for predicting recessions both in-sample and out-of-sample. This result is robust to adding typical leading indicators, such as the term spread, stock prices, consumer confidence surveys and oil prices. It is shown that residential investment is particularly useful for the prediction of recessions for countries with high home-ownership rates. Finally, in a separate exercise for the US, we show that the predictive ability of residential investment is — in a broad sense — robust to employing real-time data.

Journal of International Money and Finance 76 s. 50-67 Doi: https://doi.org/10.1016/j.jimonfin.2017.05.003

This paper explores if economic uncertainty alters the macroeconomic influence of monetary policy. We use several measures of U.S. economic uncertainty, and estimate their interaction with monetary policy shocks as identified through structural vector autoregressions. We find that U.S. monetary policy shocks affect economic activity less when uncertainty is high, in line with “real-option” effects from theory. Holding uncertainty constant, the effect on investment is approximately halved when uncertainty is in its top instead of its bottom decile.

International Journal of Forecasting 32(2) s. 283-292 Doi: https://doi.org/10.1016/j.ijforecast.2015.06.006

The Scandinavian Journal of Economics 118(1) s. 168-195 Doi: https://doi.org/10.1111/sjoe.12126

Journal of applied econometrics 30(7) s. 1013-1028 Doi: https://doi.org/10.1002/jae.2406

Journal of business & economic statistics 32(1) s. 48-68 Doi: https://doi.org/10.1080/07350015.2013.844155

In this paper, we use U.S. real-time data to produce combined density nowcasts of quarterly GDP growth, using a system of three commonly used model classes. We update the density nowcast for every new data release throughout the quarter, and highlight the importance of new information for nowcasting. Our results show that the logarithmic score of the predictive densities for U.S. GDP growth increase almost monotonically, as new information arrives during the quarter. While the ranking of the model classes changes during the quarter, the combined density nowcasts always perform well relative to the model classes in terms of both logarithmic scores and calibration tests. The density combination approach is superior to a simple model selection strategy and also performs better in terms of point forecast evaluation than standard point forecast combinations.

Quarterly Review of Economics and Finance 54(2) s. 180-193 Doi: https://doi.org/10.1016/j.qref.2013.09.003

Energy Economics 45 s. 268-279 Doi: https://doi.org/10.1016/j.eneco.2014.07.006

Empirical Economics 42(1) s. 95-119 Doi: https://doi.org/10.1007/s00181-010-0429-9

Handelshøyskolen BI

We analyze the importance of demand from emerging and developed economies as drivers of the real price of oil over the last two decades. Using a factor-augmented vector autoregressive (FAVAR) model that allows us to distinguish between different groups of countries, we find that demand from emerging economies (most notably from Asian countries) is more than twice as important as demand from developed countries in accounting for the fluctuations in the real price of oil and in oil production. Furthermore, we find that different geographical regions respond differently to oil supply shocks and oil- specific demand shocks that drive up oil prices, with Europe and North America being more negatively affected than emerging economies in Asia and South America. We demonstrate that this heterogeneity in responses is not only attributable to differences in energy intensity in production across regions but also to degree of openness and the investment share in GDP.

Handelshøyskolen BI

This paper bridges the new open economy factor augmented VAR (FAVAR) studies with the recent findings in the business cycle synchronization literature emphasizing the importance of regional factors. That is, we estimate and identify a three block FAVAR model with separate world, regional and domestic blocks and study the transmission of both global and regional shocks to four small open economies (Canada, New Zealand, Norway and UK). The results show that foreign shocks explain a major share of the variance in all countries, most so shocks that are common to the world. However, regional shocks also play an important role, explaining more than 20 percent of the variance in the variables. Hence in small open economies, the world is not enough. The regional factors impact the four countries differently, though, some through trade and some through consumer sentiment. Our findings of a strong transmission of both global and regional shocks to open economies are in sharp contrast to the evidence from recently developed open economy DSGE models.

Norges Bank

Handelshøyskolen BI

In this paper we use U.S. real-time vintage data and produce combined density nowcasts for quarterly GDP growth from a system of three commonly used model classes. The density nowcasts are combined in two steps. First, a wide selection of individual models within each model class are combined separately. Then, the nowcasts from the three model classes are combined into a single predictive density. We update the density nowcast for every new data release throughout the quarter, and highlight the importance of new information for the evaluation period 1990Q2-2010Q3. Our results show that the logarithmic score of the predictive densities for U.S. GDP increase almost monotonically as new information arrives during the quarter. While the best performing model class is changing during the quarter, the density nowcasts from our combination framework is always performing well both in terms of logarithmic scores and calibration tests. The density combination approach is superior to a simple model selection strategy and also performs better in terms of point forecast evaluation than standard point forecast combinations.

The 23rd Annual Congress of the European Economic Association

The 14th International Conference on Computing in Economics and Finance

The 4th Annual CIRANO Workshop on Data Revision in Macroeconomic Forecasting and Policy

The 28 International Symposium on Forecasting

The 28th International Symposium on Forecasting