Till Örebro universitet

In this paper, we investigate whether key relations in the crude oil market have been stable over time. This is done by estimating hybrid time-varying parameter structural Bayesian VAR models using monthly data ranging from February 1973 to May 2023. Model selection suggests that while stochastic volatility is preferred over homoscedasticity, the dynamics of the model are best described by constant parameters in all equations.

In this paper, we evaluate the directional interest-rate forecasts of UK households from the Bank of England’s Inflation Attitudes Survey. Employing a test for directional forecast accuracy and data on the survey balance ranging from 1999Q4 to 2023Q2, we find that the balance is not able to predict in which direction the interest rate will move over the coming year. In addition, regression models based on the balance are not able to generate forecasts for the quantitative change in the interest rate over the coming twelve months that have higher precision than a naïve forecast of no change. In order to provide information as to whether our findings are due to the inherent difficulty when it comes to forecasting interest rates or if households are not very insightful regarding interest rates, we investigate – again using data on the survey balance and testing for directional accuracy – whether households have been able to correctly assess the directional change of the interest rate over the previous twelve months; our results indicate some amount of “literacy” among the households regarding the interest rates that they face. Finally, analyses based on individual-response level data suggest that literacy regarding interest rates – proxied by the respondent having been correct regarding the directional change over the previous twelve months – does not appear helpful when forecasting.

We analyse how financial market analysts' expectations in the Czech National Bank's Financial Market Inflation Expectations survey perform relative to the random-walk forecast when it comes to predicting five financial variables. Using data from 2001 to 2022, our results indicate that the analysts are able to significantly outperform the random-walk forecast in terms of forecast precision for the repo rate and Prague Interbank Offered Rate at the one-month forecasting horizon. For the five- and ten-year interest rate swap rates, the random walk significantly outperforms the analysts at both the one-month and one-year forecasting horizons. For the CZK/EUR exchange rate, the random-walk forecast has a lower root mean squared forecast error than that of the analysts' forecast at the one-month horizon whereas at the one-year horizon the opposite is found; however, none of these differences are statistically significant.

In this paper, the natural rate of interest in Denmark, Norway and Sweden is estimated. This is done by augmenting the Laubach and Williams (Rev Econ Stat 85:1063-1070, 2003) framework with a dynamic factor model linked to economic indicators--a modelling choice which allows us to better identify business cycle fluctuations. We estimate the model using Bayesian methods on data ranging from 1990Q1 to 2022Q4. The results indicate that the natural rate has declined substantially and in all countries is at a low level at the end of the sample.

In this paper, we extend the standard Gaussian stochastic-volatility Bayesian VAR by employing the generalized hyperbolic skew Student’s t distribution for the innovations. Allowing the skewness parameter to vary over time, our specification permits flexible modelling of innovations in terms of both fat tails and – potentially dynamic – asymmetry. In an empirical application using US data on industrial production, consumer prices and economic policy uncertainty, we find support – although to a moderate extent – for time-varying skewness. In addition, we find that shocks to economic policy uncertainty have a negative effect on both industrial production growth and CPI inflation.

We assess the bivariate relation between money growth and inflation in the euro area and the United States using hybrid time-varying parameter Bayesian VAR models. Model selection based on marginal likelihoods suggests that the relation is statistically unstable across time in both regions. The effect of money growth on inflation weakened notably after the 1980s before strengthening after 2020. There is evidence that this time variation is related to the pace of price changes, as we find that the maximum impact of money growth on inflation is increasing in the trend level of inflation. These results caution against asserting a simple, time-invariant relationship when modeling the joint dynamics of monetary aggregates and consumer prices.

We estimate the trend short-term interest rate in the United States using an unobserved-components stochastic-volatility model with interest-rate and survey data from 1998Q2 to 2022Q4. Our results indicate that the trend short-term interest rate has drifted down during most of the sample and remains low in a historical perspective, despite the recent sharp increase in the short-term interest rate.

It has been claimed that the fall in US inflation during the Great Recession was surprisingly small. One possible explanation for this is that the Phillips curve is unstable and that its slope was lower around the Great Recession. We investigate the importance of time-varying parameters using Bayesian vector autoregressions for inflation and unemployment. We find support for time variation in the inflation equation and an unstable Phillips curve that was somewhat flatter between 2005 and 2013. However, conditional forecasts mostly suggest that inflation was not unexpectedly high around the Great Recession, which puts the claim of a "missing disinflation" into question.

We analyse micro-level data concerning four financial variables in Sveriges Riksbank's Prospera Survey to evaluate the precision of forecasts provided by professionals active in the Swedish fixed -income market. Our results indicate that for the SEK/EUR and SEK/USD exchange rates, and the five-year government bond yield, none of the market participants that frequently participate in the survey manage to significantly outperform the random-walk forecast. For the central bank's policy rate, the market participants typically have a statistically significant higher forecast pre-cision than the random-walk forecast at the three-month horizon; however, at the two-and five-year horizons, the random-walk forecast typically outperforms the market participants.

In this paper, we analyze how skewness and heavy tails affect the estimated relationship between the real economy and the corporate bond-yield spread-a popular predictor of real activity. We use quarterly US data to estimate Bayesian VAR models with stochastic volatility and various distributional assumptions regarding the innovations. In-sample, we find that-after controlling for stochastic volatility-innovations in GDP growth can be well described by a Gaussian distribution. In contrast, the yield spread appears to benefit from being modeled using non-Gaussian innovations. When it comes to real-time forecasting performance, we find that the yield spread is a relevant predictor of GDP growth at the one-quarter horizon. Having controlled for stochastic volatility, gains in terms of forecasting performance from flexibly modeling the innovations appear to be limited and are mostly found for the yield spread.