The cyclical properties of the Baltic Dry Index (BDI) and their implications for forecasting performance are investigated. We find that changes in BDI can lead to permanent shocks trade major exporting economies. In our exercise, we show commodities trigonometric regression improved predictions then use results perform an investment exercise how they be used risk management freight sector.

ABSTRACT This paper explores the information content of untargeted narratives Bank England (BoE), European Central (ECB), and Federal Reserve (Fed) whether it has potential to improve forecasting performance. We apply Latent Dirichlet Allocation (LDA) method extract topics from corpus text data. then evaluate impact these central bank officials' speeches on macroeconomic financial variables 1997 2018. Our results suggest that model, incorporating speeches, produces estimates with a lower...

Abstract Covid-19 and lockdowns have had spillover effects on other health outcomes. Motor vehicle collisions (MVC) are likely to been affected by the pandemic due to, among others, less traffic volume speeding empty streets. This paper studies impact of MVCs in Northern Ireland. Using monthly data injuries deaths, we find a steep decline slight serious compared what would expected absence pandemic. However, no effect number deaths. Based from tickets, plausible explanation for differential...

Abstract Empirical evidence is presented about the properties of economic sentiment cycle synchronization for G ermany, F rance and UK they are compared with “crisis” countries I taly, S pain, P ortugal reece. Instead using output data it preferred to focus on indicator ( ESI ), a forward‐looking, survey‐based variable consistently available from 1985. The cyclical nature allowed analyis presence or not synchronicity among country pairs before after onset financial crisis. results show that...

This paper aims at providing a primer on the use of big data in macroeconomic nowcasting and early estimation. We discuss: (i) typology characteristics relevant for estimates, (ii) methods features extraction from unstructured to usable time series, (iii) econometric that could be used with data, (iv) some empirical results key target variables four EU countries, (v) ways evaluate nowcasts ash estimates. conclude by set recommendations assess pros cons specic context.

A bootstrap methodology suitable for use with stationary and non‐stationary fractionally integrated time series is further developed in this article. The resampling algorithm involves estimating the degree of fractional integration, applying differencing operator, resulting approximation to underlying short memory and, finally, cumulating obtain a resample original process. This approach extends existing methods literature by allowing general schemes including blockwise bootstraps....

We consider the issue of Block Bootstrap methods in processes that exhibit strong dependence. The main difficulty is to transform series such way implementation these techniques can provide an accurate approximation true distribution test statistic under consideration. bootstrap algorithm we suggest consists following operations: given xt ~ I(d0), 1) estimate long memory parameter and obtain dˆ, 2) difference dˆ times, 3) apply block on above finally, 4) cumulate sample times. Repetition...

Abstract This paper empirically studies the reversal pattern following formation of trend‐following signals in time series context. is statistically significant and usually occurs between 12 24 months after signals. Employing a universe 55 liquid futures, we find that instruments with sell portfolio (‘losers’) contribute to this type reversal, even if their profits are not realised. The buy (‘winners’) much less. A double‐sorted investment strategy based on both return continuation yields...

We consider forecasting key macroeconomic variables using many predictors extracted from the Eurostat PEEIs dataset. To avoid curse of dimensionality, we rely on model selection and reduction. For use heuristic optimisation information criteria, including simulated annealing, genetic algorithms, MC^3 sequential testing. reduction employ methods principal components, partial least squares Bayesian shrinkage regression. The problem unbalanced datasets is discussed potential solutions are...

This article considers a multivariate system of fractionally integrated time series and investigates the most appropriate way for estimating Impulse Response (IR) coefficients their associated confidence intervals. The extends univariate analysis recently provided by Baillie Kapetanios (2013 Baillie, R. T., Kapetanios, G. (2013). Estimation inference impulse response functions form strongly persistent processes. Econometrics Journal 16:373–399.[Crossref], [Web Science ®] , [Google Scholar]),...

This paper assesses the forecasting performance of various variable reduction and selection methods. A small a large set wisely chosen variables are used in industrial production growth for four Euro Area economies. The results indicate that Automatic Leading Indicator (ALI) model performs well compared to other methods datasets. However, Partial Least Squares using heuristic optimisations information criteria along with ALI could be averaging methodologies.

This paper proposes a modified version of the widely used price and moving average cross-over trading strategies. The suggested approach (presented in its 'long only' version) is combination 'buy' signals dynamic threshold value which acts as trailing stop. behavior performance from this strategy different standard with results showing that, on average, proposed modification increases cumulative return Sharpe ratio investor while exhibiting smaller maximum drawdown duration than strategy.