Interpretable sports team rating models based on the gradient descent algorithm

We introduce several new sports team rating models based on the gradient descent algorithm. More precisely, the models can be formulated by maximising the likelihood of match results observed using a single step of this optimisation heuristic. The proposed framework is inspired by the prominent Elo rating system, and yields an iterative version of ordinal logistic regression, as well as different variants of Poisson regression-based models. This construction makes the update equations easy to interpret, and adjusts ratings once new match results are observed. Thus, it naturally handles temporal changes in team strength. Moreover, a study of association football data indicates that the new models yield more accurate forecasts and are less computationally demanding than corresponding methods that jointly optimise the likelihood for the whole set of matches.     

Are low-frequency data really uninformative? A forecasting combination perspective

In this study, we investigate whether low-frequency data improve volatility forecasting when high-frequency data are available. To answer this question, we utilize four forecast combination strategies that combine low-frequency and high-frequency volatility models and employ a rolling window and a range of loss functions in the framework of the novel Model Confidence Set test. Out-of-sample results show that combination forecasts with GARCH-class models can achieve high forecast accuracy. However, the combination forecast methods appear not to significantly outperform individual high-frequency volatility models. Furthermore, we find that models that combine low-frequency and high-frequency volatility yield significantly better performance than other models and combination forecast strategies in both a statistical and economic sense.     

Would information on consumer confidence have helped to predict UK household expenditure during the recent economic crisis?

The objective is to investigate whether access to data on consumer confidence would have aided forecasts of the growth of UK household consumption expenditure over the recent period of economic crisis. A disaggregated study is performed on the basis that consideration is given not only to household spending in total but also to expenditure on each of durable goods, semi-durable goods, nondurable goods and services. The empirical analysis demonstrates how modifications which are made to the harmonized indicator of the European Commission are able to enhance predictive accuracy. However, the benefit which is derived from consulting consumer survey data does not extend to an earlier interval over which the behaviour of consumer sentiment was far less volatile.     

Rethinking weather station selection for electric load forecasting using genetic algorithms

Demand forecasting is and has been for years a topic of great interest in the electricity sector, being the temperature one of its major drivers. Indeed, one of the challenges when modelling the load is to choose the right weather station, or set of stations, for a given load time series. However, only a few research papers have been devoted to this topic. This paper reviews the most relevant methods that were applied during the Global Energy Forecasting Competition of 2014 (GEFCom2014) and presents a new approach to weather station selection, based on Genetic Algorithms (GA), which allows finding the best set of stations for any demand forecasting model, and outperforms the results of existing methods. Furthermore its performance has also been tested using GEFCom2012 data, providing significant error improvements. Finally, the possibility of combining the weather stations selected by the proposed GA using the BFGS algorithm is briefly tested, providing promising results.     

GEFCom2012: Electric load forecasting and backcasting with semi-parametric models

We sum up the methodology of the team tololo for the Global Energy Forecasting Competition 2012: Load Forecasting. Our strategy consisted of a temporal multi-scale model that combines three components. The first component was a long term trend estimated by means of non-parametric smoothing. The second was a medium term component describing the sensitivity of the electricity demand to the temperature at each time step. We use a generalized additive model to fit this component, using calendar information as well. Finally, a short term component models local behaviours. As the factors that drive this component are unknown, we use a random forest model to estimate it.     

产融结合进程、研究动态与发展趋势——基于我国经济体制改革的逻辑

改革开放以来,我国走出了一条不同于英美的市场主导型、德日的银行主导型和韩国的政府主导型产融结合之路.目前我国企业集团的产融结合呈现出央企挑大梁、地方国企崛起、民营企业集团争持金融牌照的态势,新兴高科技互联网企业利用技术优势新晋产融结合.本文遵循制度-体制-机制的演化规律,研究发现我国产融结合的广度和深度与经济体制改革紧密相关,具有鲜明的阶段性特征,历经萌发期、形成期和发展期几个阶段.因此,在我国当前经济发展新常态和全面深化改革要求下,产融结合应从系统性、协同性、突破性和前瞻性四个方面作更深入的研究.     

Rare shocks vs. non-linearities: What drives extreme events in the economy? Some empirical evidence

A small-scale vector autoregression (VAR) is used to shed some light on the roles of extreme shocks and non-linearities during stress events observed in the economy. The model focuses on the link between credit/financial markets and the real economy and is estimated on US quarterly data for the period 1984–2013. Extreme shocks are accounted for by assuming t-distributed reduced-form shocks. Non-linearity is allowed by the possibility of regime switch in the shock propagation mechanism. Strong evidence for fat tails in error distributions is found. Moreover, the results suggest that accounting for extreme shocks rather than explicit modeling of non-linearity contributes to the explanatory power of the model. Finally, it is shown that the accuracy of density forecasts improves if non-linearities and shock distributions with fat tails are considered.     

Earnings Volatility and Earnings Prediction: Analysis and UK Evidence

This paper confirms that US evidence of a negative relationship between earnings persistence and earnings volatility applies to UK firms over the period 1991–2010. Our analytical framework highlights the possibility that this result may reflect downward estimation bias in earnings persistence (and persistence of cash flow and accruals components of earnings) related to transitory earnings elements. Out‐of‐sample forecasts, based on models estimated for earnings volatility quartiles, suggest significant improvement in earnings forecasts for lower volatility firms. The results also suggest that the negative association between earnings persistence and volatility may be due to both estimation bias and variation in core earnings persistence.     

Future changes in age and household patterns: Some implications for public finances

Using stochastic forecasting techniques, this paper assesses the consequences for public finances of changes in age and household structures in Denmark over the period 2008–2037. Focusing on components of welfare provisions and tax payments with noticeable differences across age and household status, we show that, based on a point forecast, the fiscal impact of changes in household structures amounts to an annual negative effect of 0.5% of GDP, and the effect of changes in age structures is forecast to worsen the public budget by 3.7% of GDP per year. While being subject to a considerable amount of uncertainty, the prospect of such a dramatic weakening of public finances is likely to trigger demands for welfare reforms characterized by a more individualized system of public transfer and tax payments, in addition to the measures that have already been taken to address the fiscal effects of population ageing.     

Modelling exchange rate volatility with random level shifts

Recent literature has shown that the volatility of exchange rate returns displays long memory features. It has also been shown that if a short memory process is contaminated by level shifts, the estimate of the long memory parameter tends to be upward biased. In this article, we directly estimate a random level shift model to the logarithm of the absolute returns of five exchange rates series, in order to assess whether random level shifts (RLSs) can explain this long memory property. Our results show that there are few level shifts for the five series, but once they are taken into account the long memory property of the series disappears. We also provide out-of-sample forecasting comparisons, which show that, in most cases, the RLS model outperforms popular models in forecasting volatility. We further support our results using a variety of robustness checks.