An adaptive approach to forecasting three key macroeconomic variables for transitional China

The macroeconomic forecasts for emerging economies often suffer from the constraints of instability and limited data. In light of these constraints, we propose the use of a local autoregressive (LAR) model with a data-driven estimation window, i.e., a local homogenous interval, that is adaptively identified to strike a balance between information efficiency and stability. When applied to three key macroeconomic variables of China, the LAR model substantially outperforms the alternative models for various forecast horizons of 3 to 12 months, with forecast error reductions of between 4% and 64% for the IP growth, and between 1% and 68% for the inflation rate. The one-quarter ahead performance of the LAR model matches that of a well-known survey forecast. The patterns of the identified local intervals also coincide with the characteristic evolution of the gradual reforms and monetary policy shifts in China. In short, the LAR model is suitable for not only forecasting, but also the real-time monitoring of the effects of regime and policy changes in emerging economies.     

Do analysts who understand accounting conservatism exhibit better forecasting performance?

This study investigates the performance of analysts when they match the asymmetric timeliness of their earnings forecast revisions (i.e., asymmetric forecast timeliness) with the asymmetric timeliness of firms’ reported earnings (i.e., asymmetric earnings timeliness). We find that better timeliness‐matching analysts produce more accurate earnings forecasts and elicit stronger market reactions to their forecast revisions. Further, better timeliness‐matching analysts issue less biased earnings forecasts, more profitable stock recommendations and have more favorable career outcomes. Overall, our results indicate that analysts’ ability to incorporate conditional conservatism into their earnings forecasts is an important reflection of analyst expertise and professional success.     

Combining multiple probability predictions using a simple logit model

This paper begins by presenting a simple model of the way in which experts estimate probabilities. The model is then used to construct a likelihood-based aggregation formula for combining multiple probability forecasts. The resulting aggregator has a simple analytical form that depends on a single, easily-interpretable parameter. This makes it computationally simple, attractive for further development, and robust against overfitting. Based on a large-scale dataset in which over 1300 experts tried to predict 69 geopolitical events, our aggregator is found to be superior to several widely-used aggregation algorithms.     

A model of West African millet prices in rural markets

In this article we specify a model of millet prices in the three West African countries of Burkina Faso, Mali, and Niger. Using data obtained from USAID’s Famine Early Warning Systems Network (FEWS NET) we present a unique regional millet price forecasting model that takes advantage of the panel nature of our data and accounts for the distance of rural markets from capital cities. Another novel aspect of our analysis is our use of the Normalized Difference Vegetation Index (NDVI) to detect and control for variation in conditions for productivity. We find that including NDVI information significantly improves price forecasts.     

A gradient boosting approach to the Kaggle load forecasting competition

We describe and analyse the approach used by Team TinTin (Souhaib Ben Taieb and Rob J Hyndman) in the Load Forecasting track of the Kaggle Global Energy Forecasting Competition 2012. The competition involved a hierarchical load forecasting problem for a US utility with 20 geographical zones. The data available consisted of the hourly loads for the 20 zones and hourly temperatures from 11 weather stations, for four and a half years. For each zone, the hourly electricity loads for nine different weeks needed to be predicted without having the locations of either the zones or stations. We used separate models for each hourly period, with component-wise gradient boosting for estimating each model using univariate penalised regression splines as base learners. The models allow for the electricity demand changing with the time-of-year, day-of-week, time-of-day, and on public holidays, with the main predictors being current and past temperatures, and past demand. Team TinTin ranked fifth out of 105 participating teams.     

Crime prediction by data-driven Green’s function method

We develop an algorithm that forecasts cascading events, by employing a Green’s function scheme on the basis of the self-exciting point process model. This method is applied to open data of 10 types of crimes happened in Chicago. It shows a good prediction accuracy superior to or comparable to the standard methods which are the expectation–maximization method and prospective hotspot maps method. We find a cascade influence of the crimes that has a long-time, logarithmic tail; this result is consistent with an earlier study on burglaries. This long-tail feature cannot be reproduced by the other standard methods. In addition, a merit of the Green’s function method is the low computational cost in the case of high density of events and/or large amount of the training data.     

The role of text-extracted investor sentiment in Chinese stock price prediction with the enhancement of deep learning

Whether investor sentiment affects stock prices is an issue of long-standing interest for economists. We conduct a comprehensive study of the predictability of investor sentiment, which is measured directly by extracting expectations from online user-generated content (UGC) on the stock message board of Eastmoney.com in the Chinese stock market. We consider the influential factors in prediction, including the selections of different text classification algorithms, price forecasting models, time horizons, and information update schemes. Using comparisons of the long short-term memory (LSTM) model, logistic regression, support vector machine, and Naïve Bayes model, the results show that daily investor sentiment contains predictive information only for open prices, while the hourly sentiment has two hours of leading predictability for closing prices. Investors do update their expectations during trading hours. Moreover, our results reveal that advanced models, such as LSTM, can provide more predictive power with investor sentiment only if the inputs of a model contain predictive information.     

Growth and real business cycles in Vietnam and the Asean-5. Does the trend shock matter?

We examine Vietnam’s economy in comparison with its closest trade partners. We show that capital accumulation has been the primary growth engine since the start of its transition to the pro-market economy in 1986 – the Doi Moi. We also show that the cyclical behavior of its macro aggregates is similar to that of its ASEAN-5 peers and other developing countries. We extend the standard small open economy RBC model by considering habit persistence and government consumption, which allows a close match of the moments of the growth variables. At the business cycle frequency, transitory productivity shocks account for approximately one-half of Vietnam’s output variance, while country risk and non-transitory productivity shocks account to close to one-fifth each. Regarding the Solow residual’s volatility, we find that the trend component merely accounts for 12 % of this variance in Vietnam, while in Thailand it is only 6 %. These findings refute the “the cycle is the trend” hypothesis in Aguiar and Gopinath (2007) and align with the hypotheses in García-Cicco et al. (2010) and Rhee (2017), where the stationary component is overwhelmingly dominant. We claim that technological progress and productivity-enhancing measures are fundamental for Vietnam’s economy to sustain high growth.     

基于指数平滑法的G市主要货物流通量预测

随着国家发展西部地区的力度不断加大,贵州省作为西部地区省份也乘着这股力度不断在各方面完善自己,G市作为贵州省的主要城市之一近几年在经济、政治等方面都有显著的提高。物流的发展在一定程度上可以带动一个地区产业的发展,物流的构架在一定程度上可以反映出一个地区产业的构架。通过对G市物流运营现状的分析,发现G市的产业结构的特性,然后,通过分析建立符合G市物流发展的预测模型,最后结合模型和分析对G市物流发展提出合理化建议。     

Evaluating quantile forecasts in the M5 uncertainty competition

Probabilistic forecasts are necessary for robust decisions in the face of uncertainty. The M5 Uncertainty competition required participating teams to forecast nine quantiles for unit sales of various products at various aggregation levels and for different time horizons. This paper evaluates the forecasting performance of the quantile forecasts at different aggregation levels and at different quantile levels. We contrast this with some theoretical predictions, and discuss potential implications and promising future research directions for the practice of probabilistic forecasting.