Crude oil price forecasting incorporating news text

Sparse and short news headlines can be arbitrary, noisy, and ambiguous, making it difficult for classic topic model LDA (latent Dirichlet allocation) designed for accommodating long text to discover knowledge from them. Nonetheless, some of the existing research about text-based crude oil forecasting employs LDA to explore topics from news headlines, resulting in a mismatch between the short text and the topic model and further affecting the forecasting performance. Exploiting advanced and appropriate methods to construct high-quality features from news headlines becomes crucial in crude oil forecasting. This paper introduces two novel indicators of topic and sentiment for the short and sparse text data to tackle this issue. Empirical experiments show that AdaBoost.RT with our proposed text indicators, with a more comprehensive view and characterization of the short and sparse text data, outperforms the other benchmarks. Another significant merit is that our method also yields good forecasting performance when applied to other futures commodities.     

Forecasting bulk prices of Bordeaux wines using leading indicators

Agricultural price forecasting has been being abandoned progressively by researchers ever since the development of large-scale agricultural futures markets. However, as with many other agricultural goods, there is no futures market for wine. This paper draws on the agricultural prices forecasting literature to develop a forecasting model for bulk wine prices. The price data include annual and monthly series for various wine types that are produced in the Bordeaux region. The predictors include several leading economic indicators of supply and demand shifts. The stock levels and quantities produced are found to have the highest predictive power. The preferred annual and monthly forecasting models outperform naive random walk forecasts by 27.1% and 3.4% respectively; their mean absolute percentage errors are 2.7% and 3.4% respectively. A simple trading strategy based on monthly forecasts is estimated to increase profits by 3.3% relative to a blind strategy that consists of always selling at the spot price.     

Quasi‐rational and ex ante price expectations in commodity supply models: an empirical analysis of the US broiler market

A statistically optimal inference about agents' ex ante price expectations within the US broiler market is derived using futures prices of related commodities along with a quasi‐rational forecasting regression equation. The modelling approach, which builds on a Hamilton‐type framework, includes endogenous production and allows expected cash price to be decomposed into anticipated and unanticipated components. We therefore infer the relative importance of various informational sources in expectation formation. Results show that, in addition to the quasi‐rational forecast, the true supply shock, future prices, and ex post commodity price forecast errors have, at times, been influential in broiler producers' price expectations. Copyright © 2003 John Wiley & Sons, Ltd.     

The effects of economic news on commodity prices

We assess how commodity prices respond to macroeconomic news and show that commodities have been relatively insensitive to such news over daily frequencies between 1997 and 2009 compared to other financial assets and major exchange rates. Where commodity prices are influenced by news, there is a pro-cyclical bias and these sensitivities have risen as commodities have become increasingly financialized. However, models based on news still do a relatively poor job of forecasting commodity prices at daily frequencies. We also find some asymmetries in how commodity prices respond to news, most notably for gold, which alone among commodities acts as a safe-haven when “bad” economic news emerges.     

Excess co-movement of agricultural futures prices: Perspective from contagious investor sentiment

This study investigates the excess co-movement of agricultural futures prices from a new perspective of contagious investor sentiment. This study shows that contagious investor sentiment is a key determinant of excess co-movement of agricultural futures prices, by using contagious investor sentiment among different agricultural futures. Further, this study decomposes contagious investor sentiment into expected and unexpected contagious investor sentiment. Results show that both of them can positively affect excess co-movement of agricultural futures prices. More interestingly, expected contagious investor sentiment outperforms unexpected contagious investor sentiment in soybean 1 future, soymeal future, and strong wheat future. In general, the results of this study can provide strong support for the significant roles of contagious investor sentiment in asset pricing applications.     

Text-based crude oil price forecasting: A deep learning approach

This study proposes a new, novel crude oil price forecasting method based on online media text mining, with the aim of capturing the more immediate market antecedents of price fluctuations. Specifically, this is an early attempt to apply deep learning techniques to crude oil forecasting, and to extract hidden patterns within online news media using a convolutional neural network (CNN). While the news-text sentiment features and the features extracted by the CNN model reveal significant relationships with the price change, they need to be grouped according to their topics in the price forecasting in order to obtain a greater forecasting accuracy. This study further proposes a feature grouping method based on the Latent Dirichlet Allocation (LDA) topic model for distinguishing effects from various online news topics. Optimized input variable combination is constructed using lag order selection and feature selection methods. Our empirical results suggest that the proposed topic-sentiment synthesis forecasting models perform better than the older benchmark models. In addition, text features and financial features are shown to be complementary in producing more accurate crude oil price forecasts.     

THE INFLUENCE OF FUNDAMENTALS ON EXCHANGE RATES: FINDINGS FROM ANALYSES OF NEWS EFFECTS

Abstract As we survey the literature of macroeconomic news in the foreign exchange market, we can by now look back on nearly 30 years of research. The first studies which analysed news effects on exchange rates were established in the early 1990s (see, for example, Dornbusch). Almost at the same time Meese and Rogoff published their influential paper, revealing the forecasting inferiority in exchange rates of structural models against the random walk. This finding has shocked the pillars of exchange rate economics and thus cast general suspicion on research focusing on fundamentals in this field. The eventual rising popularity of event studies can partly be attributed to the re‐establishment of the raison d’être of exchange rate economics. This work focuses on systematically surveying this literature with particular respect to its primary goal, i.e. shedding light on the analytical value of fundamental research. Thus, its major findings are, first, fundamental news does matter, whereas non‐fundamental news matters to a lesser degree. Second, news influences exchange rates via two separated channels, i.e. incorporating common information into prices directly or indirectly based upon order flow. Third, with a few exceptions the impact of fundamental news on exchange rates is fairly stable over time.     

Adaptive learning forecasting,with applications in forecasting agricultural prices

We introduce a new forecasting methodology, referred to as adaptive learning forecasting, that allows for both forecast averaging and forecast error learning. We analyze its theoretical properties and demonstrate that it provides a priori MSE improvements under certain conditions. The learning rate based on past forecast errors is shown to be non-linear. This methodology is of wide applicability and can provide MSE improvements even for the simplest benchmark models. We illustrate the method’s application using data on agricultural prices for several agricultural products, as well as on real GDP growth for several of the corresponding countries. The time series of agricultural prices are short and show an irregular cyclicality that can be linked to economic performance and productivity, and we consider a variety of forecasting models, both univariate and bivariate, that are linked to output and productivity. Our results support both the efficacy of the new method and the forecastability of agricultural prices.     

Jumps,cojumps and macro announcements

We use recently proposed tests to extract jumps and cojumps from three types of assets: stock index futures, bond futures, and exchange rates. We then characterize the dynamics of these discontinuities and informally relate them to US macroeconomic releases before using limited dependent variable models to formally model how news surprises explain (co)jumps. Nonfarm payroll and federal funds target announcements are the most important news across asset classes. Trade balance shocks are important for foreign exchange jumps. We relate the size, frequency and timing of jumps across asset classes to the likely sources of shocks and the relation of asset prices to fundamentals in the respective classes. Copyright © 2010 John Wiley & Sons, Ltd.     

Forecasting returns in the VIX futures market

This paper introduces a new forecasting model for VIX futures returns. The model is structural in nature and parsimonious, and contains parameters that are relatively easy to estimate. The forecasts of next day VIX futures returns based on this model are superior to those produced by a linear forecasting model that uses the same set of predictors. Moreover, the profits to a market-timing model based on the proposed forecasts are statistically and economically significant, and are robust to both the method used for adjusting for risk and transaction costs (up to around 15 basis points). In contrast, the forecasts generated by the linear forecasting model are not.     

The role of oil futures intraday information on predicting US stock market volatility

This study investigates the role of oil futures price information on forecasting the US stock market volatility using the HAR framework. In-sample results indicate that oil futures intraday information is helpful to increase the predictability. Moreover, compared to the benchmark model, the proposed models improve their predictive ability with the help of oil futures realized volatility. In particular, the multivariate HAR model outperforms the univariate model. Accordingly, considering the contemporaneous connection is useful to predict the US stock market volatility. Furthermore, these findings are consistent across a variety of robust checks.     

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.     

Thin trading and mispricing profit opportunities in the Canadian commodity futures

This study examines whether thin trading problems in the Canadian futures market can create mispricing profit opportunities for canola and feed wheat futures traded over the period 1981 through 1993. A forecasting model is developed using historical and publicly available information to predict futures closing prices for these contracts, then two trading rules (a confidence interval and a percentage price change filter) are used to determine their profit potentials. The size of profits generated from trading canola futures under either rule during the period 1987–1993 is consistent with C. Carter's (1989) earlier results that no market inefficiency was detected during the 1980–1987 period. Similarly, profits from the Canadian feed wheat thinly traded contracts and from a control group using the highly-liquid American soybean oil and wheat contracts do not violate the efficiency theory. The average gross profit per trade analysis further suggests that net positive profits may not be viable for marginal investors.     

Combining forecasts: Performance and coherence

There is general agreement in many forecasting contexts that combining individual predictions leads to better final forecasts. However, the relative error reduction in a combined forecast depends upon the extent to which the component forecasts contain unique/independent information. Unfortunately, obtaining independent predictions is difficult in many situations, as these forecasts may be based on similar statistical models and/or overlapping information. The current study addresses this problem by incorporating a measure of coherence into an analytic evaluation framework so that the degree of independence between sets of forecasts can be identified easily. The framework also decomposes the performance and coherence measures in order to illustrate the underlying aspects that are responsible for error reduction. The framework is demonstrated using UK retail prices index inflation forecasts for the period 1998–2014, and implications for forecast users are discussed.     

Central bank policy announcements and changes in trading behavior: Evidence from bond futures high frequency price data

We present an analytical framework to investigate surprises in financial markets. The framework enables us to simultaneously identify and quantify surprises in security price data. By applying the framework to the tick-by-tick data on Japanese government bond futures prices, we find that the Bank of Japan’s introduction of quantitative and qualitative monetary easing in 2013 was one of the most surprising episodes during the period from 2005 to 2016. We also show that traders’ sensitivity to the Bank’s announcements has strengthened since the introduction of the negative interest rate policy in 2016, whereas their sensitivity to economic indicators and surveys has weakened substantially.     

Forecasting election results by studying brand importance in online news

This study uses the semantic brand score, a novel measure of brand importance in big textual data, to forecast elections based on online news. About 35,000 online news articles were transformed into networks of co-occurring words and analyzed by combining methods and tools from social network analysis and text mining. Forecasts made for four voting events in Italy provided consistent results across different voting systems: a general election, a referendum, and a municipal election in two rounds. This work contributes to the research on electoral forecasting by focusing on predictions based on online big data; it offers new perspectives regarding the textual analysis of online news through a methodology which is relatively fast and easy to apply. This study also suggests the existence of a link between the brand importance of political candidates and parties and electoral results.     

Online big data-driven oil consumption forecasting with Google trends

The rapid development of big data technologies and the Internet provides a rich mine of online big data (e.g., trend spotting) that can be helpful in predicting oil consumption — an essential but uncertain factor in the oil supply chain. An online big data-driven oil consumption forecasting model is proposed that uses Google trends, which finely reflect various related factors based on a myriad of search results. This model involves two main steps, relationship investigation and prediction improvement. First, cointegration tests and a Granger causality analysis are conducted in order to statistically test the predictive power of Google trends, in terms of having a significant relationship with oil consumption. Second, the effective Google trends are introduced into popular forecasting methods for predicting both oil consumption trends and values. The experimental study of global oil consumption prediction confirms that the proposed online big-data-driven forecasting work with Google trends improves on the traditional techniques without Google trends significantly, for both directional and level predictions.     

Hedging futures performance with denoising and noise-assisted strategies

Noise processing is very important to improve hedging effectiveness. However, the existing methods are mainly considered from the view of denoising strategy, and the research on noise-assisted strategy is limited. In this paper, a framework that includes both denoising and noise-assisted strategies is proposed to comprehensively analyze the impact of noise proceeding on hedging effectiveness. In detail, the EMD technology is utilized to decompose the futures and spot original returns. Then, the decomposition terms are stepwise removed or added in the opposite way to obtain the denoised and noise-assisted returns. Finally, under the minimum-CVaR framework, the dynamic hedged portfolios based on original and processed returns are constructed to test the hedging effectiveness. Based on the daily prices of CSI300, S&P500, WTI crude oil, and gold futures contract which range from February 9, 2007, to January 10, 2020, the empirical results indicate that both denoising and noise-assisted hedging strategies can decrease CVaR compare with using original return. Furthermore, denoising or adding high-intensity noise has better hedging performance than low-intensity noise, adding uncorrelated noise performs better than adding correlated noise Robustness results by changing confidence level validate the above conclusions.     

Now-casting inflation using high frequency data

This paper proposes a methodology for now-casting and forecasting inflation using data with a sampling frequency which is higher than monthly. The data are modeled as a trading day frequency factor model, with missing observations in a state space representation. For the estimation we adopt the methodology proposed by Bańbura and Modugno (2010). In contrast to other existing approaches, the methodology used in this paper has the advantage of modeling all data within a single unified framework which allows one to disentangle the model-based news from each data release and subsequently to assess its impact on the forecast revision. The results show that the inclusion of high frequency data on energy and raw material prices in our data set contributes considerably to the gradual improvement of the model performance. As long as these data sources are included in our data set, the inclusion of financial variables does not make any considerable improvement to the now-casting accuracy.