Google Trends Topic-Based Uncertainty: A Multi-National Approach

Several studies have shown that uncertainty among economic actors influences business cycle dynamics. This paper uses Google Trends topic queries to construct an uncertainty proxy that can be applied to every country where Google is active. Using a VAR approach, this paper demonstrates that the obtained impulse-response functions of main economic indicators to a one-standard deviation shock to the constructed indicator, are similar to those from an already-existing uncertainty proxy, the EPU. This is true for the G7 countries and Russia. On average, the uncertainty indicator constructed for this paper leads to more statistically significant responses than does the EPU. Thus, this paper shows that Google Trends is a helpful tool for obtaining timely information about uncertainty among economic actors. The main improvement in this uncertainty proxy is in its language independence. Existing uncertainty-measurement approaches, in contrast, rely on certain keywords that often vary across countries.

Investigating inefficiencies of bookmaker odds in football using machine learning

The efficient-market hypothesis states that it is impossible to beat the market, as the price reflects all available information. Applied to bookmaker odds for football games, there should not be a systematic way of winning money on the long run.However, we show that by using simple machine learning models we can systematically outperform the markets belief manifested through the bookmakers odds. The effect of this inefficiency is diminishing over time, which indicates that the knowledge that has been derived from and the pure amount of the data is also reflected in the odds in recent times.We give some insights how this effect differs across major football leagues in Europe, which algorithms are performing best and statistics on the ROI using machine learning in football betting. Additionally, we share how the simulation study has been designed in more detail.

Political Polarization and Movie Ratings: Web Scraping The Brazilian Contemporary Scenario

Our work aims to analyze the impact of political polarization on movieratings at the IMDB platform. For that we explore the concepts of Word ofMouth and Buzz marking the important role they play on polarized opinionsin movie ratings.We develop a code on python to perform web-scrapping onthe sample scope of Brazilian movies and interpret the data collected using acontroversiality index based on standard deviation. The outcome sheds somelight into the relation between Buzz and Controversiality within theframework of Brazil’s current political scenario.

Proposal of a composite indicator for measuring social media presence in the wine market

Cybermetrics field is attracting considerable interest due to its utility as a data-oriented technique for research, though it may provide misleading information when used in complex systems. This paper outlines a new approach to market research analysis through the definition of composite indicators for cybermetrics, applied to the Spanish wine market. Our findings show that the majority of cellars were present in only one or two social media networks: Facebook, Twitter or both. Besides, the presence on the Web can be summarized into three principal components: website quality, presence on Facebook, and presence on Twitter. Three groups of cellars were identified according to their position in these components: cellars with a high number of errors in their website with complete absence of information in social media, cellars with strong presence in social media, and cellars in an intermediate position. Our results constitute an excellent initial step towards the definition of a methodology for building composite indicators in cybermetrics. From a practical approach, these indicators may encourage cellar managers to make better decisions towards their transition to the digital market.

Data granularity in mid-year life table construction

Life tables have a substantial influence on both public pension systems andlife insurance policies. National statistical agencies construct life tables fromhypotheses death rate estimates to the (mx aggregated ), or death figures probabilities of demographic (q x ), after applying events (deaths, variousmigrations and births). The use of big data has become extensive acrossmany disciplines, including population statistics. We take advantage of thisfact to create new (more unrestricted) mortality estimators within the familyof period-based estimators, in particular, when the exposed-to-riskpopulation is computed through mid-year population estimates. We useactual data of the Spanish population to explore, by exploiting the detailedmicrodata of births, deaths and migrations (in total, more than 186 milliondemographic events), the effects that different assumptions have oncalculating death probabilities. We also analyse their impact on a sample ofinsurance product. Our results reveal the need to include granular data,including the exact birthdate of each person, when computing period mid-year life tables.

Mining News Data for the Measurement and Prediction of Inflation Expectations

In this paper we use high frequency multidimensional textual news data andpropose an index of inflation news. We utilize the power of text mining and itsability to convert large collections of text from unstructured to structured formfor in-depth quantitative analysis of online news data. The significantrelationship between the household’s inflation expectations and news topics isdocumented and the forecasting performance of news-based indices isevaluated for different horizons and model variations. Results suggest that withoptimal number of topics a machine learning model is able to forecast theinflation expectations with greater accuracy than the simple autoregressivemodels. Additional results from forecasting headline inflation indicate that theoverall forecasting accuracy is at a good level. Findings in this paper supportthe view in the literature that the news are good indicators of inflation and areable to capture inflation expecta-tions well.

Donald Trump, investor attention and financial markets

Information attracts attention but attention is costly. Social media has been at the forefront ofinformation dissipation due to the sheer number of users propagating information in a fast but cheap way. We look into one specific case where Donald Trump’s tweets on companies have had effect on retail investors whose only source of information is internet. We find that retail investor attention spike as indicated by surge in Google Search Volume Index following Donald Trump’s tweet, irrespective of the tone in the tweet. We also find that Trump’s tweet facilitates wealth transfer due to selling from the retail investors followed by buying by the institutional investors in low retail investor attention environment. Finally, we see no effect in intra-day returns for the stocks irrespective of the attention they are receiving.

Setting Crunchbase for Data Science: Preprocessing, Data Integration and Feature Engineering

In order to support equity investors in their decision-making process, researchers are exploring the potential of machine learning algorithms to predict the financial success of startup ventures. In this context, a key role is played by the significance of the data used, which should reflect most of the variables considered by investors in their screening and evaluation activity. This paper provides a detailed description of the data management process that can be followed to obtain such a dataset. Using Crunchbase as the main data source, other databases have been integrated to enrich the information content and support the feature engineering process. Specifically, the following sources has been considered: USPTO PatentsView, Kauffman Indicators of Entrepreneurship, Academic Ranking of World Universities, CB Insights ranking of top-investors. The final dataset contains the profiles of 138,637 US-based ventures founded between 2000 and 2019. For each company the elements assessed by equity investors have been analyzed. Among others, the following specific areas were considered for each company: location, industry, founding team, intellectual property and funding round history. Data related to each area have been formalized in a series of features ready to be used in a machine learning context.

Communicating Corporate Social Responsibility through Twitter: a topic model analysis on selected companies

Social media are fundamental in creating new opportunities for firms and they represent a relevant tool for the communication and the engagement with customers. The purpose of this paper is to analyse the communication of Corporate Social Responsibility (CSR) activities on Twitter. We consider the listed companies included in the Dow Jones Industrial Average Index and we implement a topic model analysis on their timelines. In order to identify the topic discussed, their correlation, and their evolution over time and sectors,we apply the Structural Topic Model algorithm, which allows estimating the model including document-level metadata. This model proves to be a powerful tool for topic detection and for estimating the effects of document-level metadata. Indeed, we find that the topics are overall well identified, and the model allows catching signals from the data. Finally, we discuss issues related to the validity of the analysis, including data quality problems.

Comparative multivariate forecast performance for the G7 Stock Markets: VECM Models vs deep learning LSTM neural networks

The prediction of stock prices dynamics is a challenging task since these kind of financial datasets are characterized by irregular fluctuations, nonlinear patterns and high uncertainty dynamic changes.The deep neural network models, and in particular the LSTM algorithm, have been increasingly used by researchers for analysis, trading and prediction of stock market time series, appointing an important role in today’s economy.The main purpose of this paper focus on the analysis and forecast of the Standard & Poor’s index by employing multivariate modelling on several correlated stock market indexes and interest rates with the support of VECM trends corrected by a LSTM recurrent neural network.