[en] Research methods in economics and social sciences are evolving with the increasing availability of Internet and Big Data sources of information. As these sources, methods, and applications become more interdisciplinary, the 2nd International Conference on Advanced Research Methods and Analytics (CARMA) is an excellent forum for researchers and practitioners to exchange ideas and advances on how emerging research methods and sources are applied to different fields of social sciences as well as to discuss current and future challenges.

[en] Recent regulatory initiatives such as the European Deposit Insurance Scheme propose a change in the coverage and backing of deposit insurances. An assessment of these proposals requires a thorough understanding of what drives depositors' withdrawal decisions. We show that Google searches for 'deposit insurance' and related strings reflect depositors' fears and help to predict deposit shifts in the German banking sector from private banks to fully guaranteed public banks. After the introduction of blanket state guarantees for all deposits in the German banking system this fear driven reallocation of deposits stopped. Our findings highlight that a heterogeneous insurance of deposits can lead to a sudden, fear induced reallocation of deposits endangering the stability of the banking sector even in absence of redenomination risks.

[en] This paper analyses the potential usefulness of big data in official statistics starting from four key questions such as Why? When? How? and What for - should we use big data in official statistics? To derive some answers related to empirical cases. This paper presents a big data classification by types, which is then used to identify how big data can answer to specific information needs in key policy areas. Based on the findings of these investigations, some very provisional and subjective answers to the questions raised above are derived. Keywords: Big data, nowcasting, indicators, policy areas.

[en] We use computational linguistic methods and a novel dataset to measure the sentiment component of central bank communications in 23 countries over the 2002-2016 period. We first construct a Granger causality network to identify how sentiment is transmitted across central banks. The network structure suggests that comovement in sentiment is not reducible to comovement in output across countries. We also show that some central banks in the network, such as the Federal Reserve and the Bundesbank, tend to cause sentiment shifts in other central banks; whereas other central banks, such as the European Central Bank and the Bank of Japan, tend to be shifted by other central banks. Finally, we use a structural VAR to demonstrate that sentiment shocks generate cross-country spillovers in sentiment, policy rates, and real variables.

[en] Scientists usually try to find relevant and updated documents for their research. Also, they face an abundance of information. Most of the methodologies and algorithms look Backwards, so they suffer an inevitable time delay. We propose a recommendation algorithm combining Forward and Backward citation entire networks and Macro, Meso and Micro metrics that concludes in a strategic map and a heuristic reading path. Underlying it, we found an asymmetric bowtie scientific advance model that informs all, solving the abundance problem with a triple reduction and a heuristic reading path.

[en] Economic policymaking relies upon accurate forecasts of economic conditions. Current methods for unconditional forecasting are dominated by inherently linear models that exhibit model dependence and have high data demands. We explore deep neural networks as an opportunity to improve upon forecast accurac y with limited data and while remaining agnostic as to functional form. We focus on predicting civilian unemployment using models based on four different neural network architectures. Each of these models outperforms benchmark models at short time horizons. One model, based on an Encoder Decoder architecture outperforms benchmark models at every forecast horizon (up to four quarters).

[en] In this paper we analyze five big data algorithmic trading systems based on artificial intelligence models that uses as predictors stats from Google Trends of dozens of financial terms. The systems were trained using monthly data from 2004 to 2017 and have been tested in a prospective way from January 2017 to February 2018. The performance of this systems shows that Google Trends is a good metric for global Investors' Mood. Systems for Ibex and Eurostoxx are not profitable but Dow Jones, S&P 500 and Nasdaq systems has been profitable using long and short positions during the period studied. This evidence opens a new field for the investigation of trading systems based on big data instead of Chartism.

[en] A typical analytical lifecycle in data science projects starts with the process of data generation and collection, continues with data preparation and preprocessing and heads towards project specific analytics, visualizations and presentations. In order to ensure high quality trusted analytics, every relevant step of the data-model-result linkage needs to meet certain quality standards that furthermore should be certified by trusted quality gate mechanisms. We propose “blockchain-backed analytics”, a scalable and easy-to-use generic approach to introduce quality gates to data science projects, backed by the immutable records of a blockchain. For that reason, data, models and results are stored as cryptographically hashed fingerprints with mutually linked transactions in a public blockchain database. This approach enables stakeholders of data science projects to track and trace the linkage of data, applied models and modeling results without the need of trust validation of escrow systems or any other third party.

[en] Following recent research, we explore virtually contemporaneous, and geographically granular, user online activity related to entrepreneurship. In this paper, we present evidence that data harvested by Dstillery can complement efforts of, and data collected by, government agencies and organizations advocating for entrepreneurship, business formation and economic growth, e.g., the Kauffman Foundation. Our website-based behavior data is close to real time and at a geographically granular level. We find that the concentration of a region’s visits to website resources for entrepreneurship and business development are statistically related to business start-up and, particularly, growth activity. Visits to websites related to entrepreneurship are more strongly associated with growth entrepreneurship, in contrast to start-up entrepreneurship. While data capture and analysis related to entrepreneurship website activity is in its infancy, this analysis points to the potential of this data source to nowcast business formation and growth at a regional level.

[en] With the wide application of the smart card technology in public transit system, traveller’s daily travel behaviours can be possibly obtained. This study devotes to investigating the pattern of individual mobility patterns and its relationship with social-demographics. We first extract travel features from the raw smart card data, including spatial, temporal and travel mode features, which capture the travel variability of travellers. Then, travel features are fed to various supervised machine learning models to predict individual’s demographic attributes, such as age group, gender, income level and car ownership. Finally, a case study based on London’s Oyster Card data is presented and results show it is a promising opportunity for demographic study based on people’s mobility behaviour.