CARMA 2018 - 2nd International Conference on Advanced Research Methods and Analytics

Universitat Politècnica de València July 12, 2018 – July 13, 2018

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) aims to become a 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.

URI permanente para esta colecciónhttps://riunet.upv.es/handle/10251/110570

Examinar

Mostrando 1 - 20 de 51

Access status: Abierto ,
Using big data in official statistics: Why? When? How? What for?
(Editorial Universitat Politècnica de València, 2018-09-07) Mazzi, Gian Luigi
[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.
Access status: Abierto ,
Fear, Deposit Insurance Schemes, and Deposit Reallocation in the German Banking System
(Editorial Universitat Politècnica de València, 2018-09-07) Fecht, Falko; Thum, Stefan; Weber, Patrick
[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.
Access status: Abierto ,
Measuring Technology Platforms impact with search data and web scraping
(Editorial Universitat Politècnica de València, 2018-09-07) Blazquez-Soriano, Amparo; Domenech, Josep; García Alvarez-Coque, José María; Facultad de Administración y Dirección de Empresas; Departamento de Economía y Ciencias Sociales; Escuela Técnica Superior de Ingeniería Agronómica y del Medio Natural; Grupo de Investigación de Economía Internacional y Desarrollo
[EN] In recent years, European research policies and priorities in the agricultural sector have been developed through industry-based partnerships sponsored by the European Commission (EC). In 2004, the EC regulated a form of partnership called European Technology Platform (ETP) with the aim to define research agendas that would attract private investment. Monitoring the impact and performance of public policies, such as the implementation of ETPs, is basic for policy-makers. However, assessing the performance of ETPs frequently result into costly efforts given the current lack of indicators to monitor their variety of activities. In addition, since most ETPs have been set up recently it is difficult to assess their results, which are typically revealed after some time and take a considerable amount of time to be captured and processed with traditional methods such as surveys. In this study, we propose to assess the dynamics of ETPs through measures based on online information, given that it is fresh, available in real-time and is a publicly reflect of the activities of organizations. We firstly consider an ETP as an innovation intermediary and define its functions according to innovation literature. Then, we enumerate the particular activities within each function in which the ETP may be involved. To monitor such functions and activities, some indicators based on online data are proposed. This conceptual basis has been put into practice with a particular case study based on the agri-food technology platform “TP Organics”. Preliminary results show that the online-based indicators are able to measure the level of activity of the platform, if its scope is expanding or reducing, and how the importance of the different functions has evolved over time.
Access status: Abierto ,
Mining for Signals of Future Consumer Expenditure on Twitter and Google Trends
(Editorial Universitat Politècnica de València, 2018-09-07) Pekar, Viktor
[EN] Consumer expenditure constitutes the largest component of Gross Domestic Product in developed countries, and forecasts of consumer spending are therefore an important tool that governments and central bank use in their policy-making. In this paper we examine methods to forecast consumer spending from user-generated content, such as search engine queries and social media data, which hold the promise to produce forecasts much more efficiently than traditional surveys. Specifically, the aim of the paper is to study the relative utility of evidence about purchase intentions found in Google Trends versus those found in Twitter posts, for the problem of forecasting consumer expenditure. Our main findings are that, firstly, the Google Trends indicators and indicators extracted from Twitter are both beneficial for the forecasts: adding them as exogenous variables into regression model produces improvements on the pure AR baseline, consistently across all the forecast horizons. Secondly, we find that the Google Trends variables seem to be more useful predictors than the semantic variables extracted from Twitter posts, the differences in performance are significant, but not very large.
Access status: Abierto ,
Fishing for Errors in an Ocean Rather than a Pond
(Editorial Universitat Politècnica de València, 2018-09-07) Wilson, John; Te'eni, Dov
[EN] In the internet age, a proliferation of services appear on the web. Errors in using the internet service or app are dynamically introduced as new devices/interfaces/software are produced and are found to be incompatible with an app that is perfectly good for other devices. The number of users who can detect various errors changes dynamically: for instance, there may be new adopters of the software over time. It may also happen that an old user might upgrade and thus run into new incompatibility errors. Allowing new users and errors to enter dynamically poses considerable modeling and estimation difficulties. In the era of Big Data, methods for dynamically updating as new observations arise are important. Traditional models for detecting errors have generally assumed a finite number of errors. We provide a general model that allows for a procedure for finding maximum likelihood estimators of key parameters where the number of errors and the number of users can change.
Access status: Abierto ,
From Twitter to GDP: Estimating Economic Activity From Social Media
(Editorial Universitat Politècnica de València, 2018-09-07) Indaco, Agustín
[EN] This paper shows how the use of data derived from Twitter can be used as a proxy for measuring GDP at the country level. Using a dataset of 270 million geo-located image tweets shared on Twitter in 2012 and 2013, I find that: (i) Twitter data can be used as a proxy for estimating GDP at the country level and can explain 94 percent of the variation in GDP; and (ii) that the residuals from my preferred model are negatively correlated to a data quality index which assesses the capacity of a country’s statistical system. This suggests that my estimates for GDP are more accurate for countries which are considered to have more reliable GDP data. Taken together, these findings show that institutions and individuals could use social media data to corroborate official GDP estimates; or alternatively for government statistic agencies to incorporate social media data to complement and further reduce measurement errors.
Access status: Abierto ,
Identification of helpful and not helpful online reviews within an eWOM community using text-mining techniques
(Editorial Universitat Politècnica de València, 2018-09-07) Olmedilla, Maria; Martinez-Torres, Rocio; Toral, Sergio L.
[EN] Consumers represent today a significant source of information to learn about products and services quality thanks to the proliferation of user-generated content in the form of online reviews. It is thus of paramount to understand what makes online reviews helpful to consumers as this evaluation might affect their purchase decisions. In this regard, this research has applied textmining techniques by extracting the characteristics from online reviews' texts of an eWOM community, and further utilized these characteristics to train a logistic classifier using three classes: helpful, neutral and not helpful. The aim is identifying which unique attributes determine whether an online review is helpful or not. Findings reveal that there are much more unique attributes classified as helpful than attributes classified as not helpful. Additionally, the unique attributes associated to helpful reviews exhibit more objective appraisal while those associated to not helpful reviews show more subjective appraisal. The proposed methodology can be used to predict the helpfulness of posted reviews and to obtain their unique attributes.
Access status: Abierto ,
Measuring Retail Visual Cues Using Mobile Bio-metric Responses
(Editorial Universitat Politècnica de València, 2018-09-07) Dishman, Paul; Groves, Joshua; Jolley, Dale
[EN] This research provides the results of a comprehensive in-store study that utilized eye tracking to determine the initial eye attractiveness of signage and displays used in a Toyota retail dealership. Potential car buyers (n = 24) walked randomly through the showroom for the first time, and were asked to view the various signs, displays, video monitors, decorations, and vehicles on display. Research was conducted while the dealership was open in order to include distractions from human interaction. Subjects’ eye movements and the objects viewed were captured using Tobii II eye tracking glasses at 60Mhz. A typical showroom self-tour lasted approximately 4:31 minutes. Subjects were then shown their results and Retrospective Think Aloud interviews were conducted with the subjects to determine positive and negative reactions to the observed objects. Signage measured included those required by Toyota, as well as those created by the dealership. Types of signage measured included digital, video, posters, stand-up cards, and ads placed on the vehicle. Each potential eye attractive object was identified and classified by type (signage, décor, digital signage, vehicle information, etc.). Every subject’s results were analyzed by the number of fixations and the time spent viewing each object. The study revealed that video or digital messaging was not any more effective than static signage, but that placement of the signage was a determining factor in the effectiveness of message receptivity. Many of the non-signage objects received more attention than did certain types of advertising signage. The various attributes of the objects and signs that received positive attention were analyzed as to their eye attactiveness characteristics. Although signage in a retail showroom is believed to be critical in providing advertising and product messages, this study (in its particular environment) demonstrated that signage is not viewed by customers as often as previouly thought.
Access status: Abierto ,
Empirical examples of using Big Internet Data for Macroeconomic Nowcasting
(Editorial Universitat Politècnica de València, 2018-09-07) Kapetanios, George; Marcellino, Massimiliano; Papailias, Fotis
[EN] In this paper we present results for nowcasting and one-month-ahead forecasting three key macroeconomic variables: inflation (measured by the month on month growth rate in the Harmonized Index of Consumer Prices), retail sales (measured by the Retail Trade Index), and the Unemployment Rate. The exercise is conducted recursively in a pseudo out of sample framework, using monthly data for three economies: Germany, Italy and the UK. We assess the relative performance of Big Data (proxied via weekly Google Trends) and standard indicators (based on a large set of economic and financial variables). We also evaluate the role of several econometric methods and alternative specifications for each of them (with or without big data), for a total of 279 models and model combinations. In general, we find that Google Trends tend to slightly improve the forecasts of factor models and penalized regressions. Furthermore, a data-driven automated model selection strategy, where the forecasts from a set of best performing models over the recent past are pooled, performs particularly well, with Big Data present in about 65% of the pooled models (on average across the cases where the strategy is the best model).
Access status: Abierto ,
The Catalonian Crises through Google Searches: A Regional Perspective
(Editorial Universitat Politècnica de València, 2018-09-07) Artola, Concha; Pérez, Javier J.
[EN] In this paper we focus in the period of political turmoil starting in September 2017 in Catalonia. Our research question is the following: can the Catalan crisis be tracked by the searches done by the public on different consumption items in the Internet? We do so by focusing in two set of consumption categories: Travel to Catalonia from the main international markets (France, Germany and United Kingdom) and searches on the main consumption categories done from Catalonia and from other five big regions (Madrid, Valencia, Aragón, Andalucía and Basque Country). The preliminary results show that the uncertainty in the political situation has translated unto a decline in searches on terms associated with tourism activities in Barcelona, one broad measure shows that searches for the term “Barcelona hotel” has declined by 12%, year on year for September 2017 to January 2018, by comparison searches for hotel in other comparable Spanish regions have increased slightly. When comparing searches done from Catalonia with other regions through simple time series models, a sizeable negative residual for Catalonia is present in October 2017 –the most difficult period in the Catalan conundrum- which is not observed in other geographical areas. This is the case for some search topics associated to durable goods and Catering and Accommodation services. The political turmoil in Catalonia had significant negative effects in two consumption categories: Theaters and Restaurants.
Access status: Abierto ,
Spread the Word: International Spillovers from Central Bank Communication
(Editorial Universitat Politècnica de València, 2018-09-07) Armelius, Hanna; Bertsch, Christoph; Hull, Isaiah; Zhang, Xin
[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.
Access status: Abierto ,
Macroeconomic Indicator Forecasting with Deep Neural Networks
(Editorial Universitat Politècnica de València, 2018-09-07) Cook, Thomas; Smalter Hall, Aaron
[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).
Access status: Abierto ,
What should a researcher first read? A bi-relational citation networks model for strategical heuristic reading and scientific discovery
(Editorial Universitat Politècnica de València, 2018-09-07) Moreno Pascual, Cesar; Martínez de Ibarreta Zorita, Carlos
[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.
Access status: Abierto ,
Access and analysis of ISTAC data through the use of R and Shiny
(Editorial Universitat Politècnica de València, 2018-09-07) González-Martel, Christian; Cazorla-Artiles, José M.; Pérez-González, Carlos
[EN] The increasing availability of open data resources provides opportunities for research and data science. It is necessary to develope tools that take advantage of the full potential of new information resources. In this work we developed the package for R istacr that provides a collection of eurostat functions to be able to consult and discard the data that Eurostat, including functions to retrieve, download and manipulate the data set available through the ISTAC BASE API of the Canary Institute of Statistics (ISTAC). In addition, A Shiny app was designed for a responsive visulization of the data. This develope is part of the growing demand for open data and ecosystems dedicated to reproducible research in computational social science and digital humanities. With this interest, this package has been included within rOpenSpain, a project that aims to promote transparent research methods mainly through the use of free software and open data in Spain.
Access status: Abierto ,
Italian general election 2018: digital campaign strategies. Three case studies: Movimento 5 Stelle, PD and Lega
(Editorial Universitat Politècnica de València, 2018-09-07) Calò, Ernesto; Faggiano, Maria Paola; Gallo, Raffaella; Mongiardo, Melissa
[EN] The advent of the Network Society has brought substantial transformations also in the politics, which, like other areas of society, is affected by important changes. The network, which regulates social relations, has become the place of political discussion and that is where the most substantial part of the electoral campaign for the 2018 general election took place. The object of our research is the observation of the political propaganda of the Movimento 5 Stelle, the Partito Democratico and the Lega (the three most voted parties in the Italian elections) through the institutional accounts of the political parties on Facebook. Once collected a research sample of 1,397 posts officially published online on the three monitored accounts, the aim of our analysis is to investigate the communication strategies of the parties in a phase of hybrid democracy crossed by a deep crisis of political representation. From our analysis it emerges how the three political forces, that refer to different electorates, organize their electoral propaganda, each according to their own strategy
Access status: Abierto ,
Towards an Automated Semantic Data-driven Decision Making Employing Human Brain
(Editorial Universitat Politècnica de València, 2018-09-07) Fensel, Anna; Österreichische Forschungsförderungsgesellschaft
[EN] Decision making is time-consuming and costly, as it requires direct intensive involvement of the human brain. The variety of expertise of highly qualified experts is very high, and the available experts are mostly not available on a short notice: they might be physically remotely located, and/or not being able to address all the problems they could address time-wise. Further, people tend to base more of their intellectual labour on rapidly increasing volumes of online data, content and computing resources, and the lack of corresponding scaling in availability of the human brain resources poses a bottleneck in the intellectual labour. We discuss enabling direct interoperability between the Internet and the human brain, developing "Internet of Brains", similar to "Internet of Things", where one can semantically model, interoperate and control real life objects. The Web, "Internet of Things" and "Internet of Brains" will be connected employing the same kind of semantic structures, and work in interoperation. Applying Brain Computer Interfaces (BCIs), psychology and behavioural science, we discuss the feasibility of a possible decion making infrastructure for semantic transfer of human thoughts, thinking processes, communication directly to the Internet
Access status: Abierto ,
Technical Sentiment Analysis: Measuring Advantages and Drawbacks of New Products Using Social Media
(Editorial Universitat Politècnica de València, 2018-09-07) Chiarello, Filippo; Bonaccorsi, Andrea; Fantoni, Gualtiero; Ossola, Giacomo; Cimino, Andrea; Dell'Orletta, Felice
[EN] In recent years, social media have become ubiquitous and important for social networking and content sharing. Moreover, the content generated by these websites remains largely untapped. Some researchers proved that social media have been a valuable source to predict the future outcomes of some events such as box-office movie revenues or political elections. Social media are also used by companies to measure the sentiment of customers about their brand and products. This work proposes a new social media based model to measure how users perceive new products from a technical point of view. This model relies on the analysis of advantages and drawbacks of products, which are both important aspects evaluated by consumers during the buying decision process. This model is based on a lexicon developed in a related work (Chiarello et. al, 2017) to analyse patents and detect advantages and drawbacks connected to a certain technology. The results show that when a product has a certain technological complexity and fuels a more technical debate, advantages and drawbacks analysis is more efficient than sentiment analysis in producing technical-functional judgements.
Access status: Abierto ,
A combination of multi-period training data and ensemble methods to improve churn classification of housing loan customers
(Editorial Universitat Politècnica de València, 2018-09-07) Seppälä, Tomi; Thuy, Le
[EN] Customer retention has been the focus of customer relationship management in the financial sector during the past decade. The first and important step in customer retention is to classify the customers into possible churners, those likely to switch to another service provider, and non-churners. The second step is to take action to retain the most probable churners. The main challenge in churn classification is the rarity of churn events. In order to overcome this, two aspects are found to improve the churn classification model: the training data and the algorithm. The recently proposed multi-period training data approach is found to outperform the single period training data thanks to the more effective use of longitudinal data. Regarding the churn classification algorithms, the most advanced and widely employed is the ensemble method, which combines multiple models to produce a more powerful one. Two popularly used ensemble techniques, random forest and gradient boosting, are found to outperform logistic regression and decision tree in classifying churners from non-churners. The study uses data of housing loan customers from a Nordic bank. The key finding is that models combining the multi-period training data approach with ensemble methods performs the best.
Access status: Abierto ,
Validation of innovation indicators from companies’ websites
(Editorial Universitat Politècnica de València, 2018-09-07) Heroux-Vaillancourt, Mikael; Beaudry, Catherine
[EN] In this exploratory study, we use a web mining technique to source data in order to create innovation indicators of Canadian nanotechnology and advanced materials firms. 79 websites were extracted and analysed based on keywords related to the concepts of R&D and intellectual property. To understand what our web mining indicators actually measure, we compare them with those from a classic questionnaire-based survey. Formative indices from the surveys variables were built to better represent all the possibilities resulting from the web mining indicators. A MTMM matrix lead us to conclude that the formative indices are a good representation of the web mining indicators. As a consequence, the data extracted via our web mining technique can be used as proxies for the relative importance of R&D and the importance of IP, which would have previously only been measured using conventional methods such as government administrative data or questionnaire-based surveys.
Access status: Abierto ,
Big Data and Data Driven Marketing in Brazil
(Editorial Universitat Politècnica de València, 2018-09-07) Finger, Vítor; Reichelt, Valesca; Capelli, João
[EN] The main purpose of this article is the understanding of which marketing strategies related to big data are being implemented by Brazilian companies in different sectors, in addition to assessing these actions within an already established construct. To reach the proposed objectives, an exploratory, qualitative research was conducted using the multiple case study method. Thus, data were collected through bibliographical, documentary and semi-structured interviews, with the intent of formulating the construct by which the companies are studied. The study unit interviewed consisted of market professionals and big data specialists. As the main result, it was widely noticed the application of strategies related to big data by the companies surveyed. The classification of these actions within an already established construct, however, was not possible, since it was understood the existence of distinct stages of adoption for this technology, and it was not possible to label these companies as users of big data.

Examinar

Envíos recientes