Testing big data in a big crisis: Nowcasting under Covid-19

Barbaglia, Luca; Frattarolo, Lorenzo; Onorante, Luca; Pericoli, Filippo Maria; Ratto, Marco; Pezzoli, Luca Tiozzo

doi:10.1016/j.ijforecast.2022.10.005

Cited by 10 publications

(8 citation statements)

References 53 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has been highlighted that during the pandemic, policymakers need short-term predictions and nowcasts of the current state of the economy to design timely policy actions and assess the success of various activities in mitigating the pandemic's negative effects (Barbaglia et al, 2022;Ferrara et al, 2022), in protecting societal well-being, and also to have a better understanding of citizens evolving concerns and information needs (World Health Organisation, 2021). Big Data have become an important support system in dealing with global healthcare crisis as well as a tool for economic sustainability, and social well-being during period of crisis.…”

Section: Discussionmentioning

confidence: 99%

“…(Xu et al, 2022), and create the need to manage and monitor these data in order to obtain relevant information (Piccarozzi & Aquilani, 2022) as well as to manage the changing needs of many economic sectors. Moreover, for COVID-19 pandemic no past observations could be used to provide a relevant signal about its potential economic (Barbaglia et al, 2022), and not only, impact.…”

Section: Big Data and Covid-19 Related Analysismentioning

confidence: 99%

See 1 more Smart Citation

Big Data, Sentiment Analysis, and Examples during the COVID-19 Pandemic

2022

View full text Add to dashboard Cite

Applied research in Big Data has gained popularity and is already transforming corporations, public sector, health care and subsequently everyday life. Big Data are being analysed for a variety of reasons, e.g., predict Brexit negotiating outcomes, optimise operations in agriculture, map and analyse human mobility trends under non-pharmaceutical interventions during the recent pandemic. The period of the COVID-19 pandemic has been characterised also by an ‘infodemic’, meaning an overabundance of both good and bad information. This information needs to be managed effectively as it can yield valuable insights when analysed. In this paper the terms of Big Data, Geospatial Big Data and Sentiment Analysis are presented along with selected cases, from the international literature, of the use of Big Data and analytics during the COVID-19 pandemic.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Big Data and Covid-19 Related Analysismentioning

confidence: 99%

Big Data, Sentiment Analysis, and Examples during the COVID-19 Pandemic

2022

View full text Add to dashboard Cite

show abstract

“…The data was collected for a year, from January 1 to December 31, 2021. The reason for this data scrapping duration was due to the effect from COVID-19 pandemic season that happened since end of 2019 and at its peak throughout the year 2021, thus, the data are more adequate to explore the economic impact through the public's sentiment, particularly towards cryptocurrency-related matter [43], [44]. During this period, social media and online news are optimally used due to Movement Control Order (MCO) that restrict people from going out.…”

Section: Methodsmentioning

confidence: 99%

Crypto-sentiment Detection in Malay Text Using Language Models with an Attention Mechanism

Zamani,

Kamaruddin

2023

J. Inf. Syst. Eng. Bus. Intell.

View full text Add to dashboard Cite

Background: Due to the increased interest in cryptocurrencies, opinions on cryptocurrency-related topics are shared on news and social media. The enormous amount of sentiment data that is frequently released makes data processing and analytics on such important issues more challenging. In addition, the present sentiment models in the cryptocurrency domain are primarily focused on English with minimal work on Malay language, further complicating problems. Objective: The performance of the sentiment regression model to forecast sentiment scores for Malay news and tweets is examined in this study. Methods: Malay news headlines and tweets on Bitcoin and Ethereum are used as the input. A hybrid Generalized Autoregressive Pretraining for Language Understanding (XLNet) language model in combination with Bidirectional-Gated Recurrent Unit (Bi-GRU) deep learning model is applied in the proposed sentiment regression implementation. The effectiveness of the proposed sentiment regression model is also investigated using the multi-head self-attention mechanism. Then, a comparison analysis using Bidirectional Encoder Representations from Transformers (BERT) is carried out. Results: The experimental results demonstrate that the number of attention heads is vital in improving the XLNet-GRU sentiment model performance. There are slight improvements of 0.03 in the adjusted R2 values with an average MAE of 0.163 (Malay news) and 0.174 (Malay tweets). In addition, an average RMSE of 0.267 and 0.255 were obtained respectively for Malay news and tweets, which show that the proposed XLNet-GRU sentiment model outperforms the BERT sentiment model with lesser prediction errors. Conclusion: The proposed model contributes to predicting sentiment on cryptocurrency. Moreover, this study also introduced two carefully curated Malay corpora, CryptoSentiNews-Malay and CryptoSentiTweets-Malay, which are extracted from news and tweets, respectively. Further works to enhance Malay news and tweets corpora on cryptocurrency-related issues will be expended with implementing the proposed XLNet Bi-GRU deep learning model for greater financial insight. Keywords: Cryptocurrency, Deep learning model, Malay text, Sentiment analysis, Sentiment regression model

show abstract

“…The macroeconomic forecasting itself during crises is a challenging task, much more complex than in normal times (Ferrara and Sheng 2022). The economic shock represents an unexpected and unprecedented reaction of the economy to the changes, and no past observations could provide a relevant signal about its potential economic impact (Barbaglia et al 2022). Furthermore, the uncertainty around government restrictions and policy support made it very difficult to assess their impact on national economies (Ferrara and Sheng 2022).…”

Section: Nowcasting Economic Activity Under Uncertain Timementioning

confidence: 99%

“…Many challenges remain for nowcasting during uncertain times (Barbaglia et al 2022;Huber et al 2023); however, they can be divided into two broad categories: (a) the new massive and high-frequency alternative datasets and (b) associated models for forecasting. Usually, the nowcasting challenges with and without uncertain times aspect are similar, however, in a different scale.…”

Section: Nowcasting Economic Activity Under Uncertain Timementioning

confidence: 99%

Nowcasting Economic Activity Using Electricity Market Data: The Case of Lithuania

et al. 2023

View full text Add to dashboard Cite

Traditional forecasting methods usually rely on historical macroeconomic indicators with significant delays. To address this problem, new opportunities for economic modeling and forecasting are emerging by using real-time data and making nowcasting of economic activity. This research aims to assess the usefulness of electricity market data to nowcast the economic activity in Lithuania. Various MIDAS regression models are used to nowcast nine monthly macroeconomic indicators. In general, electricity market indicators are useful to nowcast certain macroeconomic indicators. Electricity consumption is the most useful among electricity market indicators and brings benefits when nowcasting imports, industrial production, consumer confidence, wholesale and retail trade, and the repair of motor vehicles and motorcycles. Electricity production is beneficial in nowcasting the industrial production. Meanwhile, electricity price is useful for nowcasting exports, exports of goods of Lithuanian origin, imports, and industrial production. Meanwhile, electricity market data do not improve the prediction of the unemployment rate, economic sentiment indicator, and CPI-based consumer price in comparison with an autoregressive model.

show abstract

Testing big data in a big crisis: Nowcasting under Covid-19

Cited by 10 publications

References 53 publications

Big Data, Sentiment Analysis, and Examples during the COVID-19 Pandemic

Big Data, Sentiment Analysis, and Examples during the COVID-19 Pandemic

Crypto-sentiment Detection in Malay Text Using Language Models with an Attention Mechanism

Nowcasting Economic Activity Using Electricity Market Data: The Case of Lithuania

Contact Info

Product

Resources

About