Recurrence Analysis of the Economic Behavior of Agents of the Frequency of Online Exchange Rate in the Information Environment

Abstract

The behavior of the socioeconomic agents of the foreign exchange market has always been characterized by insufficient predictability, therefore today researchers do not leave the search for new methods and indicators for their prediction. With the development of the information society and online technologies, one of the manifestations of the behavior of agents in the foreign exchange market has become their semantic-thematic online queries in search engines. Considering that the time series of exchange rates have been sufficiently studied, insufficient attention has been paid to the dynamics of the frequency of requests for exchange rates as to the parameters of the online behavior of foreign exchange market agents. Since the dynamics of the frequency of requests for exchange rates (dollar, euro) is non-linear, a recurrence analysis was applied. The results of this analysis proved that the tendencies of informational activity of agents are characterized by an antipersistent type of behavior with periodicity and the presence of drift, which are manifested in varying degrees for each exchange rate. If for the indicator of the dollar exchange rate a concentration of interest of market agents near the trajectory of a certain attractor with other deviations due to other factors is shown, then a similar indicator of the euro exchange rate is characterized by a more pronounced and stable over time cyclical manifestation of information activity of agents. In further researches, it is planned to implement a quantitative analysis of the constructed recurrence plots in order to take into account the behavior of the agents of the foreign exchange market.

Recurrence analysis of innovation behavior of bitcoin market agents in conditions of COVID-19

“…After the formation of the set of initial data, the next step is the choice of the reasonable delay of the time series (G). Choosing G should take into account the following [18]:…”

“…A significant number of scientific and practical works are devoted to the research of time series of cryptocurrencies, carried out by methods of machine learning and forecasting [3,4], nonlinear autoregression [5], binomial logistic regression [6], recurrent neural network, ARIMA model [7], Bayesian neural networks [8], theory of complex systems [9][10][11][12][13], fractal analysis [9,[14][15][16][17][18][19] and others. However, the identification of the behavior of digital economy agents requires a search for specific parameters that would uniquely determine it.…”

“…The search for such parameters is difficult, since their characterized by weak structuring and significant subjectivity. So, to date, the results of the influence of social networks on the dynamics of the indicators of the digital market [7] have been obtained, including a quantitative relationship between social networks and the bitcoin price [14][15][16], a connection between the parameters of bitcoin and online searches in Google has been confirmed [18], Google queries and exchange rates [18]. Based on the results of the research, the authors confirm the thesis that the indicators of online search and social networks really determine the dynamics of the rate of cryptocurrencies, therefore, this direction requires further research.…”

Research of the Behavior of National Economy Subjects on-Line Frequency Request for Exchange Rates Methods of Analysis of Pseudophase Space

“…Втім, вектор наукових досліджень тільки починає формуватися. Так, наразі доведено зв'язок між он-лайн активністю та соціальноекономічними показниками, зокрема, соціальними мережами та курсом біткоіну [1,2], курсами валют та частотою запитів он-лайн, впливу даних Вікіпедії [3,4] тощо. Тема торкається площини завжди актуальних питань дослідження нелінійної динаміки [5][6][7][8] значно мінливих, але циклічних рядів соціально-економічних показників online та offline.…”

“…Частота запитів за ключовими словами «курс валют» та «євро» майже на рівні. Спостерігаються максимальні пікові значення частоти запитів за ключовим словом «долар» у періоди : грудень 2014 р., серпень 2015 р., січень 2016 р., після значення показника коливалося в певному коридорі, пожвавлення спостерігалося з квітня 2018 р. За результатами досліджень курсу валют та їх частоти запитів он-лайн у роботах [3,4] доведено наявність тісного зв'язку саме у період дестабілізації тенденцій курсу валют. У період, коли динаміка курсу валют досягає певної стабільності, такого тісного зв'язку із частотою їх запиту не спостерігається.…”