Introduction

The global financial crisis of 2008 after decades of the great moderation policy governance opened the discussion of the financial stability importance for economic activity. Identification of the earning warning signals to predict business cycles and avoid instability in the economy was always the primary goal of the policymakers. Our paper is the first to combine financial big data to forecast financial cycles.
First to mention the importance of the big data for macroeconomic management was \cite{Phillips1962}. Phillips (1962) put forward the assumption that policymakers can not design effective economic policy without quantitative knowledge behind the main economic forces governing the economic system.  Main economic forces consist of employment, inflation and growth as instruments and policy goals at the same time. Therefore, it is not possible to design an optimal economic policy without knowing the quantitative relations that hold between these economic forces. To be able to measure and identify the correct quantitive relationship among employment, inflation and growth, a big data approach is essential. 
Fluctuations in financial variables always had a significant impact on real economic activities. That is the case today but was all the same 3000 years ago. Oscillatory behaviour in savings and investments dynamics and price targeting articulate economic cycle, see Ji Ran \cite{Milburn2007}. Consequently, policymakers must always monitor the saving-investments cycle keeping prices under control to ensure the smooth running of the economy.  The same concept highlighting the importance of financial conditions for business cycles is extended in \cite{Keynes1883-19461936}\cite{Schumpeter1939}\cite{Minsky1963,Minsky1975}.
Inherent dynamics in financial conditions articulate the dynamics of the business cycles driving economic booms and downturns. Financial markets offer a large number of financial data reflecting financial conditions still the issue of financial cycles definition, measurement and forecasting remain undisclosed. We can understand the complexity of the problem from the fact that today we still miss a generalised definition of financial cycles. General literature uses Minsky's financial instability hypothesis to define financial cycles in the form of the Minsky's cycle. Pioneering works of \cite{Borio2011,Drehmann2012,Drehmann_2013} define financial cycles as long swings in financial conditions (credit and assets price). A large body of research focuses on the issue of finding an optimal method for measuring financial cycles  \cite{Schuler2015b}\cite{Strohsal2019b}\cite{Hiebert2014}\cite{Runstler2018}\cite{Strohsal2019}\citep*{marinko}.  
In this paper, we use quarterly big data for the United Kingdom, the United States, Japan and China from Q1 2004 Q1 2019. We use web-based indicators from the Google trends, financial market data (house price index and credit volume, share in the GDP), long- interest rates on 10-year government treasury bond, consumer and business confidence index and share price index. Using frequency domain time-series methods singular  (SSA) and multichannel singular spectrum analysis (MSSA), we test the impact of big data statistics on financial cycles forecasting. Our results show that policymakers can obtain better macro-aggregates to use in financial cycles forecasting and thus better forecasting accuracy and validity. Assuring better information support and prognosis on financial cycles can help economic agents in the decision-making process on financial markets. In the same time, using relevant micro-financial data improve the relevance of macro-financial aggregates. Financial cycles reflect fluctuations in macro-financial aggregates. Thus, more precise measurement and identification of financial cycles support central bank policies meeting information needs for price and macro-financial stability. Comparing the forecasting results in this study, we can see that the results of the \cite{Diebold1995} test support the hypothesis that more accurate forecasting of the financial cycles can be obtained using financial big data. Big data financial cycles can serve the purpose of developing early warning signals of financial distress as financial crisis predictor. Since more contemporary studies show financial conditions articulate real economic activity dynamics and thus business cycles, understanding the true nature of the financial cycles is of crucial importance for unravelling business cycles. Our study results in financial big data can have an essential role in this attempt.  
The rest of the papers is structured as follows: after the introduction on financial cycles and big data link, literature with interest in financial big data is presented. In section three, we describe data used in the study and methodology we apply to obtain the results on tested hypothesis. Empirical results discussion and implications are presented in section four, while part five bring up conclusions on the role and importance of financial big data for financial and economic stability analysis. 

'Financial Big Data' Implication for Financial Cycles: A Review

The literature researching financial big data importance for monitoring financial conditions and financial cycles just recently show an increasing interest in financial big data. We divide the literature on financial variables importance in three sets. The first set, study financials variables such as savings, investments, assets price and their role in the real economic activity. Theoretical studies on savings and investments cyclical behaviour and importance for economic booms and downturns have a long tradition in writings of Ji Ran \cite{Milburn2007} and Kautilya \cite{ISI:000214562700004}.  Later studies of \cite{Smith1723-17902000}\cite{Keynes1883-19461936}\cite{Schumpeter1939} in the form of their theories explore the impact of financial variables on cyclical economic dynamics. They observe the market economic system present episodes of booms and bust trying to understand the nature behind this cyclical behaviour in the business activity. \cite{Minsky1963,minsky2016can,Minsky1975,Minsky1986,Minsky1977} goes a step further by defining a Minsky's cycle as a consequence of the speculative behaviour of economic agents on the financial markets (financial instability hypothesis), see \cite{Palley2011a}. As pointed out by \cite{Palley2011a}, financial cycles work in the form of Minsky's 'super-cycles' lasting over several business cycles at a system-level generally ending in financial crisis. 
The second set of literature explores the oscillatory behaviour of financial cycles in various countries. 
According to the research of \citep{Drehmann2012}\citep{Drehmann_2013}\cite{Scharnagl2016}\cite{Strohsal2019b}\cite{Galati2016b}\cite{Claessens2014a} financial cycles exhibit a medium-term pattern (ranging from 10-19 years). Cycles show the tendency to be longer and with a larger amplitude after the post-1985 period \citep{Drehmann2012}.  Studies above, from a large set of financial variables, isolate three primary financial cycles determinants. According to their study results, financial cycles can be approximated using financial data from property markets (housing prices) and credit markets (volume of credits and credits share in the GDP). Financial cycles can be measured using data on residential property prices, real credit and credit-to-GDP ratio \citep{Drehmann2012}. Measurement methods on financial cycles range from frequency to time-domain analysis. \cite{claessens2011financial,Claessens2012}  and \citep{Drehmann2012} adapt the turning points methods developed by \citep*{bry1971cyclical}\cite{Harding2002} to measure financial cycles. \citet{Strohsal2019b} use the indirect estimation of spectral densities to isolate financial cycles in the United Kingdom (UK), United States (US) and Germany.  Their results show strong statistical evidence for the existence of financial cycles (7-15 years) in the US. Weaker statistical evidence supports the existence of a financial cycle in the UK and limited evidence for Germany is found. 
\citet{Drehmann2012}\citep{Aikman2015} measure financial cycles using frequency-based filters advanced by \citep*{Baxter1999} and \citep*{Christiano2003}.  Their study isolates financial cycles in Australia, Germany, Japan, Norway, Sweden, the United Kingdom and the United States) over the period 1960-2011. Results support the existence of the financial cycles in the medium term, lasting over the whole sample 16 years on average. \citet{Galati2016b} use the unobserved component time series model to isolate financial cycles in the US, Germany, France, Italy, Spain and the Netherlands over 1970-2014. Empirical results from the study show the existence of medium-term financial cycles lasting on average from 8 to 25 years. \citet*{Stremmel2015} study financial cycles for 11 European countries using seven different potential financial cycles components. His results show financial cycles can be best approximated by looking at Credit to GDP ratio, House prices to income ratio and Credit growth dynamics. 
\citet{Schuler2015} use multivariate and time-varying approach for 13 European Union economies from 1970 to 2013. Empirical results of their study support previous research on financial cycles finding financial cycles exist in medium-term frequencies from 8 to 20 years. \citet{RePEc:bis:biswps:755} use data over 120 years in the US using business cycle methodology to model financial cycle linking financial cycle time to calendar time observations. 
The impact of financial cycles on real economic activity (business cycles) attracted much attention in contemporary studies (third set of literature).  Studies of \cite{Borio2014}\citep*{Braun2005a}\citep{Claessens2009a}\cite{Paccagnini2019}\citep*{Borio_2011}\citep{Borio2018}\citep{Nowotny2014} explore the link between financial and business cycles. Empirical results from the studies show financial and business cycles are highly synchronised. Financial cycles are generally longer in relation to the business cycles, in fact, they last twice long as an average business cycle. Since financial cycles usually last as several business cycles, it is important to find if the financial cycle increases the recession risk or cause secular stagnation \citep{Borio2018}\cite{Borio2017a}
The literature on the importance of financial big data for financial cycles measurement is novel and scarce. \citet*{Hassani2015} offer a review of the pros and cons of using big data for the purpose of forecasting. \citet{alessi2009forecasting} study show large macroeconomic and financial data improve forecast performance in finance and economics. \citet{Altissimo2010} provide evidence of using big data to improve the forecast accuracy when using band-pass filters. Using data from the google trends in the study of \citet{Choi2012} improve the forecasting prediction for economic indicators. \citet{gupta2014using}  forecast employment using classical and Bayesian methods applied to eight sectors for the US economy with 143 monthly series. Forecast model using big data information outperform other models in forecast accuracy. 
Large data sets (big data) availability is essential but not sufficient condition for improving forecast accuracy or model predictions \cite{silver2012signal}\citet{Banbura2014} find forecasting with large datasets more complex for the purpose of signal extraction. 
To fully understand financial cycles we need big data on the micro and macro level. \citet{wibisonouse} 
define big data as a large volume of information resulting from the web, financial, administrative and commercial records. They find the financial big data phenomenon as a new opportunity for decision-making process improvement (complete, immediate and granural information adding new value to 'traditional' macroeconomic indicators). Therefore, big data analytics and artificial intelligence offer new opportunities to central banks and economic agents on the financial markets. Opportunities but challenges as well (resources availability, unresolved changing decision-making processes, best policy introduction). \citet*{subrahmanyam2019big} review the body of literature on big data in the finance application.  Big data application in finance ranges from the basic application as in \cite{TETLOCK_2007} to complex \citep{Tetlock2008}. According to their study, text analysis in the financial press capture hard-to-quantify aspects of firms' fundamental supporting robust low firm' earning forecasting. Big data based on quantifying language application can help to improve methods for measuring firms' fundamentals or forecast market trends. Big data in finance also help to fill in the gap between qualitative and quantitative data application \cite{Huang_2018}. His study results show big data on consumers opinions (from Amazon. com) when used provide a more robust prediction about firms' cash flow and stock pricing.  \citet{Chordia_2018}  show that using newsfeed data on macroeconomic conditions only minorly increase profits from fast trading. According to the study of \citep{Fan_2012}, using high-frequency financial data improve the efficiency and stability of portfolio selection. The literature on big data application for policy purpose is much more limited \citep[see][]{Tissot2019}. He lists the advantages and disadvantages of using financial big data to obtain macro-relevant micro information, better macro aggregates to improve policy design and assessment. 
Financial big data opened its way to portfolio and decision making processes on the micro-level and it is just moving (slowly) on the macro-level (central banks and financial authorities). 

Dataset and Stylized Facts on Financial Cycles in the UK, USA, Japan and China 2005 - 2019

Our paper is first to combine financial big data for the purpose of financial cycles predictions. We follow  \citet{wibisonouse} definition of financial big data and use the following variables: 
We use data for the United Kingdom (UK), United States (US), Japan (Jap) and China (CHI) from Q1 2004 to Q1 2019 (quarterly data).  Previous studies on financial cycles use data on residential property prices, credits to private non-financial sectors and credits to private non-financial sectors share in the GDP (gross domestic product). Facing limited data availability on financial conditions and web we had to choose between longer time series and a variety of data sources. For the purpose of testing the importance of financial big data in financial cycles prediction, we use data from various sources: web-based indicators - social networks (financial crisis search term importance), financial market data (share prices, residential property prices, credits), administrative data (consumer confidence index, business confidence index, long-term interest rates), commercial data sets (government bonds yields). 
To measure financial cycle components, we use spectral analysis techniques - Eviews 10.0 adds-in for Spectral analysis \citet*{Ronderos2014}. Prior to analysis, we normalize and centre all the series for the purpose of eliminating noise in the series. Since normalization procedure increases the possibility of spurious cycles existence in the series, we apply \citet*{Ronderos2014} significant pass filter (SPF) method of the form:
\(F_{t}=\sum_{k=0}^{n / 2}\left[A_{k} \cos \left(\omega_{k} t\right)+B_{k} \sin \left(\omega_{k} t\right)\right]\)(1)
with Ft = filtered series, \((A_{k}=a_{k}), (B_{k}=b_{k}) ordinate (I\left(\omega_{k}\right))\) to test statistical significance of the deterministic component in the spectrum.