Применение Google Trends для прогнозирования миграции из России: агрегация поисковых запросов и учет лаговой структуры

Purpose: To analyze the predictability of Google's search queries in the Brazilian financial market. Originality/gap/relevance/implications: Despite a growing foreign literature using Google's search query data, there is no acknowledgement of work on this area in Brazil. An application to the Brazilian financial market shows new sources of information about market movements and may contribute to researchers and practitioners to understand how changes in specific search queries affect the market. Key methodological aspects: Following previous studies, we estimate VAR models and Granger causality tests to investigate the effects over three variables in both stock and fixed income markets: traded volume, return and volatility. Following this procedure, we verify both the hypothesis of financial variables being affected by search queries, as well as the opposite relationship. Weekly data from Google's search queries and financial markets was gathered for the period between 2007 and 2014. Summary of key results: The existence of a predictive effect between search query data and financial variables, particularly in the stock market, is evident. However, this result was not robust in all cases studied. It is noteworthy that, for the inverse relationship, i.e. financial market impacting search queries on Google, strong evidence of a causal relationship has been found. A trading strategy based on this type of data yielded higher returns than the defined benchmarks. Key considerations/conclusions: A significant relationship between Google's search query data and the financial market has been discovered. Results provide a new source of information that affects the Brazilian financial market.

The ability to estimate the current mood states of web users has considerable potential for realizing user-centric opportune services in pervasive computing. However, it is difficult to determine the data type used for such estimation and collect the ground truth of such mood states. Therefore, we built a model to estimate the mood states from search-query data in an easy-to-collect and non-invasive manner. Then, we built a model to estimate mood states from mobile sensor data as another estimation model and supplemented its output to the ground-truth label of the model estimated from search queries. This novel two-step model building contributed to boosting the performance of estimating the mood states of web users. Our system was also deployed in the commercial stack, and large-scale data analysis with >11 million users was conducted. We proposed a nationwide mood score, which bundles the mood values of users across the country. It shows the daily and weekly rhythm of people's moods and explains the ups and downs of moods during the COVID-19 pandemic, which is inversely synchronized to the number of new COVID-19 cases. It detects big news that simultaneously affects the mood states of many users, even under fine-grained time resolution, such as the order of hours. In addition, we identified a certain class of advertisements that indicated a clear tendency in the mood of the users who clicked such advertisements.

Annual suicide figures are critical in identifying trends and guiding research, yet challenges arising from significant lags in reporting can delay and complicate real-time interventions. In this paper, we utilized Google Trends search volumes for behavioral forecasting of national suicide rates in Ireland between 2004 and 2015. Official suicide rates are recorded by the Central Statistics Office in Ireland. While similar investigations using Google trends data have been carried out in other jurisdictions (e.g., United Kingdom, United Stated of America), such research had not yet been completed in Ireland. We compiled a collection of suicide- and depression-related search terms suggested by Google Trends and manually sourced from the literature. Monthly search rate terms at different lags were compared with suicide occurrences to determine the degree of correlation. Following two approaches based on vector autoregression and neural network autoregression, we achieved mean absolute error values between 4.14 and 9.61 when incorporating search query data, with the highest performance for the neural network approach. The application of this process to United Kingdom suicide and search query data showed similar results, supporting the benefit of Google Trends, neural network approach, and the applied search terms to forecast suicide risk increase. Overall, the combination of societal data and online behavior provide a good indication of societal risks; building on past research, our improvements led to robust models integrating search query and unemployment data for suicide risk forecasting in Ireland.

Seasonal trends in restless legs symptomatology: evidence from Internet search query data

Veterinary Education and Problem-based Learning

Varicella is usually a mild disease in children but may be life-threatening, especially in adolescents and adults. Infection control measures implemented during the Coronavirus Disease 2019 (COVID-19) pandemic may have suppressed varicella transmission, potentially creating an 'immunity debt', particularly in countries without universal varicella vaccination. To assess trends in Google search engine queries for varicella keywords as a proxy for varicella infection rates and to evaluate the effect of universal varicella vaccination on these trends. A further objective was to assess the impact of the COVID-19 pandemic on varicella keyword search query trends in countries with and without universal varicella vaccination. This study used the keyword research tool, Google Trends, to evaluate trends in time series of the relative search query popularity of language-specific varicella keywords in 28 European countries from January 2015 through December 2021. The Google Ads Keyword Planner tool was used to evaluate absolute search volumes from March 2018 through December 2021. The relative search query popularity of varicella keywords displayed marked seasonal variation. In all 28 countries, the relative search query popularity of varicella keywords declined after the start of the COVID-19 pandemic (March 2020), compared with pre-pandemic levels (range, -18% to -70%). From April 2020 to July 2021, a period of intense COVID-19 transmission and infection control, absolute search volumes for varicella keywords were lower than pre-pandemic levels but rebounded after July 2021, when infection control measures were relaxed. This evaluation of search query trends demonstrated that search query data could be used as a proxy for trends in varicella infection rates and revealed that transmission of varicella may have been suppressed during the COVID-19 pandemic. Consideration should be given to using search query data to better understand the burden of varicella, particularly in countries where surveillance systems are inadequate.

BackgroundAs suggested as early as in 2006, logs of queries submitted to search engines seeking information could be a source for detection of emerging influenza epidemics if changes in the volume of search queries are monitored (infodemiology). However, selecting queries that are most likely to be associated with influenza epidemics is a particular challenge when it comes to generating better predictions.ObjectiveIn this study, we describe a methodological extension for detecting influenza outbreaks using search query data; we provide a new approach for query selection through the exploration of contextual information gleaned from social media data. Additionally, we evaluate whether it is possible to use these queries for monitoring and predicting influenza epidemics in South Korea.MethodsOur study was based on freely available weekly influenza incidence data and query data originating from the search engine on the Korean website Daum between April 3, 2011 and April 5, 2014. To select queries related to influenza epidemics, several approaches were applied: (1) exploring influenza-related words in social media data, (2) identifying the chief concerns related to influenza, and (3) using Web query recommendations. Optimal feature selection by least absolute shrinkage and selection operator (Lasso) and support vector machine for regression (SVR) were used to construct a model predicting influenza epidemics.ResultsIn total, 146 queries related to influenza were generated through our initial query selection approach. A considerable proportion of optimal features for final models were derived from queries with reference to the social media data. The SVR model performed well: the prediction values were highly correlated with the recent observed influenza-like illness (r=.956; P<.001) and virological incidence rate (r=.963; P<.001).ConclusionsThese results demonstrate the feasibility of using search queries to enhance influenza surveillance in South Korea. In addition, an approach for query selection using social media data seems ideal for supporting influenza surveillance based on search query data.

BackgroundIndia is home to 20% of the world’s suicide deaths. Although statistics regarding suicide in India are distressingly high, data and cultural issues likely contribute to a widespread underreporting of the problem. Social stigma and only recent decriminalization of suicide are among the factors hampering official agencies’ collection and reporting of suicide rates.ObjectiveAs the product of a data collaborative, this paper leverages private-sector search engine data toward gaining a fuller, more accurate picture of the suicide issue among young people in India. By combining official statistics on suicide with data generated through search queries, this paper seeks to: add an additional layer of information to more accurately represent the magnitude of the problem, determine whether search query data can serve as an effective proxy for factors contributing to suicide that are not represented in traditional datasets, and consider how data collaboratives built on search query data could inform future suicide prevention efforts in India and beyond.MethodsWe combined official statistics on demographic information with data generated through search queries from Bing to gain insight into suicide rates per state in India as reported by the National Crimes Record Bureau of India. We extracted English language queries on “suicide,” “depression,” “hanging,” “pesticide,” and “poison”. We also collected data on demographic information at the state level in India, including urbanization, growth rate, sex ratio, internet penetration, and population. We modeled the suicide rate per state as a function of the queries on each of the 5 topics considered as linear independent variables. A second model was built by integrating the demographic information as additional linear independent variables.ResultsResults of the first model fit (R2) when modeling the suicide rates from the fraction of queries in each of the 5 topics, as well as the fraction of all suicide methods, show a correlation of about 0.5. This increases significantly with the removal of 3 outliers and improves slightly when 5 outliers are removed. Results for the second model fit using both query and demographic data show that for all categories, if no outliers are removed, demographic data can model suicide rates better than query data. However, when 3 outliers are removed, query data about pesticides or poisons improves the model over using demographic data.ConclusionsIn this work, we used search data and demographics to model suicide rates. In this way, search data serve as a proxy for unmeasured (hidden) factors corresponding to suicide rates. Moreover, our procedure for outlier rejection serves to single out states where the suicide rates have substantially different correlations with demographic factors and query rates.

Forecasting plays a pivotal role in economic planning, particularly in aligning supply with demand and informing production decisions. This study aims to compare the performance of the Autoregressive Integrated Moving Average (ARIMA) and Seasonal ARIMA (SARIMA) models in forecasting the non-oil and gas export values of East Java, a region known for its dynamic trade activity. Using monthly time series data spanning from January 2007 to January 2024, sourced from the Central Statistics Agency (BPS) of East Java Province, this research conducts an in-depth analysis of forecasting accuracy and model suitability. Before model implementation, the dataset underwent several preprocessing steps to ensure its quality, including the handling of missing values and outlier adjustments. Both ARIMA and SARIMA models were developed, calibrated, and evaluated using standard forecasting performance metrics, namely Root Mean Square Error (RMSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). The ARIMA model exhibited consistently lower error rates across all three metrics, indicating its robustness in capturing the underlying patterns within the export data. In contrast, while the SARIMA model incorporated seasonal components, its performance did not surpass that of ARIMA in this specific case. The comparative findings suggest that, despite the seasonal nature of trade, the ARIMA model is more suitable for short-term forecasting of East Java’s non-oil and gas exports. This research contributes to the broader literature on economic forecasting by emphasizing the importance of selecting appropriate models based on data characteristics. Furthermore, the results provide valuable insights for policymakers and stakeholders engaged in export planning and regional trade development In this result the ARIMA model overcome the SARIMA with MAPE 0.116 to 0.983.

The global outbreak of coronavirus disease 2019 (COVID-19) has spread to more than 200 countries worldwide, leading to severe health and socioeconomic consequences. As such, the topic of monitoring and predicting epidemics has been attracting a lot of interest. Previous work reported search volumes from Google Trends are beneficial in decoding influenza dynamics, implying its potential for COVID-19 prediction. Therefore, a predictive model using the Wiener methods was built based on epidemic-related search queries from Google Trends, along with climate variables, aiming to forecast the dynamics of the weekly COVID-19 incidence in Washington, DC, USA. The Wiener model, which shares the merits of interpretability, low computation costs, and adaptation to nonlinear fluctuations, was used in this study. Models with multiple sets of features were constructed and further optimized by the highest weight selecting strategy. Furthermore, comparisons to the other two commonly used prediction models based on the autoregressive integrated moving average (ARIMA) and long short-term memory (LSTM) were also performed. Our results showed the predicted COVID-19 trends significantly correlated with the actual (rho <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$=$</tex-math> </inline-formula> 0.88, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$p $</tex-math> </inline-formula> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$&lt;$</tex-math> </inline-formula> 0.0001), outperforming those with ARIMA and LSTM approaches, indicating Google Trends data as a useful tool in terms of COVID-19 prediction. Also, the model using 20 search queries with the highest weighting outperformed all other models, supporting the highest weight feature selection as a feasible criterion. Google Trends search query data can be used to forecast the outbreak of COVID-19, which might assist health policymakers to allocate health care resources and taking preventive strategies.

Prior studies have shown that Internet search query data have great potential to improve tourism forecasting. As such, selecting the most relevant information from large amounts of search query data is crucial to enhancing forecasting accuracy and reducing overfitting; however, such feature selection methods have not been considered in the tourism forecasting literature. This study employs four machine learning–based feature selection methods to extract useful search query data and construct relevant econometric models. We examined the proposed methods based on monthly forecasting of tourist arrivals in Beijing, China, along with weekly forecasting of hotel occupancy in the city of Charleston, South Carolina, USA. Our findings indicate that the forecasting model with the selected search keywords outperformed the benchmark ARMAX model without feature selection in forecasting tourism demand and hotel occupancy. Therefore, machine learning methods can identify the most useful search query data to significantly improve forecasting accuracy in tourism and hospitality.

BackgroundA variety of obstacles including bureaucracy and lack of resources have interfered with timely detection and reporting of dengue cases in many endemic countries. Surveillance efforts have turned to modern data sources, such as Internet search queries, which have been shown to be effective for monitoring influenza-like illnesses. However, few have evaluated the utility of web search query data for other diseases, especially those of high morbidity and mortality or where a vaccine may not exist. In this study, we aimed to assess whether web search queries are a viable data source for the early detection and monitoring of dengue epidemics.Methodology/Principal FindingsBolivia, Brazil, India, Indonesia and Singapore were chosen for analysis based on available data and adequate search volume. For each country, a univariate linear model was then built by fitting a time series of the fraction of Google search query volume for specific dengue-related queries from that country against a time series of official dengue case counts for a time-frame within 2003–2010. The specific combination of queries used was chosen to maximize model fit. Spurious spikes in the data were also removed prior to model fitting. The final models, fit using a training subset of the data, were cross-validated against both the overall dataset and a holdout subset of the data. All models were found to fit the data quite well, with validation correlations ranging from 0.82 to 0.99.Conclusions/SignificanceWeb search query data were found to be capable of tracking dengue activity in Bolivia, Brazil, India, Indonesia and Singapore. Whereas traditional dengue data from official sources are often not available until after some substantial delay, web search query data are available in near real-time. These data represent valuable complement to assist with traditional dengue surveillance.

(ProQuest: ... denotes formulae omitted.)I. INTRODUCTIONThis paper estimates dynamic relationships between the Korean stock market index and the related online search queries within a multivariate GARCH framework. In addition, the paper also attempts to investigate whether the information from search query data can be served as a potential source for designing profitable trading strategies in the Korean stock market.The emergence of internet and social networking services combined with the extensive dissemination of smart phones have revolutionized the way we communicate and exchange information. Consequently, big data continuously flowing from increasing online activities by users have become a buzz word for recent years because of its potentials for various uses including marketing, political predictions, disease epidemics, social dynamics, etc.1Economists are one of the late professions who delves into the investigation of possible use of big data mainly for forecasting market dynamics and related issues.2 The seminal paper by Choi and Varian (2012) shows the use of Google search query data as a key predictor for various economic activities including auto sales, travels, etc.3 It stimulates subsequent studies in the field of economics, which mostly concentrates on exploiting big data to increase the prediction power of forecasting models for economic variables of own interests.The idea that the search query might contain information about subsequent actions by users is based on the premise that economic agents living in a contemporary society largely rely on the prior information-search process before making important economic decisions such as the purchase of durable goods and financial investments.However, the motivaton of information demand does not always run in this direction; the heightened information-gathering activity itself can be the manifestation of a simple endogenous response to major events in markets in quest of more information, which might yield possible effects on market developments in next rounds. These intricacies in the causal relationship between the information demand and the market outcomes make it difficult to assess correctly the real importance of big data. In spite of this complicacy, the predictive power of information generated from online big data for market activity is supported by numerous studies ranging from stock markets to housing markets.4The empirical analysis on the Korean stock market in this study reveals that the search frequency related to the Korean stock market has negative contemporaneous correlations with the KOSPI return for the majority of time with the occasional tightening of its magnitude. Furthermore, a negative association between the search query and the one-week-ahead stock return is observed, while the stock return has no statitically significant impact on the level of the future search query.Based on these observations, we experiment with a hypothetical trading strategy that interprets the increased level of online search activity as a negative signal for future stock returns so as to examine whether profitable trading schemes can be constructed out of big data. The result from this simple exercise demonstrates that the big data-based strategy outperforms the benchmark strategy in terms of the expected utility over wealth. The other experiment also shows that the big data-based option trading strategy can beat the market for certain KOSPI200 option contracts.As a result, this study is the first attempt to analyse the relationship between the KOSPI return and the degree of investor attention5 using the Korean stock market data and a Korean-based internet platform. We conjecture that the contribution is not limited to quantifying the dynamic relationships between stock retums and investor attention and estimating time-varying volatilities; it also explores the potential of big data as one of the practical tools for financial investment strategies, which is rarely pursued in the related literature. …

The use of Internet-based systems for infectious disease surveillance has been increasingly explored in recent years. However, few studies have used Internet search query or social media data to monitor spatial and temporal trends of avian influenza in China. This study investigated the potential of using search query and social media data in detecting and monitoring avian influenza A (H7N9) cases in humans in China. We collected weekly data on laboratory-confirmed H7N9 cases in humans, as well as H7N9-related Baidu Search Index (BSI) and Weibo Posting Index (WPI) data in China from 2013 to 2017, to explore the spatial and temporal trends of H7N9 cases and H7N9-related Internet search queries. Our findings showed a positive relationship of H7N9 cases with BSI and WPI search queries spatially and temporally. The outbreak threshold time and peak time of H7N9-related BSI and WPI searches preceded H7N9 cases in most years. Seasonal autoregressive integrated moving average (SARIMA) models with BSI (β = 0.008, p < 0.001) and WPI (β = 0.002, p = 0.036) were used to predict the number of H7N9 cases. Regression tree model analysis showed that the average H7N9 cases increased by over 2.4-fold (26.8/11) when BSI for H7N9 was > = 11524. Both BSI and WPI data could be used as indicators to develop an early warning system for H7N9 outbreaks in the future.

Making sense of antimicrobial use and resistance surveillance data: application of ARIMA and transfer function models