With the rapid development of information technology，Various data that can reflect human activities is explosive。People pay more attention to how to dig out valuable information from massive data，The understanding of big data is also deepening。At the beginning of the rise of big data，People often only put them with storage、Transmission、Network、Calculate the relationship，Therefore, its development and application areas are quite limited。Although the computer at this time、Automation、Scientific and technological fields such as communication science attaches great importance to big data，But the field of social sciences (such as economics) does not involve related research on big data，Instead, it still uses structural statistics as a sample，Use traditional measurement economic methods to conduct empirical research。In recent years，The advantages of big data in the field of social sciences are becoming more and more prominent。Big data has a sample mass、Strong real -time processing ability、Data non -structured characteristics，It can solve problems that some traditional statistical survey data cannot be solved。Economic research is inevitably combined with big data，Using big data methods to analyze the problem and gradually be sought after。Economists use real -time features of big data to establish prediction models，Carrying out policy influence analysis and other work，Use big data to verify the economic theory。So，Systematic summary of the research on big data metering economics is important。

　bet365 best casino games

Research on Empirical Economics is based on data，Data is verifying economic theory、Forecast macroeconomic trends、Formulate policy、Enterprise or individuals to make economic decisions，All have an indispensable role。Traditional statistics and measurement economic bet365 Play online games methods are often helpless for certain problems。At this time，Give full play to the large amount of samples of big data、Update real time、Advantages such as data source channels to solve difficult problems that are difficult to overcome in the past，It is the basic starting point for scholars to use big data methods to study economic problems。Overall，The application of big data in economics can be roughly classified as three categories: optimizing traditional economic indicators or building its advanced indicators、Build a new economic prediction indicator、Create contact between economic variables。

In terms of optimizing traditional economic indicators or building its advanced indicators，The role of big data should not be underestimated。For certain economic indicators (such as GDP growth rate、unemployment rate、Inflation rate, etc.) acquisition pathway，The traditional approach is to obtain these indicators through a large number of statistical surveys，Both need to invest huge costs in both the statistical process and data processing。and，Because the statistical process requires a long time，Data release usually exists，will lead to real -time monitoring of economic operation、Policy Formulation、Strategic decisions, etc., have different degrees of errors。So，With the continuous improvement of big data processing capabilities，Many economists have begun to optimize the original economic indicators using big data，or build new indicators with economic significance。For example，2008 Alberto Cavallo and others launched the "billion price plan"，Grasp the online price of more than 15 million kinds of products sold by more than 900 retailers on the website，and thereby calculating the daily real -time inflation index。The inflation index calculated by online price data can avoid the various disadvantages of traditional surveys，Therefore, the inflation index constructed through this method is often closer to the real level of inflation。Ding Huanfeng and others found the relationship between night light intensity data and economic activities，It is believed that economic operation (especially bet365 Play online games GDP) can largely reflect the light intensity。Under certain conditions，The brightness of the light can even be used as an alternative variable of GDP。Li Fengqi and others proposed a search prediction algorithm，It can automatically dig the relationship between Baidu search query index and economic indicators，From this to screen out the representative query data，and use this data as an advanced indicator，Forecast CPI and other macroeconomic indicators。Shen Shu and others based on consumer behavior theory，It proposed a network big data prediction model based on KPLSR methods and Lasso machine learning methods，and thus predict the consumer confidence index。Benjamin Edelman uses changes in the job supply and demand on the job search website，Predict the unemployment rate。

In terms of building a new economic prediction indicator，Due to rich channels for getting big data、Massive content of the income、Update efficient and timely，Economists have built many new economic indicators through big data。Compared with traditional data，indicators built by big data have stronger economic prediction capabilities。Cui Xianying (Hyunyoung Choi) thinks，Google trend data is better for short -term prediction，Searching data through the Internet can explore the user's behavioral characteristics，Easy to discover economic changes early，Therefore, a seasonal AR model containing Google trend variables，For economic indicators nearby (such as travel destinations、Car sales、Consumption confidence, etc.) Predictive。Discovery after verification，Compared with traditional methods，The accuracy of this method has increased by 5%-20%。Xu Wei combines online news emotions with Google search data，Established a data mining integrated model，and add the lagging item of the real estate price index to the model，Then use the support vector to return to the SVR model to predict the real estate price index。

In terms of establishing connections between economic variables，Some seemingly irrelevant data actually hides all kinds of contact bet365 Play online games between things，Analysis of correlation based on big data methods，It can often reveal the internal connection between different things。This can even subvert people's traditional cognition，So appropriately correct the existing theory。Establish connection between economic variables based on big data methods，It has become the most important and core part of big data in economic research。Li Yin (Yin Li) and others to study the government artificially、The effect of the tightness between the industry and the academic circles on the performance of the performance of small and medium -sized enterprises at the micro level，Information data data from 271 small and medium -sized green food companies on related websites，and the government、3989_4015，Finally verified the government、The closer the connection between the industry and the academic circles，The faster the sales growth of an enterprise。Helen Susannah Moat and others studied the relationship between the search data related to the financial -related and the stock market on the Google and Wikipedia website，Establish an advance indicator through online search data，As a result, a valid prediction of the stock market。Alain Yee Loong Chong, etc., or whether the two variables of online promotion and online comments can predict consumers' purchase behavior，What are the differences in the prediction effect of the two variables，Data based on Amazon website，Building product sales、Discount、Free shipping，and user praise、Neural network prediction model between variables such as poor review，It is found that the above variables can predict consumers' purchase behavior，But the best prediction effect of user review。Lai Yibing is based on the theory of regional economics，Based on big data and space measurement economic model theory，Empirical analysis of the regional economic development status and influencing factors of Guangdong Province。Explorest Space data analysis results indicate，The regional economic development of Guangdong Province has significant space correlation，bet365 live casino games The results of the measurement economic model incorporated into the space effect indicate the added value of the second industry、Add value of industries above designated size、Technology and government policy is the main influencing factor of the regional economic development of Guangdong Province。Matthew J. Schneider and others used the Bag-OF-Words to automatically process the user review text，And user review based on Amazon website predicts the sales of laptops after a week，Inspection found that the prediction results are more accurate than the model when the user comments are not considered。

　　Facing problems

Although the research on the application of big data in the field of economics in recent years has risen quickly，But there are still some problems in the analysis of big data metering economic analysis，The specific performance is as follows。

First，There are still various obstacles in the obtaining channels for big data。Current，Big data is mainly government、Banks and Internet companies, master the control，Not easily sold or disclosed these data。So，Many times economists can hardly obtain core big data resources to support their research。Not only that，For some big data involved in personal information，Due to the constraints of many aspects such as law and morality，Economists can hardly make full use of it。This leads to in the current economic research，The acquisition of big data can often only be searching for data such as data。So，In the future, you need to focus on solving related problems obtained by big data，This is also the most critical part of promoting the application of big data in economic research。

2，Mastering big data processing technology is not easy。Big data is the term from the computer field，Economists have not really paid attention to big data for more than ten years。Big data -based research combines each discipline，Among them, computers、Mathematics、Economics、Knowledge in the fields of psychology。So，For many economists，Machine learning、Data mining and other methods are not its director，Some complex research requires Bet365 app download the help of experts and mathematics experts to complete，This has caused certain restrictions to the application of big data in economics research。Future，Economists need to grasp the relevant technologies related to the acquisition and processing of big data，and provide more opportunities for scholars' cross -disciplinary cooperation，This highlights the power of big data in the economic field。

third，The theoretical foundation of the economic analysis method based on big data has not been consolidated。On the one hand，High -dimensional big data can easily generate correlation between variables。Although you can relieve this problem by reducing dimensions，But different from the traditional measurement economy，The theoretical significance of big data reduction and dimension is still controversial。On the other hand，The essence of the academic world has not reached an agreement on the essence of big data research methods: whether the research method based on big data is still a sample analysis like traditional statistics，Is it a general analysis? For the real overall，Big data can only be counted as one of the big samples，Maybe a large sample? at present，Economic analysis based on big data is mainly discussing the correlation between variables，cannot involve causal relationship，Therefore, we need to further improve the economic interpretation ability based on big data。

　Future Outlook

More than 40 years of reform and opening up，my country's economic development is rapid、The level of science and technology continues to improve。"Digital China" and "Smart Society" are important development goals in my country。Big Data Merit Economics has a broad development prospect，Further breakthroughs in the following two aspects。

First，Academic and Disciplines。To cultivate leading and original academic achievements in the field of big data metering economics，To strengthen the training of academic talents，Actively build a national think tank，Carry out academic and subject innovation，Therefore a batch of strategic needs that can serve national bet365 best casino games strategic needs、Taking quantitative analysis as the core、The important achievement of the combination of basic theory and application research。

2，policy application。In order to give full play to the role of big data measurement economy in terms of national governance and scientific decision -making，Should start from the actual situation of our country，Use the appropriate policy analysis and evaluation method in specific economic issues。In order to improve the ability to prevent and control my country's economic and financial risks in the new period，It should make full use of the prediction method based on big data measurement modeling，Provide scientific and effective policy recommendations。

　(Author Unit: Institute of Economic, Technical and Economics of the Chinese Academy of Social Sciences; Institute of Economics and Social Development of Nankai University)