How the CPI survey develops in the big data age

Along with the community on the CPI (consumer Price index) the increasing concern, the rapid development of the Internet, an information explosion characterized by a large data era is coming. This puts more demands and challenges on the CPI survey, actively explores how to deal with the arrival of large data age, and realizes the CPI survey and the era of convergence has become a top priority.

Problems in CPI survey

First, the investigation object does not cooperate the phenomenon to happen frequently. Price survey of fresh goods and other parts of the specification requirements "every five every 10" mining price, the frequency of higher prices, a wide range of investigations, and so in the market with farmers, traders and other frequent contact process, the price encounter "cold" phenomenon occurs.

Second, part of the principle of mining price is difficult to implement thoroughly. From the actual business work of CPI, it is difficult to carry out the principle of the main mining price is homogeneous comparable and the actual transaction price principle. For example, due to technical updates, product upgrades and other reasons, such as mobile phones, computers and other specifications of the price of goods is very common, the auxiliary is often faced with a reasonable price or a product performance indicators close to the dilemma, the same quality can be compared to the principle can only grasp the price trend of lax implementation, at the same time because the auxiliary staff only to buy, and clothing, Furniture and other specifications there is a larger bargaining space, so you want to accurately obtain the actual price of specifications, unless the merchant with a higher degree of cooperation, is willing to provide sales vouchers, otherwise accurate to the actual transaction price difficulties.

Thirdly, it is difficult to find the same specifications at different mining prices. According to the program requirements for the same industrial consumer goods and service items specific specifications to select more than 2 survey points, in fact, most of the specifications can meet the requirements of individual industrial consumer goods such as packaged food, clothing, home appliances, etc. for supermarkets, stores and so on to obtain product advantages in the competition to customize the different requirements of the purchase specifications, Some specifications even become special products, these different business strategies for the different price points to find brands, models, performance and other identical specifications bring certain difficulties.

Four is the price point changes affect the quality of data. Through the spot check of the price data and the field mining price, we found that there is a big difference in the price of different prices, especially for the fresh goods, first of all the farmers market stalls, the second market fixed stalls and scattered stalls coexist, because the fixed stall stalls higher, better food, the higher the cost of purchase Scattered stalls of the majority of vegetables sold, and therefore the same specifications of the price differences between the larger, and the auxiliary to the market when the market for the price of the market to replace the situation, resulting in price artificially.

The challenges of CPI in the big Data age

First, the large data on the CPI collection methods have impact. Under the background of large data, the development and application of Internet and IoT of things have accelerated the infiltration of informatization to all aspects of social economy and the daily life of the residents, the data source range is more extensive, and the collection channel is more diversified, which will impact on the traditional way that CPI hires the regular price data collection.

Second, large data on the CPI data quality requirements more stringent. In the large data age, information centralized sharing, information platform transparency will be the inevitable trend of development, in the social public to statistical information products and services, convenient, rapid query and timely access, the statistical data quality will be more sensitive, more attention, higher requirements.

Third, large data on the CPI price data analysis requirements higher. Large data age, the investigation team in the traditional sense of professional division of labor will be broken, the investigation of professional integration and crossover will be an inevitable, from the CPI price movements derived from the data processing analysis chain in addition to the traditional sense of the closely related to the production field PPI, IPI, The price of CPI data processing analysis requires a higher level and depth broader.

Research on the development model of CPI business in large data age

--CPI Price data acquisition mode changes. The current CPI data generation process is mainly through the use of handheld data collector, management system, summary program through the layer of summary report to generate CPI various indices, and large data era faced with a huge amount of price data, CPI data production mode of data entry should be in the merchant trading platform or sales system to join the query, screening, Extraction, import and a series of intelligent technology to operate the port to obtain the required specifications price and information, then through the corresponding data processing program to generate CPI index, so on the one hand to reduce the auxiliary to pick up the price link, to avoid the auxiliary to the price of the process of active or passive mining price errors caused by data distortion, On the other hand, by querying and searching on the platform, we can quickly and accurately locate the different kinds of information of the specification products between the various price points. In order to represent the selection of specifications, the correct replacement of the specifications, the same specifications in different price points of the exact price and the same quality comparable, the number of principles, The actual transaction price, such as the principle of better implementation to provide a more favorable enforcement environment.

--CPI investigation technical means innovation. Under the background of large data, the first is that the CPI data collection method relies on the information technology means comprehensively; the second is to set up a centralized management platform for CPI statistics, including a series of statistical data management systems such as CPI data collection, data processing, data summarization, data evaluation, data release, public inquiries, etc. To achieve the perfect convergence and use of large data.

--electric dealers and other online stores gradually become the main object of CPI collection. At present, Taobao, Cat, when, Jingdong Mall, Suning easy to buy and many other network trading platform, the goods involved in food, clothing, electrical appliances, furniture, building materials and many other closely related to the consumer goods, the transaction amount is huge, and the CPI survey of the eight categories of goods and services in most of the specifications of the match, Electric dealers and other online stores gradually become the main channel for future CPI data collection, of course, in addition to the large-scale, national-type trading platform, in order to achieve the CPI better represent the region's price trend, but also should be integrated regional network trading platform, mining price points, such as web data online information, such as Lianyungang appeared "Lianyungang Basket Net" Online shopping platform, related to eggs meat, cereals and oils, vegetables, fruits and other commonly used agricultural and sideline products, CPI food 16 categories in more than 60% representative of the price of the specification product information can be in the transaction platform, in the full assessment of its and the real store price differences and the public feel the matching degree of the premise, It can be used as the exploration of some food price network in this region.

--CPI weight adjustment frequency is accelerated. China's current CPI adjustment frequency is relatively low, according to the programme system requirements, adjusted once every 5 years, the standard mainly based on the national income and expenditure of 140,000 households and households survey data, supplemented by a typical survey to supplement, from the adjustment of frequency to see clearly not with the pace of economic development. In the background of large data, the function of data sharing will be gradually embodied, through the construction of Shared data center, the independent and decentralized information of each investigation business will be integrated and utilized to realize the sharing of household income and expenditure data of urban and rural households, and to speed up the adjustment of CPI weight becomes inevitable, Realization in each CPI survey natural year moderate weight adjustment, and the CPI weights in addition to urban and rural households income and expenditure data, should also be based on trade, services, health care and other relevant departments of data, and through strengthening with the large supermarkets, shopping malls, electric dealers and other survey points, access to large supermarkets, Shopping malls and other historical sales details for the CPI survey of small categories, the basic classification of weight distribution provides a more persuasive reference.

--CPI Business personnel function change. In the big Data age, the price data is mainly obtained from the electricity trader transaction, electric dealer's sales strategy such as Taobao second kill, timed snapping, holiday promotions and other activities led to the same specifications of a certain price fluctuations in prices, a large range, coupled with the shop to circumvent the physical store personnel costs, field rent, utilities and other conventional operating expenses expenses, Commodity prices are significantly lower than physical stores, this is a huge challenge to the CPI business personnel ability, business personnel need specialized data mining and analysis processing ability, one is able to trade platform, record, system and so on the massive price data carries on the accurate screening localization, and to the price data comprehensive control and the key audit, Second, the CPI data and other related professional statistical data can be integrated analysis and processing.

--CPI data Publishing is more open and transparent. First, the CPI data release way more perfect. The current CPI data mainly by the use of television, network, newspapers and other news media to publish regularly, large data age with data visualization, human-computer interaction, machine intelligence and other advanced technology development and application, data release form will be more diverse. Second, the CPI data is more open and transparent. Because of the massive storage and sharing of data, data information is no longer a secret, so for the more attention of the CPI data should be as open as possible raw data, weight measurement data, compilation methods, all kinds of price indices and other data, the public's attention to CPI data and demand from the traditional passive acceptance into active query and acquisition.

--CPI Investigation Service function promotion. In the large data age, the CPI service subject should abandon the past government statistics for a long time to serve as a single service object of the concept, to meet more social groups, investors, business operators, self-employed and even farmers in the market stall, such as the objective needs of the investigation object-oriented, better publicity and interpretation of CPI While emphasizing the legal investigation, to guide the public to make full use of price data products to serve their own production operations and consumption activities, and actively participate in the price data collection, analysis, use of the link, so as to establish a new cooperative relationship between investigators and survey objects, to achieve a true sense of mutual benefit and win.