The Study Of China’s Copper Industry Composite Index And Early Warning System

Copper plays an important role in the modern society, and its development trend will have a great impact on the national economy. As its prosperity and warning systems are rarely studied, it’s conducive to reflect the actual comprehensive status of copper industry, and to find out the turning point in time. It is of great theoretical and practical value to carry out the study, with suitable index and methods.In this paper, we use composite index, early warning system and RS-NN model to monitor the copper industry. Firstly, on basis of the index system and processed data, we mark out the leading, consistent and lagging indicator group on the application of the time difference correlation analysis and peak-valley corresponding method. Combining theoretical analysis and industrial characteristics, we choose a total of45primary indicators, including four main parts, namely macro economy, energy and equipment supply, downstream and copper’s core indicators. Then, preprocess the indicators and conduct seasonal adjustment considering Chinese mobile holidays, with copper output as benchmarks, and finalized nine leading indicators, four consistent indicators and four lagging indicators.Secondly, we establish the composite index to describe and predict the state of the overall development of copper industry, among which, the leading composite index for prediction, the coincident one reflecting the current trends, and the lagging one for ex-verification. The research shows that the turning points of consistent composite index have averaged a2-3months lag over the leading one, two months lead over the lagging index.Then, combine3a theory and expertise to decide the critical value of early warning indicators, and to establish the early warning system. We use a similar manner as traffic lights to indicate each period of each indicator and the integrated early warning index, thus describing their state of the economy running hot or cold degree, playing early warning purposes, identifying the main cause of economic fluctuations, and providing reference for countermeasures. Through empirical research on China’s copper industry form January,2004to December2014, we conclude that the established warning system can accurately estimate the industry conditions.This paper also construct the RS-NN model to predict the warning degree of the copper industry, in which, artificial neural network model for predicting the future trends, and the rough set is used to simplify the input layer of artificial neural network model to enhance early warning accuracy. The RS-NN model complements each other’s advantages, and enriches the intelligent forecasting techniques in the field of economic monitoring and early warning. Comparing with the BP neural network, we find that the RS-NN model is superiority to BP. The RS-NN model can be used in short-term prediction or a longer period, and determine the warning degree.