Market Trend and AnalysisMeasuring and Forecasting Steel Market Conditions with the POSRI Steel Index
조회수! : 3785
추천하기 : 0
It is difficult to be certain precisely what the “market conditions” may be since meanings and definitions vary. Over the history of the field of economics, a number of attempts have been made at explaining a particular state or status using indicators. The Steel Industry Index developed by the United States Geological Survey (USGS) is a representative composite index for the steel industry.
As a research institute specializing in steel, POSCO Research Institute(POSRI) has been conducting research into methodology that would be able to accurately and astutely predict steel market conditions: the POSRI Steel Index. The POSRI Steel Index explicitly uses four sectors—the economy, steel-consuming industries, steel demand/supply, and raw materials—to reflect steel market conditions. The POSRI Steel Index uses five indicators each for the four sectors, or twenty in total.
The primary advantage of the POSRI Steel Index is that it requires neither specialized statistical analysis nor econometric techniques. It rather uses a simple calculation of +1, -1, or 0 for a comparison of changes in indicators and scores among sectors. By transforming sector scores into radar charts, the POSRI Steel Index makes it easy to compare economic imbalances and intuitively grasp market conditions.
In the comparison between the POSRI Steel Index and actual steel price fluctuations, the POSRI Steel Index moves closely with steel prices, leading by three to four months. This outcome suggests that the Chinese steel market is highly likely to slow after the third quarter. However, with robust scores in the economy and steel-consuming industry sectors, a sudden fall is unlikely to occur in the second half of 2017. Due to the time differences in the collection of the statistics, indicators used to calculate the index are only publicly released one to two months later. To solve this problem, higher frequency data should be used rather than monthly data as a means to enhance predictability.