Data for: Nonlinear effects of financial factors on fluctuations in nonferrous metals prices: A Markov-switching VAR analysis

Our dataset consists of monthly data from August 2004 to October 2016 on international copper futures prices (P_CU), global refined copper production (GRCP), global refined copper consumption (GRCC), China’s copper imports (CCI), the percent position held by non-commercial traders (NCPP), federal funds rate (FFR), broad dollar index (BDI), and crude oil prices (COP). The copper futures closing prices of the LME are selected to represent international copper futures prices. Changes in global copper supply and demand are reflected by the global refined copper production and global refined copper consumption selected based on monthly data provided by the International Copper Study Group (ICSG, http://www.icsg.org/). We select China’s copper imports to represent “Chinese factor”, and these data are obtained from the average monthly data of copper ore and concentrate provided by China’s customs authorities. Following Sanders et al. (2004) and Fan and Xu (2011), we use the percent position held by non-commercial traders (NCPP) to measure the financial speculation, which is calculated by (non-commercial long position+ non-commercial short position+2* non-commercial spread position)/ (2* total open interest), the data are sourced from the Commodity Futures Trading Commission (CFTC). We use the federal funds rate as a proxy variable for the interest rate. The change in USD exchange rate is measured by the broad dollar index issued by the Federal Reserve Board, and it measures the change in the exchange rate of USD against a basket of foreign currencies. Regarding the oil price variable, as it is generally considered to be a good proxy for the global oil price market, we use the West Texas Intermediate (WTI) crude oil futures prices for the empirical analysis. The data on international copper futures prices, federal funds rate, broad dollar index and WTI crude oil prices are obtained from the WIND database. To eliminate heteroscedasticity, all the variables except financial speculation and federal funds rate are expressed in natural logarithms.

本数据集汇聚了自2004年8月至2016年10月的月度数据，涵盖国际铜期货价格（P_CU）、全球精炼铜产量（GRCP）、全球精炼铜消费（GRCC）、中国铜进口（CCI）、非商业交易者持有的百分比仓位（NCPP）、联邦基金利率（FFR）、广义美元指数（BDI）以及原油价格（COP）等多个维度。伦敦金属交易所（LME）的铜期货收盘价被选为国际铜期货价格的代表。全球铜供求关系的变动通过国际铜研究小组（ICSG，http://www.icsg.org/）提供的月度数据所选取的全球精炼铜生产和全球精炼铜消费指标得以体现。本数据集选取中国铜进口数据以表征“中国因素”，这些数据来源于中国海关当局提供的铜矿石和精炼铜平均月度数据。参照Sanders等（2004年）以及Fan和Xu（2011年）的研究，我们采用非商业交易者持有的百分比仓位（NCPP）来衡量金融投机，其计算公式为（非商业多头仓位+非商业空头仓位+2倍的非商业跨期仓位）/（2倍的总持仓量），数据源自商品期货交易委员会（CFTC）。联邦基金利率被用作利率的代理变量。美元汇率变动通过联邦储备委员会发布的广义美元指数来衡量，该指数反映了美元与一篮子外币之间的汇率变动。至于原油价格变量，鉴于其通常被视为全球石油市场的一个良好代理，我们使用西得克萨斯中质原油（WTI）期货价格进行实证分析。国际铜期货价格、联邦基金利率、广义美元指数和WTI原油价格的数据均来源于WIND数据库。为消除异方差性，除金融投机和联邦基金利率外的所有变量均以自然对数形式表示。