Forecasting commodity prices out-of-sample: Can technical indicators help? |
| |
Institution: | 1. School of Economics and Management, Nanjing University of Science and Technology, China;2. School of Finance, Nanjing Audit University, China;3. Antai College of Economics and Management, Shanghai Jiao Tong University, China |
| |
Abstract: | Economic variables are often used for forecasting commodity prices, but technical indicators have received much less attention in the literature. This paper demonstrates the predictability of commodity price changes using many technical indicators. Technical indicators are stronger predictors than economic indicators, and their forecasting performances are not affected by the problems of data mining or time changes. An investor with mean–variance preference receives utility gains of between 104.4 and 185.5 basis points from using technical indicators. Further analysis shows that technical indicators also perform better than economic variables for forecasting the density of commodity price changes. |
| |
Keywords: | Forecasting Commodity price Technical indicators Predictive regression Forecast combination |
本文献已被 ScienceDirect 等数据库收录! |
|