Contents lists available at ScienceDirect Resources Policy journal homepage: www.elsevier.com/locate/resourpol Copper price estimation using bat algorithm Hesam Dehghani a, ⁎ , Dejan Bogdanovic b a Mining Engineering Faculty, Hamedan University of Technology, Hamedan, Iran b Technical Faculty, University of Belgrade, Belgrade, Serbia ARTICLE INFO Keywords: Copper price Bat algorithm Time series Estimation ABSTRACT The most effective parameter on the value of mining projects is metal price volatility. Therefore, knowing the metal price volatility can help the managers and shareholders of the mining projects to make the right decisions for extending or restricting the mining activities. Nowadays, classical estimation methods cannot correctly es- timate the metal prices volatility due to their frequent variations in the past years. For solving this problem, it is necessary to use the artificial algorithms that have a good ability to predict the volatility of the various phe- nomena. In this paper, the Bat algorithm was used to predict the copper price volatility. Accordingly, Brownian motion with mean reversion (BMMR) was chosen as the most suitable time series function with the root mean square error (RMSE) of 0.449. Then, the estimation parameters of the equation were modified using Bat algo- rithm. Finally, it is concluded that the determined equation with 0.132 of RMSE can predict the copper price better than the classic estimation methods. 1. Introduction Product price is the most important and effective parameter in various project evaluation. Mining projects are no exception and the value of the project is the most sensitive to changes in mineral prices. Therefore, knowing the mineral price changes may play an important role in making the right decisions for applying the administrative op- tions for extending or restricting the mining activities via mining pro- jects managers and shareholders. Significant volatilities in the minerals price, especially in recent years, have led that the classic prediction approaches do not have the ability to correctly estimate the price changes. Hence, numerous researchers have tried to predict the mi- nerals price using artificial methods. Xie et al. (2006) proposed a new method for crude oil price prediction based on a support vector ma- chine (SVM) model. They compared their model with other models, which were developed using artificial neural networks (ANN) and ge- netic algorithm (GA). The obtained results show that like ANN and GA, SVM is a capable method for forecasting the crude oil price. Hadavandi et al. (2010) developed a time series model for gold price and exchange rate forecasting based on particle swarm optimization (PSO). Dehghani and Ataee-pour (2012) predicted the copper price using binomial tree. Dehghani et al. (2014) estimated the price and operating cost in a copper mine using multidimensional binomial tree. Kriechbaumer et al. (2014) used an improved combined wavelet-autoregressive integrated moving average (ARIMA) to forecast monthly price of aluminum, copper, lead and zinc. Chen et al. (2016a), (2016b) investigated the changes of the various metals price using grey wave forecasting method. Liu and Li (2017) forecasted the gold price and analyzed the related influence factors based on random forest. Li and Li (2015) studied the volatility of the copper price using time series functions. Lasheras et al. (2015) used the ARIMA and artificial neural networks methods for predicting the copper spot price in New York Commodity Exchange (COMEX). The results of this research show that the estima- tion error of the neural network is always less than time series model. Liu et al. (2017) predicted the copper price using decision tree learning. Their model forecast the copper price using price volatility of the sev- eral materials such as crude oil, gold, silver, etc. Table 1 shows some of the research-works in field of minerals price prediction. Numerous studies on the mineral price prediction indicate the challenge and importance of this process. Copper price plays vital roles in various aspects in today's econo- mies. Copper price has a significant impact on gold and other precious metals' prices (Morales and Andreosso-O′Callaghan, 2011). Copper is strongly associated with many industries, such as electrical wiring, construction, and equipment manufacturing; and therefore, copper price has become a significant impact factor on the performance of related companies and economies (Lasheras et al., 2015). On the other hand, for some countries such as Chile and Zambia, whose economy relies extensively on copper production, fluctuations in copper price is very important (Lasheras et al., 2015). Forecasting copper price is particularly important for policymakers and participants in the market. Budgeting is crucial for mining http://dx.doi.org/10.1016/j.resourpol.2017.10.015 Received 23 August 2017; Received in revised form 25 October 2017; Accepted 27 October 2017 ⁎ Corresponding author. E-mail addresses: dehghani@hut.ac.ir (H. Dehghani), dnbogdanovic@yahoo.com (D. Bogdanovic). Resources Policy xxx (xxxx) xxx–xxx 0301-4207/ © 2017 Elsevier Ltd. All rights reserved. Please cite this article as: Dehghani, H., Resources Policy (2017), http://dx.doi.org/10.1016/j.resourpol.2017.10.015