PROBABILITY AND STATISTIC SOFTWARE FOR ENGINEERING APPLICATION: MODELING AND FORECASTING COPPER PRICES FOR INDUSTRIES

• Main Author: MAZANI, MOHD FAIZAL
• Format: Final Year Project
• Language: English
• Published: Universiti Teknologi PETRONAS 2007
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://utpedia.utp.edu.my/9598/1/2007%20-%20Probability%20and%20Statistic%20Software%20for%20Engineering%20Application%20Modelling%20and%20Forecasting.pdf; http://utpedia.utp.edu.my/9598/

Theobjective of thisfinal year project is to develop a probability and statistics software for engineering application. The chosen engineering application is modeling and forecasting copper prices for industries. The software may be used by managers, market researchers, and survey companies, in making decisions in copper-related business. The price of copper has been volatile over the years due to competing materials, remaining stocks available, and economic uncertainties. Therefore, me software modeling tool will be useful to forecast the future price of copper to enable the user in making someadjustments or preparation in their business. The proposed framework of the system consists of three inter related components, the database that will provide input to the model, the forecasting modeling and user interface that provides a medium for the userto communicate with the system. Three stages have been identified in order to develop the system. They are variable identification, statistical model development and the development of the software. Several variables have been identified but only eight variables are finalized as the independent variables ofthe model. The models mat were identified are multiple regression analysis and time series. The best possible R2 obtained in the modeling is 0.939 which is quite high. The accuracy ofthe forecasted price ofcopper isapproximately higher than 87%.The model is incorporated in an interactive user-friendly interface.