Hsi-Cheng Li

A COMPARISON OF VAR AND NEURAL NETWORKS WITH GENETIC ALGORITHM IN FORECASTING PRICE OF OIL

This study applies VAR and ANN techniques to make ex-post forecast of U.S. oil price movements. The VAR-based forecast uses three endogenous variables: lagged oil price, lagged oil supply and lagged energy consumption. However, the VAR model suggests that the impacts of oil supply and energy consumption has limited impacts on oil price movement. The forecast of the genetic algorithm-based ANN model is made by using oil supply, energy consumption, and money supply (M1). Root mean squared error and mean absolute error have been used as the evaluation criteria. Our analysis suggests that the BPN-GA model noticeably outperforms the VAR model.

Gold Price, Neural Networks and Genetic Algorithm

Economic theory has failed to provide sufficient explanation of the dynamicpath of price movement over time. Therefore, the use of any linear ornon-linear functional form to model the gold price movement is bound to bearbitrary in nature. Neural Networks equipped with genetic algorithm have theadvantage of simulating the non-linear models when little a priori knowledgeof the structure of problem domains exist. Studies suggest that such a systemprovides better predictions when compared with traditional econometric models.The NeuroGenetic Optimizer software is applied to the NYMEX database of dailygold cash price covering 12/31/1974–12/31/1998 period. Among differentmethods, back-propagation neural networks with genetic algorithms is used topredict gold price movement. The results indicate that prices in the past, upto 36 days, strongly affect the gold prices of the future. This confirms thefact that there is short-term time dependence in gold price movements.

Job search and employment

Abstract A multistage search method based on fuzzy preference is used to analyze the processes of job search. An employment decision based on fuzzy preference results in an exact choice. The incorporation of fuzzy interval preference into employment decision, however, may lead to multiple choices, which provides a theoretical justification for the existence of fuzzy supply of labor. Thus, fuzziness, in addition to randomness, offers an explanation to wage dispersion.

An analysis of energy futures

This study investigates the distributions of returns on crude oil, heating oil and natural gas futures. Given substantial evidence of kurtosis and skewness in the returns, wavelet analysis is used to investigate the local correlation in time-frequency space between these three types of energy futures. In all three series, volatility is much more prevalent at shorter frequencies. This suggests that the departure from normality, and thus the trading risk, is greater at shorter timescales. In addition, the wavelet coherences between the returns on energy futures indicate high correlations over lower frequencies between 2005 and 2010, with the returns on crude oil futures often serving as the volatility leader. The results demonstrate the capability of wavelet analysis in processing nonstationary data and its ability to reveal causality. In addition, this study suggests that risk management based on the assumption of a normal distribution has limited applicability.