FORECASTING THE BEHAVIOR OF MULTIVARIATE TIME-SERIES USING NEURAL NETWORKS

被引：285

作者：

CHAKRABORTY, K ^{[1
]}

MEHROTRA, K ^{[1
]}

MOHAN, CK ^{[1
]}

RANKA, S ^{[1
]}

机构：

[1] SYRACUSE UNIV,SCH COMP & INFORMAT SCI,CTR SCI & TECHNOL,4-116,SYRACUSE,NY 13244

来源：

NEURAL NETWORKS | 1992年 / 5卷 / 06期

关键词：

NEURAL NETWORKS; BACK PROPAGATION; MULTIVARIATE TIME-SERIES; STATISTICAL MODELS; TRAINING; ONE-LAG PREDICTION; MULTILAG PREDICTION; COMBINED MODELING; FORECASTING;

D O I：

10.1016/S0893-6080(05)80092-9

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper presents a neural network approach to multivariate time-series analysis. Real world observations of flour prices in three cities have been used as a benchmark in our experiments. Feedforward connectionist networks have been designed to model flour prices over the period from August 1972 to November 1980 for the cities of Buffalo, Minneapolis, and Kansas City. Remarkable success has been achieved in training the networks to learn the price curve for each of these cities and in making accurate price predictions. Our results show that the neural network approach is a leading contender with the statistical modeling approaches.

引用

页码：961 / 970

页数：10

共 12 条

[1]

[Anonymous], 1987, LEARNING INTERNAL RE

[2]

ARMER JD, 1987, PHYS REV LETT, V59, P845

[3]

BOX GEP, 1970, TIME SERIES ANAL FOR

[4]

GRANGER C. W. J., 1986, FORECASTING EC TIME

[5]

HARVEY AC, 1989, FORECASTING TIME SER

[6]

Li M., 1990, Proceedings. 5th IEEE International Symposium on Intelligent Control 1990 (Cat. No.90TH0333-5), P524, DOI 10.1109/ISIC.1990.128507

[7]

SAIKKONEN P, 1991, STAT SINICA, V1, P453

[8]

Tesauro G., 1988, Complex Systems, V2, P39

[9]

TIAO GC, 1989, J ROY STAT SOC B MET, V51, P157

[10]

Tong H., 1990, NONLINEAR TIME SERIE

← 1 2 →