mpra paper 2855
TRANSCRIPT
MP A RMunich Personal RePEc Archive
A study on the factors aecting gold price and a neuro-fuzzy model of forcastSarfaraz, Leyla and Afsar, Amir UNSPECIFIED
2005
Online at http://mpra.ub.uni-muenchen.de/2855/ MPRA Paper No. 2855, posted 07. November 2007 / 02:46
. ARIMA . . - . .
:
1 - 2-
1- . . . . . . ) ( . . . . ) (ARIMA . .
1
- Neuro Fuzzy
. .
2- ) 4191-0881( . . 007 0781 0008 3191 . 3191 3922 3321 0301 934 ]42[. . 33-8291 . . . 4391 76/02 53 . . 0006 5291 00081 56% ]42[.
. 1 2002- 8491 . " -" ]9[. . 53 .
: )(World Gold Council
1: 2002 8491 )(
0491 0591 0591 0691 . . . . . 00002 0591 0009 1791 ]42 4002[.
1791 . 0791 0991 002 008 ]9[. 5791 1 . 6791 3
]7[. 8791 " " . . 0891 0991 ]7[. 1 4/36 0891 9/51 7991 . 01% 0891 9991 . 1 ) ( 9991 9/34 1/74 4/26 8/54 7/23 4/32 9/51 2/51 6/41 7/21 65 681 014 544 709 2141 9761 2371 4281 0691 0/44 4/561 2/976 2/573 3/034 5/034 2/613 7/903 8/113 4/682 : 0791 5791 0891 5891 0991 5991 7991 8991 9991 0002
1002 Gold in the Official Sector
3- ]8[. . ]11[. . ) ( . 1791 ) ( 53 ]6[. 1 . . . 0691 1791 . 0791 . 9791 . 72 ]41[. 2891 4891 )51 3891( 5891 )61 4891( ]51[ 5891 013 ]4[.
6891 9/51 863 . 023 6891 074 7891 005 7891 ]5[. 7891 7 )04 5891( . . 8891 . . 2 . 052 9991 75 . 1 03 . 5791 0031 2002 5791 ]71[.
: )(World Gold Council
2: ) (1
- Millennium Gold Fund
4- 1 2 . ) ( . ) ( . .
5- 3 . . . . . 4 ) (MFNN . .
1 2
- Financial Assets - Real Assets 3 - Artificial Neural Networks 4 - Multilayered Feedforward Neural Network
]32[ ) MFNN( . . MFNN . . ) Zk ( dk . . . ]418122[. ) (MSE . . ) ( . . .
. . ) (ANFIS Jaris ]01,21,31[. 3 .
3:
ANFIS . . .
6- . 6-1 6-2 .1 2
- Backpropagation - Artificial Nero Fuzzy Inference Systems
6-1- . ) (Pd ) (Pw ) (Ps ) (CPI )$( . . Ps CPI Pw $ . ) 3( 3 - Ps Pw Dollar CPI Pd
058/0 946/0
059/0 757/0
804/0 901/0-
247/0 1
1 247/0
Pd CPI
. . . ) ( . ) (Ps ) (CPI
. CPI 4 . 4: Sig t B R2
R
Pw
300/0 100/0 010/0 120/0 200/0
091/3- 014/4 807/2 904/2 413/3 635/0 822/0 202/0 891/0
9/744728- 152/224 693/904 781/03 591/021 039/0 469/0
CPI Ps $ Dollar
39% Ps CPI Pw . $ 591/021 + Pd = 827447/9 + 1422/251Pw + 409/396CPI + 30/187 Ps
. .
6-2- ) (MFNN . . )1 53 51(
1
- Takagi-Sugeno
. . 63 " " - - . 025 05 52 52 )( . RMSE 4 . . . : 1( 2( .5.9 9 5.8 8 5.7 7 1 72 41 35 04 97 66 691 381 071 751 441 131 811 501 29 Epoch
RMSE1 2
4: RMSE
- Difference of two sigmoid - Over fitting
. : ) (MSE ) (RMSE ) .(NMSE . 2 R NMSE 2.NMSE=1-R 2 R . ) (MAE . . 5 . 5: 2R MAE NMSE MSE RMSE
4989/0
2973/5
9010000/0
1298/36
3399/7
5 6 . 062 062 . 5 6 .
0038 0028 0018 0008 0097 0087 0077 1 352 532 712 991 181 361 541 721 901 19 37 55 73 91FNN Actual
5: 0018 FNN Actual
0508
0008 Price 0597 0097 0587 1 352 532 712 991 181 361 541 721 901 19 37 55 73 91 Time
6:
6-1 39% 6-2 32/99 . .
7- . ]61[. 1 3
]1[. . .
8- . . .
. . . .
9- ]1[ 9731 ]2[ 8531 ]3[ 56-3631. ]4[ . " " 311 0731. ]5[ )4731(." " . 3. ]6[ . )6731( . .,[7] Apple yard & Field (1995). "International Economics, Payments, Exchange Rates .Macroeconomic Policy, University of North Carolina at chapel Hill, USA ,[8] Bogner .Stephan .)3002( "Gold .vs US ."Dollarhttp://216.239.37.104/translate_c?hl=en&sl=de&u=http://www.goldseiten.de/ansichten/bonger...&03.htm
[9] Carbaugb, R. J. (1992). "International Economies", Wadsworth Publishing .Company [10] Dhar,V.& chou,D.(2001). A comparison of nonlinear methods for predicting .129-709,)4(21 ,earnings surprises and returns. IEEE Transactions on Neural networks
[11] Hamilton,
Adam.
(2003),
"Gold
in
Euros
and
Yen".
http://www.zeall/c.com/2003/goldfx.htm.
[12] Jang, J. R. and Sun, C. (1995). Nero Fuzzy Modelling and Control, Proc. of the IEEE, P.P: 378-405. [13] Kuo, R.J., Chen, C.H. and Hwang, Y.C. (2001). An intelligent stock trading decision support system through integration of genetic algorithm based fuzzy neural network and artificial neural network, Fuzzy Sets and Systems, 118 . [14] Le Cun, Y. (1985). Une procedure d'apprentissage pour reseau a seuil assymetrique. Cognitive, 85, 599-604. [15] Lee, C.C.(1990). Fuzzy logic in control systems: fuzzy logic controller, Part I, IEEE Trans. Systems Man Cybernet. 20 (2) , 404 - 418. [16] Lippmann, R.P. (1987). An introduction to computing with neural nets, IEEE Mag. 3 (4), P.P: 4-22. [17] Morrison, Kevin. (2003), "Gold bulls wait for further dollar slide". http://www.siliconinvestor.com/stocktallc/msg.gsp?msgid=19529668. [18] Parker, D.B. (1985). Learning logic: Casting the cortex of the human brain in silicon. Technical Report TR-47. Cambridge, MA: Center for Computational Research in Economics and Management, MIT. [19] Qi,M.(2001). Predicting US reccessions with leading indicators via neural network models. International Journal of Forecastinig,17(3),383-401. [20] Tong, R. M. (1997). A control engineering review of fuzzy systems, Automatic, 13(6): P.P: 559-569. [21] Wang, L. X., Mendel J. M. (1992). Fuzzy basis functions, universal approximation and orthogonal least-squares learning, IEEE Transaction on Neural Networks, 807-814. [22] Werbos, P.J. (1974). Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences. Cambridge, MA: Harvard University, Ph.D. dissertation. [23] Wong, B.K., Jiang, L. and Lam, J. (2000). A bibliography of neural network business application research: 1994-1998. Computers and Operations Research, 27(11), 1045-1076. [24] World Gold Council, Gold as a Reserve asset. 2003, http://www.gold.org/value/reserve-asset/gold-as/background.html. [25] Zadeh, L.A. (1973). Outline of a new approach to the analysis of complex systems and decision processes, IEEE Trans. Systems Man Cybernet. SMC-3 (1), P.P: 28-44. [26] Zimmermann, M. J. (1996). Fuzzy Set Teory and its application, Kluwer Academic Publishers, Boston.