Versión de HTML Básico
Elsy Gómez-Ramos y Francisco Venegas-Martínez
Analíti a
k
6
Revista de Análisis Estadístico
Journal of Statistical Analysis
-0,001
1,13E-17
0,001
0,002
0,003
0,004
0,005
0,006
3
3
4
6
7
10
Training
Forecast
Figura 3.
Training and forecast errors. Source: author elaboration.
7 Conclusions
This paper carried out an exhaustive review of the
specialized literature on ANNs and made a comparati-
ve analysis according to their performances in forecasting
stock indices (or stocks) and exchange rates. In this regard,
it is important to point out that the MLP is one of the most
used networks in finance, because it is a feedforward mul-
tilayer network with non-linear node functions. In order to
support this, we have reviewed thirty applications in the li-
terature. We found that more than 40 % of the analyzed re-
searches support the idea that the MLP is the best network
or at least it has the same performance with respect to the
proposal networks. However, it is shown that the MLP has
important delimitations in several respects: network archi-
tecture, basic functions and initialization weights. One way
to improve the performance of the MLP is to apply intelli-
gent methods. As a result we get a hybrid network which
is expected to provide a more accurate forecast.
Referencias
Adeodato, P., A. Arnaud, G. Vasconcelos, R. Cunha y
D. Monteiro. 2011.
Mlp ensembles improve long term pre-
diction accuracy over single networks
. International Journal
of Forecasting, 27(3): 661–671.
Anderson, J. A. 2007.
An Introduction to Neural Networks
.
The IMT Press.
Bahrammirzaee, A. 2010.
A comparative survey of artificial in-
telligence applications in finance: Artificial neural networks,
expert system and hybrid intelligent systems.
Neural Com-
puting and Applications, 19(8): 1165–1195.
Bildirici, M., E. Alp y O. Ersin. 2010.
Tar-cointegration neu-
ral network model: An empirical analysis of exchange rates
and stock return.
Expert Systems with Applications, 37(1):
57–69.
Brooks, C. 2006.
Introductory Econometrics for Finance
. Cam-
bridge University Press, 7 edición.
Burrell, P. y B. Folarin. 1997.
The impact of neural networks in
finance.
Neural Computing and Applications, 6(4): 193–
200.
Carpinteiro, O., J. Leite, C. Pinheiro y I. Lima. 2011.
Fore-
casting Models for Prediction in Time Series. Artificial Intelli-
gence Review.
Springer.
Chang, P., D. Wang y Z. D. 2012.
A novel model by evolving
partially connected neural network for stock price trend fore-
casting
. Expert Systems with Applications, 39(1): 1–15.
Cheng, C. 1996.
Fuzzy Logic and Neural Network Handbook
.
IEEE Press.
Dhamija, A. y V. Bhalla. 2011.
Exchange rate forecasting: Com-
parison of various architectures of neural networks.
Neural
Computation and Applications, 20(3): 355–363.
Enke, D. y S. Thawornwong. 2005.
The use of data mining and
neural networks for forecasting stock market returns
. Expert
Systems with Applications, 29(4): 927–940.
Ghazali, R., A. J. Hussain, D. Al-Jumeily y M. Merabti.
2007.
Dynamic Ridge Polynomial Neural Networks in Ex-
change Rates Time Series Forecasting
. ICANNGA.
Ghazali, R., A. J. Hussain y P. Liatsis. 2011.
Dynamic ridge
polynomial neural network: Forecasting the univariate non-
stationary and stationary trading signals
. Expert Systems
with Applications, 38(4): 3765–3776.
Ghazali, R., A. J. Hussain, N. M. Nawi y B. Mohamad. 2009.
Non-stationary and stationary prediction of financial time se-
ries using dynamic ridge polynomial neural network
. Neuro-
computing, 72(10-12): 2359–2367.
Guresen, E., G. Kayakutlu y T. U. Daim. 2010.
Using artifi-
cial neural network models in stock market index prediction
.
Expert Systems with Applications, 38(8): 10389–10397.
Hamid, Z., S. A. y Iqbal. 2004.
Using neural networks for fore-
casting volatility of s&p 500 index futures prices
. Journal of
Business Research, 57(10): 1116–1125.
Hansen, J. V. y R. D. Nelson. 2003.
Forecasting and recombi-
ning time-series components by using neural networks
. Jour-
nal of the Operational Research Society, 54(3): 307–317.
Haykin, S. 1994.
Neural Networks a Comprehensive Founda-
tion
. IEEE Computer Society Press.
Hussain, A., J. A. Anderson, A. Knowles, P. J. G. Lisboa y
W. El-Deredy. 2008.
Financial time series prediction using
polynomial pipelined neural networks
. Expert Systems with
Applications, 35(3): 1186–1199.
Hutchinson, J. M., A. W. Lo y T. Poggio. 1994.
A nonparame-
tric approach to pricing and hedging derivative securities via
learning networks
. Journal of Finance, 49(3): 851–889.
Ince, H. y T. B. Trafalis. 2006.
A hybrid model for exchange rate
prediction
. Decision Support Systems, 42(2): 1054–1062.
Kaastra, I. y M. Boyd. 1996.
Designing a neural network for
forecasting financial and economic time series
. Neurocompu-
ting, 10(3): 215–236.
14
Analítika,
Revista de análisis estadístico
, 3 (2013), Vol. 6(2): 7-15