See

discussions, stats, and author profiles for this publication at:

http://www.researchgate.net/publication/220346957

A high order fuzzy time series forecasting

model based on adaptive expectation and
artificial neural networks.
ARTICLE in MATHEMATICS AND COMPUTERS IN SIMULATION DECEMBER 2010
Impact Factor: 0.86 DOI: 10.1016/j.matcom.2010.09.011 Source: DBLP

CITATIONS

DOWNLOADS

VIEWS

41

89

3 AUTHORS, INCLUDING:
Cagdas Hakan Aladag

Erol Egrioglu

Hacettepe University

Giresun University

46 PUBLICATIONS 356 CITATIONS

50 PUBLICATIONS 338 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Erol Egrioglu

Retrieved on: 03 July 2015

Available online at www.sciencedirect.com

Mathematics and Computers in Simulation 81 (2010) 875882

Original articles

A high order fuzzy time series forecasting model based on adaptive

expectation and artificial neural networks
Cagdas Hakan Aladag a, , Ufuk Yolcu b , Erol Egrioglu b
b

a Department of Statistics, Hacettepe University, Ankara 06532, Turkey

Department of Statistics, Ondokuz Mayis University, Samsun 55139, Turkey

Received 27 February 2009; received in revised form 10 September 2010; accepted 18 September 2010
Available online 29 September 2010

Abstract
Many fuzzy time series approaches have been proposed in recent years. These methods include three main phases such as
fuzzification, defining fuzzy relationships and, defuzzification. Aladag et al. [2] improved the forecasting accuracy by utilizing feed
forward neural networks to determine fuzzy relationships in high order fuzzy time series. Another study for increasing forecasting
accuracy was made by Cheng et al. [6]. In their study, they employ adaptive expectation model to adopt forecasts obtained from
first order fuzzy time series forecasting model. In this study, we propose a novel high order fuzzy time series method in order
to obtain more accurate forecasts. In the proposed method, fuzzy relationships are defined by feed forward neural networks and
adaptive expectation model is used for adjusting forecasted values. Unlike the papers of Cheng et al. [6] and Liu et al. [14], forecast
adjusting is done by using constraint optimization for weighted parameter. The proposed method is applied to the enrollments of
the University of Alabama and the obtained forecasting results compared to those obtained from other approaches are available in
the literature. As a result of comparison, it is clearly seen that the proposed method significantly increases the forecasting accuracy.
2010 IMACS. Published by Elsevier B.V. All rights reserved.
Key words: Adaptive expectation model; Feed forward neural networks; Forecasting; Fuzzy relations; Fuzzy time series

1. Introduction
In recent years, fuzzy time series approach introduced by Song and Chissom [16,17] has been used widely. In the
literature, many studies have been made to improve forecasting accuracy in fuzzy time series model. Chen [4] proposed
a method which is simpler than the method proposed by Song and Chissom [16,17] in forecasting fuzzy time series.
The method proposed by Chen [4] does not include complex matrix operations in defining fuzzy relation. Huarng [11]
pointed out that the interval length influences the forecasting performance and proposed two methods, which are based
on the average and the distribution, for defining the length of interval. Egrioglu et al. [8] suggested a new approach
which is based on the optimization of the interval length. Cheng et al. [6] introduced a method based on adaptive
expectation model. In the method proposed by Cheng et al. [6], the forecasts obtained from the first order fuzzy time
series model are adjusted by employing adaptive expectation model. Cheng et al. [6] and Liu et al. [14] introduced
a method based on adaptive expectation model. In the methods proposed by Cheng et al. [6] and Liu et al. [14], the

Corresponding author. Tel.: +90 312 2977900; fax: +90 312 2977913.
E-mail addresses: aladag@hacettepe.edu.tr, chaladag@gmail.com (C.H. Aladag).

0378-4754/$36.00 2010 IMACS. Published by Elsevier B.V. All rights reserved.

doi:10.1016/j.matcom.2010.09.011

876

C.H. Aladag et al. / Mathematics and Computers in Simulation 81 (2010) 875882

forecasts obtained from the first order fuzzy time series model are adjusted by employing adaptive expectation model.
The studies mentioned above use first order fuzzy time series forecasting model.
Since first order fuzzy time series models have got a simple structure, they can be often insufficient to explain more
complex relationships. For this reason Chen [5] proposed a new method which analyze a high order fuzzy time series
forecasting model. The implementation of Chens approach becomes more difficult when the order of fuzzy time series
increases. However, neural networks can be used easily for high order fuzzy time series. Aladag et al. [2] introduced a
new approach, which uses feed forward neural networks for defining fuzzy relationships and is based on a high order
fuzzy time series forecasting model.
In this study, a new high order fuzzy time series forecasting model based on feed forward neural networks and
adaptive expectation model is proposed to gain better forecasting accuracy. The forecasts are produced by employing a
high order fuzzy time series model in which fuzzy relationships are defined by utilizing feed forward neural networks.
Then, these obtained forecasts are adapted by using adaptive expectation model in order to reach more accuracy. The
proposed method is employed to forecast the enrollments of the University of Alabama to show the considerable
outperforming results. This well known time series is also analyzed by using the methods proposed by Chen [4],
Huarng [11], Chen [5], Huarng and Yu [12] and Aladag et al. [2]. In addition, the enrollment data is analyzed by using
the conventional methods which are exponential smoothing, ARIMA and time series regression methods. The results
obtained from the proposed method are compared with the results from the methods mentioned above.
Section 2 includes the definitions of first and high order time series forecasting models. Section 3 gives the brief
information related to neural networks. The new proposed method is introduced and the implementation results of the
enrollment data are given in Sections 4 and 5, respectively. Final section concludes the paper.
2. Fuzzy time series
The definition of fuzzy time series was firstly introduced by Song and Chissom [16,17]. In fuzzy time series approach,
there is no need for various theoretical assumptions just as in conventional time series procedures. The most important
advantage of fuzzy time series approach is its ability to work with a very small set of data and no requirement for
linearity assumption. Some general definitions of fuzzy time series are given as follows:
Let U be the universe of discourse, where U ={u1 , u2 , . . ., ub }. A fuzzy set Ai of U is defined as Ai = fAi (u1 )/u1 +
fAi (u2 )/u2 + + fAi (ub )/ub , where fAi is the membership function of the fuzzy set Ai ; fAi : U [0, 1]. ua is a
generic element of fuzzy set Ai ; fAi (ua ) is the degree of belongingness of ua to Ai ; fAi (ua ) [0, 1] and 1 a b.
Denition 1. Fuzzy time series Let Y(t) (t = . . . , 0, 1, 2, . . . ) a subset of real numbers, be the universe of discourse by
which fuzzy sets fj (t) are defined. If F(t) is a collection of f1 (t), f2 (t), . . . then F(t) is called a fuzzy time series defined
on Y(t).
Denition 2. Let F(t) be a fuzzy time series. If F(t) is a caused by F(t 1), then this fuzzy logical relationship is
represented by
F (t 1) F (t)

(1)

and it is called first order fuzzy time series forecasting model.

Denition 3. Let F(t) be a fuzzy time series. If F(t) is caused by F(t 1), F(t 2), . . ., F(t m), then this fuzzy
logical relationship is represented by
F (t m), . . . , F (t 2), F (t 1) F (t)

(2)

and it is called the mth order fuzzy time series forecasting model.
3. Articial neural networks
Artificial neural networks were originally motivated by the biological structures in the brains of humans and animals,
which are extremely powerful for tasks such as information processing, learning and adaptation. The most important

C.H. Aladag et al. / Mathematics and Computers in Simulation 81 (2010) 875882

877

Fig. 1. A broad feed forward neural network architecture.

characteristics of neural networks are large number of simple units, highly parallel units, strongly connected units,
robustness against the failure of single units, and learning from data [15].
Some main elements compose artificial neural networks. Determining the elements of the artificial neural networks
has an important effect on performance of artificial neural networks. Elements of the artificial neural networks are
generally given as network architecture, learning algorithm and activation function [10].
There are various types of artificial neural networks. One of them is called as feed forward neural networks. The
feed forward neural networks have been used successfully in many studies [10]. In the feed forward neural networks,
there are no feedback connections. The broad feed forward neural network architecture that has single hidden layer
and single output is given as an illustration [1] (Fig. 1).
One critical decision is to determine the appropriate architecture, that is, the number of layers, number of nodes
in each layers and the number of arcs which interconnect with the nodes [18]. Aladag et al. [2] employ feed forward
neural network for defining fuzzy relationships. Therefore, our focus is on the feed forward networks. Determining
architecture depends on the basic problem. Since, in the literature, there are no general rules for determining the best
architecture, many architecture should be examined for the correct results [7].
Learning of artificial neural networks for a specific task is equivalent to finding all the values of the weights such that
the desired output is generated to the corresponding input. Various training algorithms have been used for determining
the optimal weight values [7].
The activation function is another element that affects the performance of artificial neural networks. It shows the
relationship between inputs and outputs of a node and a network. In general, the activation function introduces a degree
of the nonlinearity that is valuable for the most artificial neural networks applications [1].
4. The proposed method
In the literature, it is observed that the high order time series models can produce better forecasts than those obtained
from the first order fuzzy time series models. In implementations, however, it is hard to use the high ordered models
since they require usage of fuzzy relationship tables and complex matrix operations. When the artificial neural networks
are used for determining fuzzy relations in the high order models, there is no need to employ these tables and operations.
Therefore, utilizing the artificial neural networks make the process easier. Besides, better results can be obtained.
In stock markets, investors usually make their investment decision according to recent stock information such as
late market news, stock technical indicators, or price fluctuations. Reasonable investors will modify their forecasts with
recent forecasting errors [6]. Cheng et al. [6] and Liu et al. [14] showed that the forecasting accuracy can be increased
by applying adaptive expectation model to defuzzyfied forecasts. At the same time, their studies are based on the first
order fuzzy time series forecasting models.

878

C.H. Aladag et al. / Mathematics and Computers in Simulation 81 (2010) 875882

Table 1
Notations for second order fuzzy time series.
Observation no.

Ft2

Ft1

Ft

Input-1

Input-2

Target

1
2
3
4
5
6

A6
A2
A3
A7

A6
A2
A3
A7
A4

A6
A2
A3
A7
A4
A2

6
2
3
7

2
3
7
4

3
7
4
2

A novel high ordered fuzzy time series forecasting approach in which fuzzy relationships are defined by the
artificial neural networks and forecasts are adjusted by employing adaptive expectation model is proposed in this study.
The proposed methods advantages and properties different from other approaches available in the literature can be
summarized as follows:
The proposed method is more useful since instead of utilizing fuzzy relationships group tables and complex matrix
operations for defining fuzzy relationships, artificial neural networks are used.
The artificial neural networks have nonlinear structure. Therefore, the fuzzy relationships can be defined more
accurately so more accurate forecasts can be obtained.
This is the first time, adaptive expectation model is applied to high ordered fuzzy time series model.
Unlike the studies of Cheng et al. [6] and Liu et al. [14], forecast adjusting is done by using constraint optimization
for weighted parameter at the last step of the proposed approach. Therefore, the forecasting accuracy is increased.
In order to construct high order fuzzy time series model, various feed forward neural network architectures are employed
to define fuzzy relationships in the implementation. The feed forward neural networks architecture, which includes
one hidden layer and one output, is used to define fuzzy relationships. Back propagation learning algorithm is used
to train neural network models and logistic activation function is employed in all neurons. The stages of the proposed
method based on feed forward neural networks and adaptive expectation model are given below.
Stage 1. Dene and partition the universe of discourse.
The universe of discourse for observations, U = [starting, ending], is defined. After the length of intervals, l, is
determined, the U can be partitioned into equal-length intervals u1 , u2 , . . ., ub , b = 1, . . . and their corresponding
midpoints m1 , m2 , . . ., mb , respectively.
ub = [starting + (b 1) l, starting + b l],

mb =

[starting + (b 1) l, starting + b l]
2

Stage 2. Dene fuzzy sets.

Each linguistic observation, Ai , can be defined by the intervals u1 , u2 , . . ., ub .
Ai =

fAi (u1 ) fAi (u2 )

fA (ub )
+
+ + i
u1
u2
ub

(3)

Stage 3. Fuzzify the observations.

For example, a datum is fuzzified to Ai , if the maximal degree of membership of that datum is in Ai .
Stage 4. Establish the fuzzy relationship with feed forward neural network.
An example will be given to explain stage 4 more clearly for the second order fuzzy time series. Because of dealing
with second order fuzzy time series, two inputs are employed in neural network model, so that lagged variables Ft2
and Ft1 are obtained from fuzzy time series Ft . These series are given in Table 1. The index numbers (i) of Ai of
Ft2 and Ft1 series are taken as input values whose titles are input-1 and input-2 in Table 1 for the neural network
model. Also, the index numbers of Ai of Ft series are taken as target values whose title is target in Table 1 for the neural
network model. When the third observation is taken as an example, inputs values for the learning sample [A6 , A2 ] are
6 and 2. Then, target value for this learning sample is 3.
Stage 5. Defuzzify results.

C.H. Aladag et al. / Mathematics and Computers in Simulation 81 (2010) 875882

879

The defuzzyfied forecasts are middle points of intervals which correspond to fuzzy forecasts obtained by neural
networks in the previous stage.
Stage 6. Adopt the defuzzyed forecasts.
The forecasts obtained in the previous stage are adapted by utilizing adaptive forecasting equation given in (4).
t
Adaptive Forecast
= P(t 1) + (F (t) P(t 1))
(4)

where P(t 1) represents the observed value at time t 1 and F(t) is the defuzzyfied forecast calculated in Stage 5
for time t. value ( (0,1)) is determined by constrained single variable optimization. The root mean square error
(fRMSE () or RMSE) over the adaptive forecasts and the observed values is calculated as follows:


T
2 1/2
t (P(t) Adaptive Forecast(t/))
fRMSE () =
T
where T is the number of the test data. Thus, an value makes fRMSE () minimum which is chosen by solving the
optimization problem defined below.
minfRMSE (t)

(5)

subject to 0 < < 1

In the optimization process, we used a MATLAB function called fminbnd to minimize RMSE value. The function
fminbnd is used to find minimum of single-variable function on fixed interval. It finds a minimum for a problem
specified by
minf (x)
x

subject to x1 < x < x2

x, x1 , and x2 are scalars and f(x) is a function that returns a scalar. In MATLAB, x = fminbnd(f (x), x1 , x2 ) returns
a value x that is a local minimum of the scalar valued function f(x) in the interval x1 < x < x2 . In other words, to find the
minimum of the function f(x) on the interval (x1 , x2 ),
a = fminbnd(f (x), x1 , x2 )
can be used in MATLAB. f(a) gives the local minimum value in the interval (x1 , x2 ). The algorithm used by fminbnd
is based on golden section search introduced by Kiefer [13] and parabolic interpolation. Unless the left endpoint x1 is
very close to the right endpoint x2 , fminbnd never evaluates f(x) at the endpoints, so f(x) need only be defined for x in
the interval x1 < x < x2 . If the minimum actually occurs at x1 or x2 , fminbnd returns an interior point at a distance of
no more than 2 TolX from x1 or x2 , where TolX is the termination tolerance. See Brent [3] or Forsythe et al. [9] for
details about the algorithm.
To solve the optimization problem given in (5), we use the function fminbnd as follows:
= fminbnd(fRMSE (), 0, 1)
This function used in MATLAB finds an * that makes fRMSE () minimum in the interval 0 < < 1.
5. Application to the enrollment data
The proposed method is applied to the enrollment data of University of Alabama which is shown in Table 2. The
enrollment observations from 1971 to 1988 were used for the estimation, while the observations from 1989 to 1992
were used as a test data. Second, third, and fourth order fuzzy time series model are used in the implementation. In
the first stage of the proposed method, as Huarng [11] did, the lengths of intervals are chosen as 200, 300, 400, 500,
600, 700, 800, 900 and 1000. After following stages 2 and 3 in the method given in Section 4, the number of neurons
of hidden layer is altered 1 through 4 not to lose generalization ability of neural network model. Table 3 gives results
of the proposed method for second through fourth high order fuzzy time series and rounded RMSE values for the test
data. For each RMSE value given in Table 3, a corresponding optimal value exists. According to Table 3, by using

880

C.H. Aladag et al. / Mathematics and Computers in Simulation 81 (2010) 875882

Table 2
The enrollment data.
Years

Actual

Years

Actual

1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981

13,055
13,563
13,867
14,696
15,460
15,311
15,603
15,861
16,807
16,919
16,388

1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992

15,433
15,497
15,145
15,163
15,984
16,859
18,150
18,970
19,328
19,337
18,876

Table 3
RMSE values obtained from the proposed method.
Order

Hidden layer number

1
2
3
4
1
2
3
4
1
2
3
4

Length of interval
200

300

400

500

600

700

800

900

1000

499.2
503.1
225.0
503.3
332.2
225.0
503.4
502.0
503.3
503.2
494.7
503.2

455.7
269.2
269.2
503.2
503.3
269.2
503.3
503.2
503.3
503.3
501.9
494.0

472.8
503.2
202.6
503.3
503.2
202.6
503.3
104.5
502.2
202.6
502.2
202.6

383.8
444.5
444.5
503.2
80.3*
212.0
503.3
212.0
503.3
503.3
212.0
212.0

500.4
225.0
225.0
501.5
499.0
225.0
503.3
462.7
503.2
502.0
503.3
503.3

246.0
503.2
503.2
502.8
317.9
317.9
503.3
482.1
317.9
317.9
492.0
494.8

239.7
239.7
459.3
503.2
503.2
239.7
503.3
434.2
503.2
413.7
503.3
239.7

503.1
492.8
503.3
503.2
420.1
336.2
503.3
503.3
503.3
503.3
336.2
492.8

466.4
467.4
283.0
496.4
503.3
283.0
503.2
488.3
283.0
503.3
503.3
494.4

Table 4
The comparison of the results.
Method

Order

RMSE

Chen [4]
Huarng [11]a
Huarng [11]b
Chen [5]
Huarng and Yu [12]
Aladag et al. [2]
The proposed method

1
1
1
2
1
3
3

575.0
529.4
500.0
738.4
396.1
153.1
80.3

a
b

Average based method.

Distribution based method.

the proposed model, the best forecasts are obtained when 3rd order model, 500 length of interval and, 1 unit in hidden
layer are employed. For this case, obtained RMSE and optimal value are 80.3 and 0.7761, respectively.
For the aim of a comparative study, methods that Chen [4], Huarng [11], Chen [5], Huarng and Yu [12] and, Aladag
et al. [2] are also applied to the data of the enrollments of the University of Alabama and the obtained results are
summarized in Table 4. Except from the method proposed by Huarng and Yu [12], the length of interval is changed
between 200 and 1000 for the rest of all the methods. In addition, when the methods proposed by Chen [5] and Aladag
et al. [2] are employed, the order of models is taken as 2, 3 and 4.

C.H. Aladag et al. / Mathematics and Computers in Simulation 81 (2010) 875882

881

Table 5
The results obtained from conventional time series forecasting methods.
Method

RMSE

Brown exponential smoothing

Logarithmic regression
Inverse regression
Quadratic regression
Cubic regression
Compound regression
Power regression
S-curve regression
Growth regression
Exponential regression
ARIMA(0,1,0) method

2,469,727
2,332,806
3,001,306
2,865,191
2,412,646
1,605,017
3,009,670
1,605,010
1,605,017
1,605,017
477,468

Fig. 2. The forecasts obtained from the proposed method and the enrollment data.

The enrollment data also forecasted by employing conventional time series methods such as exponential smoothing,
ARIMA and various time series regression methods. In calculations, SPSS 15.0 is used. RMSE values calculated over
the test set are presented in Table 5.
According to Tables 4 and 5, the result of our proposed method has the smallest RMSE value when compared
with the other methods. In other words, the most accurate forecasts are obtained when the proposed fuzzy time series
approach is used. Thus, it can be said that the new proposed method produces better forecasts. In addition, to show the
forecasting results visually, the forecasts obtained from the proposed method and original data are drawn in Fig. 2.
6. Conclusions
An important issue in fuzzy time series approaches is to determine fuzzy relationships. Defining fuzzy relationships
in high order fuzzy time series approach is more complicated and computationally more expensive than that in first
order fuzzy time series approach when fuzzy logical relationship tables are utilized. In order to avoid this complexity,
artificial neural networks can be used to define fuzzy relationships in high order fuzzy time series. Aladag et al. [2]
proposed a high order fuzzy time series approach in which fuzzy relationships are determined by feed forward neural
networks. They also show that forecasting accuracy is increased when the feed forward neural networks are employed.
Cheng et al. [6] improved the forecasting accuracy by using the adaptive expectation model to adopt forecasts
obtained from the first order fuzzy time series approach. In this study, we propose a new high order fuzzy time series
forecasting method in which fuzzy relationships are defined by feed forward neural networks and forecasted values
are adopted by the adaptive expectation model. The proposed method is employed to forecast the well known data
of the enrollments of the University of Alabama to show the considerable outperforming results. Also, the methods
proposed by Chen [4], Huarng [11], Chen [5], Huarng and Yu [12], Aladag et al. [2] and conventional time series
methods are applied to the data for comparison. In the end of the comparison, it is obviously seen that the proposed
method produces better forecasts than those of other methods. These results indicate that utilizing feed forward neural

882

C.H. Aladag et al. / Mathematics and Computers in Simulation 81 (2010) 875882

networks for defining fuzzy relations and the adaptive expectation model for adjusting forecasts significantly improve
the forecasting accuracy in high order fuzzy time series.
Acknowledgment
We would like to thank the referee for the valuable comments, which provided insights that helped improve the
paper.
References
[1] C.H. Aladag, E. Egrioglu, S. Gunay, A new architecture selection strategy in solving seasonal autoregressive time series by artificial neural
networks, Hacettepe Journal of Mathematics and Statistics 37 (2) (2008) 185200.
[2] C.H. Aladag, M.A. Basaran, E. Egrioglu, U. Yolcu, V.R. Uslu, Forecasting in high order fuzzy time series by using neural networks to define
fuzzy relations, Expert Systems with Applications 36 (2009) 42284231.
[3] R.P. Brent, Algorithms for Minimization without Derivatives, Prentice-Hall, Englewood Cliffs, NJ, 1973.
[4] S.M. Chen, Forecasting enrollments based on fuzzy time-series, Fuzzy Sets and Systems 81 (1996) 311319.
[5] S.M. Chen, Forecasting enrollments based on high order fuzzy time series, Cybernetics and Systems 33 (2002) 116.
[6] H.C. Cheng, T.L. Chen, H.J. Teoh, C.H. Chiang, Fuzzy time-series based on adaptive expectation model for TAIEX forecasting, Expert Systems
with Application 34 (2008) 11261132.
[7] E. Egrioglu, C.H. Aladag, S. Gunay, A new model selection strategy in artificial neural networks, Applied Mathematics and Computation 195
(2008) 591597.
[8] E. Egrioglu, C.H. Aladag, M.A. Basaran, V.R. Uslu, Y. Yolcu, A new approach based on the optimization of the length of intervals in fuzzy
time series, Journal of Intelligent and Fuzzy Systems (in press).
[9] G.E. Forsythe, M.A. Malcolm, C.B. Moler, Computer Methods for Mathematical Computations, Prentice Hall, 1976.
[10] S. Gunay, E. Egrioglu, C.H. Aladag, Introduction to Single Variable Time Series Analysis, Hacettepe University Press, 2007.
[11] K. Huarng, Effective length of intervals to improve forecasting in fuzzy time-series, Fuzzy Sets and Systems 123 (2001) 387394.
[12] K. Huarng, H.K. Yu, The application of neural networks to forecast fuzzy time series, Physica A 363 (2006) 481491.
[13] J. Kiefer, Sequential minimax search for a maximum, Proceedings of the American Mathematical Society 4 (1953) 502506.
[14] J.W. Liu, T.L. Chen, C.H. Cheng, Y.H. Chen, Adaptive-expectation based multi attribute FTS model for forecasting TAIEX, Computers and
Mathematics with Applications (2010) 795802.
[15] N. Oliver, Nonlinear System Identification, Springer-Verlag, Berlin/Heidelberg/New York, 2001.
[16] Q. Song, B.S. Chissom, Fuzzy time series and its models, Fuzzy Sets and Systems 54 (1993) 269277.
[17] Q. Song, B.S. Chissom, Forecasting enrollments with fuzzy time seriesPart I, Fuzzy Sets and Systems 54 (1993) 110.
[18] J.M. Zurada, Introduction of Artificial Neural Systems, West Publishing, St. Paul, 1992.