MBE, 17(6): 7151–7166.

DOI: 10.3934/mbe.2020367
Received: 22 July 2020
Accepted: 10 October 2020
Published: 21 October 2020
http://www.aimspress.com/journal/MBE

Research article

A hybrid model combining variational mode decomposition and an

attention-GRU network for stock price index forecasting

Hongli Niu* and Kunliang Xu

School of Economics and Management, University of Science and Technology Beijing, Beijing
100083, China

* Correspondence: Email: [email protected].

Abstract: In this paper we introduce a new hybrid model based on variational mode decomposition
(VMD) and Gated Recurrent Units (GRU) network improved by attention mechanism to enhance
the accuracy of stock price indices forecasting. In the process of establishing the model, VMD is
made a use to decompose the primary series into some almost orthogonal subsequences. The
attention mechanism is introduced into GRU to assign different weights to the input elements in
advance so that better predictive results can be achieved for each component. In empirical
experiment, London FTSE Index (FTSE) and Nasdaq Index (IXIC) are adopted to examine the
performance of VMD-AttGRU model. Empirical results report that the developed hybrid model
outperforms the single models and indeed raises the accuracy of stock price indices forecasting. In
addition, the introduction of attention mechanism can increase the level predictive accuracy but
decrease the correctness of direction forecasting.

Keywords: variational mode decomposition; Gated Recurrent Units; attention mechanism; forecasting;
stock price

1. Introduction

As stock markets gradually enter the public vision, the precise prediction of stock price indices
has become one of the most promising research projects in forecasting of time series. The commonly
used forecasting methods are simply divided into two classes: econometric methods and artificial
intelligence (AI) based models. The latter, represented by artificial neural networks (ANNs), have been
proved to outperform the econometric methods in dealing with non-stationary and non-linear time
 7152

series [1–4]. As an improvement of traditional ANN, recurrent neural networks (RNN) [5] establish
connections between the hidden layer units, through which the dependency of data at different time
points can be further explained. The before-after associated structure ensures that RNN is especially
suitable for predicting time series data [6]. By introducing three gate mechanism into the hidden units
of traditional RNN, long short-term memory (LSTM) network overcomes the short comings existing
in RNN, such as gradient disappearing and exploding in long time span [7]. Recently, LSTM has been
widely utilized to predict time series and obtained outstanding results [8,9]. Gated recurrent units
(GRU) network integrates the three gates of LSTM into reset gate and update gate, which effectively
improves the computing efficiency of LSTM [10], and GRU achieved better results than LSTM in
different time series forecasting tasks [11,12]. In this paper, the attention mechanism is introduced to
assign weights to different input elements of GRU and obtain a more precise forecasting result.
To further improve the forecasting accuracy of stock price indices, hybrid models containing two
or more individual models have been developed gradually, in which the unique advantages of different
individual models can be exploited. Following “Divide-and-Conquer” principle, “Decomposition-and-
Ensemble” is a typical framework employed in time series forecasting [13], the main idea of which is
to decompose an raw complex sequence into several subseries with simple patterns so as to establish
a prediction model for every subseries, and the final result is concluded by summing up the prediction
results of the subseries [14]. Based on the excellent performance, hybrid forecasting models are
becoming the mainstream gradually [15]. As a novel multiresolution technique originated from signal
processing, variational mode decomposition (VMD) [16] is a completely non-recursive algorithm that
can decompose the original series into multiple components with a specific bandwidth in the spectral
domain. It has been proved that VMD performs better than the models of the same kind, such as
Empirical mode decomposition (EMD) [14], in noise robustness and component decomposing
accurately. In recent years, the hybrid models based on VMD have been applied successfully in several
fields. For instance, by integrating the VMD with classical ANNs, Lahmiri [17] established a
forecasting model VMD-PSO-BPNN for intraday stock prices prediction. The experimental results in
terms of six stocks suggests that the hybrid model performs better than the single PSO-BPNN model
significantly. However, there is no methodology regarding optimal selection of the number of
subcomponents of VMD. In his follow-up research [18], the newly proposed model VMD-GRNN
demonstrates higher accuracy than the EMD-based forecasting models in the predictions of WTI oil
prices, CANUS exchange rate and NASDAQ 100 VIX when the parameter of subcomponents number
ranges from 6 to 12. The similar results are proved in [19], in which the VMD is combined with a
GRNN optimized by particle swarm optimization (PSO) and the hybrid model is established to predict
the California electricity and Brent crude oil prices. The performances of EMD and VMD-based
models are assessed and the number of subseries of VMD is set to be the same as EMD. The above
researches have confirmed the applicability and superiority of VMD in practice, but it still has some
room for improvement: Firstly, the optimal number of components decomposed by VMD is still
difficult to be determined, but the empirical results of literature [18] have indicated that the forecasting
quality of the VMD-based models will vary with the change of component number decomposed.
Secondly, the above-mentioned forecasting models are all classical ANNs, which can be replaced with
the promising RNNs, such as RNN, LSTM and GRU, to further enhance the forecasting ability. Thirdly,
the evaluating metrics are only limited to error measures without considering the capability of correctly
predicting the moving direction of the time series, which is of great significance in the short-term
prediction of financial time series data.

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7153

Combining the advantages of GRU, VMD network and other variant models, there have been
several literatures utilizing the hybrid forecasting models to implement various prediction tasks. For
example, Zhu et al. [20] employed a hybrid model integrating VMD and BiGRU network to forecast
the daily natural rubber futures price and volatility, validing the effectiveness of this model. The result
indicated that the improvements in prediction performance largely depended on the time-scale
matching degree between the predicted target and the mode sub-series. Li et al. [21] introduced an
error correction strategy into VMD-GRU hybrid model to enhance the model performance in wind
speed interval prediction, and the experiments based on eight cases from two wind fields demonstrated
the proposed model is a highly qualified forecasting method. By combining GRU with VMD, Wang
et al. [22] adopted a hybrid model for addressing the wind power interval prediction problem and
proposed an optimization method based on constructed intervals for building high-quality training
labels before applying the Adam algorithm for full training, and the effectiveness of the VMD-GRU
was confirmed in comparison with other models. However, it is worth noting that the historical
elements input into the forecasting network play different roles when predicting the target value in
time series. In general, the impact of the input values closer to the target value is greater than that of
the farther time points. Moreover, the optimal number of components needs to be preset in VMD,
which is important to improve the accuracy of the final prediction result. In this work, after
decomposing the original time series into an optimal number of subseries according to a certain
standard the ratio of residual energy (rres) by VMD, an attention mechanism is introduced into the
GRU network to enhance the forecasting quality by assigning different weights to the input elements.
The contribution of this paper to the literature is to propose a novel hybrid model for the reliable
stock price indices time series prediction, namely, London FTSE Index (FTSE) and Nasdaq Index
(IXIC). The evaluations indicate that compared with the counterparts, including the single models and
the traditional GRU-based models, the proposed VMD-AttGRU model presents more accurate and
robust results demonstrated by the level forecasting indices. The introduction of attention mechanism
in the hybrid model VMD-GRU decreases the forecasting error while slightly reduce the ability of this
model to correctly predict the direction.

2. Methodologies

2.1. Variational mode decomposition (VMD)

Variational mode decomposition (VMD) is a non-recursive and adaptive data decomposition

technique developed recently [16]. VMD is utilized, in the VMD-AttGRU model, to decompose the
original stock index , 1, 2, … , into n components, , 1,2, … , , which stands for
different local vibrations ranging from high frequency to low frequency. Each mode need to
compact around a center frequency mostly. The bandwidth of a mode can be estimated by follows:
At first, for each mode , the Hilbert transform is employed to calculate the correlation analysis data
and a unilateral frequency spectrum is obtained. Then, for each mode , the spectrum of mode is
transmitted to the baseband by exponential mixing with the pulses tuned to their respective centers.
Afterwards, the Gaussian smoothness of the demodulated series is used to calculate the bandwidth.
The constraints of variational problem can be expressed in the following way:

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7154

, ∗
(1)
. .

where , ,… and , ,… respectively denote the set of the

subcomponent and its corresponding central frequency. indicates the differential processing of
t, ‖ ‖ indicates the norm processing, represents the Dirac function, and * denotes the
convolution symbol.
To solve the optimization problem of constrained variational decomposition, an augmented
Lagrangian function is introduced:

, , ∗

(2)

, ,

in which denotes the penalty parameter, and is the Lagrangian multiplier. In order to obtain
the saddle point of the above formula, which also is the solution of the original constraint conditions,
VMD adopts the Alternate direction method of multipliers (ADMM) [23].
Prior to VMD, the number of components n should be properly determined in advance. If the
number is large, additional computing resources will be occupied, but if n is small, it may lead to
an insufficient decomposition and inaccurate forecasting results finally. The ratio of residual energy
rres to original data sequence energy is used to determine the optimal number, which can be
formulated as follows:

1 ∑
(3)

where rres is the residual after decomposition, which can be used as the optimization index of VMD
process. In empirical, when rres is smaller than 1% or there is no obvious trend of downwards, the
component number can be defined [24].

2.2. Long short-term memory network and gated recurrent unit network

The long short-term memory (LSTM) network [8] creatively introduces the “gate” mechanism to
improve the conventional recurrent neural network (RNN): it replaces the hidden layer nodes of the
RNN with special memory cells. Each memory cell contains three gates: input gate , forget gate ,
and output gate that implement the filtering and processing of historical states and information,
and the problems of gradient disappearance and explosion can be effectively resolved. The LSTM has
been successfully applied in time series prediction [8,9]. The gated recurrent unit (GRU) network [10]
integrates the three gates of the LSTM into two gates: reset gate and update gate and achieves
better performance in time series forecasting tasks [25]. The reset gate measures how much the
historical information will be kept at this moment and how much the latest information will be added,

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7155

which helps to grasp the dependency of short-term existing in the series data, while the update gate
determines the degree of “forgetting” historical information, and the information with arbitrary-lengths
of the input can be memorized in this gate effectively. The basic steps of GRU can be shown in
the following:
At first, the reset gate and update gate at the current state (time t) are established by the
latest input and the hidden state produced by the previous cell , and the outputs of the two
gates are respectively given as:
(4)

(5)
Secondly, the current candidate hidden state can be formulated:
∗ (6)
Finally, the outcome of current hidden state can be computed by implementing the linear
combination of the current candidate hidden state and the previous hidden state , where the
sum of weighting coefficient is equal to 1.

1 ∗ ∗ (7)
where , , and , , represent the appropriate weight coefficient matrices, ,
and denote the corresponding bias vectors, and are the Sigmoid function and
Hyperbolic tangent function respectively, and * indicates the dot multiplication between matrices.

2.3. Attention mechanism

Attention mechanism is originated from a fact that human brain focuses on only specific parts of
their visual view when recognizing something [26]. For predicting time series, there is a fact that not
all elements in the input series contribute equally to the value of context vector at each time step t,
which is often ignored by the conventional forecasting networks. Therefore, the principle of attention
mechanism built in neural network is to select crucial elements and give more weight to them, rather
than taking all elements into account equally. That is, the attention mechanism is a deep learning
algorithm for identifying the most relevant inputs. After ignoring the irrelevant information and
amplifying the needed information, the processing efficiency of input information is greatly improved.
Recently, the attention mechanism has been applied in computational neuroscience [27], text
representations [28] and image description [29] successfully. Figure 1 depicts the calculation of
attention value in three steps, through which different weights are assigned to the elements of
input series to highlight the important subset of its inputs by training the model at different time. Every
element of the input data set is assumed to contain an address (Key) and a value (Value). The given
goal is denoted as G and the attention weight is the result to be calculated. In the figure, F (G, Key) is
adopted to calculate the relevancy between the given target G and address K. and (i = 1, 2, ...,
m) represent respectively the relevance and weight of attention for the element of input sequence
at time t. The realization of attention mechanism can be formulated as follows:
Attend , , (8)

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7156

exp
(9)
∑ exp

(10)
where denotes the attention score that is defined by input data , previous state and weight
of previous attention.
The specific implementation process of attention mechanism utilized in this work is referred to [30].
That is, in the first step, the relevancy between every previous input elements and output elements are
computed. Then, applying the softmax formula to convert the relevancies into the probability form.
Lastly in the third step, multiply the obtained probabilities by the implicit expression of the
corresponding input feature, to make it stand for the feature contribution to the forecasted load and
sum up all the input contribution features to be the input section to forecast the next load value.

Figure 1. Three steps of attention value calculation.

2.4. VMD-AttGRU network

In view of the advantages of VMD, attention mechanism and GRU network, we construct a hybrid
model named VMD-AttGRU by combining the three techniques. In this model, the VMD is utilized
to decompose the original time series into several components. The Attention-GRU (simplified as
AttGRU) is used to establish forecasting model for each component and obtain the predicted output
separately, in which the GRU layer takes the output of the attention layer as the input so that the
capability of conventional GRU network is improved. The final forecasting result is calculated by
summarizing the separate predicted outputs obtained by AttGRU. The flow chart in Figure 2 depicts
its implementation process, in which the VMD-AttGRU operation is carried out as follows.

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7157

Step 1: The VMD is utilized to decompose the stock price index series ( ), = 1, 2, ⋯ , into n
mutually independent subseries, denoted by IMF1, IMF2, ⋯ IMFn, in which the n is determined by a
specific standard. The initial series is reconstructed in terms of the IMFs as:

Step 2: Each component IMF is split into training and test datasets at a fixed ratio, and the input and
output sets are split according to the step size. The AttGRU network is utilized to train and establish the
forecasting model based on the training dataset. The forecasting output of each IMF is obtained.
Step 3: The final predicted result of the original stock price index series is calculated by
summarizing the separate predicted outputs.
Step 4: Multiple performance measures, i.e., MAE, RMSE, MAPE, TIC, and stat, are adopted to
evaluate the prediction capacity of VMD-AttGRU from different perspectives.

Figure 2. The structure of VMD-AttGRU.

3. Data selection and processing

In this work, the daily closing price of London FTSE Index (FTSE) and Nasdaq Index (IXIC) are
used to examine the validity of the proposed VMD-AttGRU model. The selected two stock price
indices are both representative in the global stock markets and regarded as important benchmarks of
social and economic development. They are collected from the global important stock price indices of
Wind database, which stored in the form of [date, price] time series. The FTSE cover the time period
from 2007/03/09 to 2020/06/05, which accounts 3348 data points, and the IXIC cover the time period
from 2007/02/20 to 2020/06/05, which also account 3348 data points. To conduct experiments, the
first 80% of each sample is used to train the model, and the remaining 20% is used as test sets. Figure 3
displays the curves of price samples of FTSE and IXIC. Table 1 exhibits the details illustration of the

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7158

selected two stock price indices. Table 2 shows the descriptive statistic information of the samples in
terms of mean, standard deviation, skewness, kurtosis, Jarque-Bera (JB) test for normality and
Augmented Dickey Fuller (ADF) test for stationarity. It is shown that with the standard deviation value
of 2114.71 for IXIC and 894.81 for FTSE, the IXIC has more volatility than the FTSE. The FTSE is
negatively skewed with skewness value of −0.56 while the IXIC is positively skewed with skewness
of 0.67. Both of them have kurtosis less than 3, implying no leptokurtosis. The results of JB test
indicate that both FTSE and IXIC price index series are distinctly non-Gaussian distributed at the 5%
confidence level. The results of ADF test suggests the significantly non-stationary of both prices.
To reduce the impact of noise and facilitate optimize the solving process, each component
, 1,2, … obtained by VMD will be normalized to the range of [0,1] by the following
maximum and minimum standardized formula:
min
′ (11)
max min

Then the normalized data is input into the AttGRU network for training and prediction. In order to
obtain the real predictive value and compare it with the actual value intuitively the normalized output
can be reverted to x(t) after prediction as follows:

max min min (12)

Table 1. Datasets of the selected stock price indices for forecasting.

FTSE IXIC
Time period 2007/03/09 ~ 2020/06/05 2007/02/20 ~ 2020/06/05
Total number 3348 3348
Train sets 2007/03/09 ~ 2017/10/13 2007/02/20 ~ 2017/10/05
Train number 2678 2678
Test sets 2017/10/16 ~ 2020/06/05 2017/10/06 ~ 2020/06/05
Test number 670 670

Figure 3. Daily closing prices of FTSE and IXIC stock indices.

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7159

Table 2. Descriptive statistics of the FTSE and IXIC.

Index Mean Std. Skewness Kurtosis JB test ADF test
FTSE 6273.01 894.81 -0.56 2.90 177.20* (0.00) −0.25 (0.56)
IXIC 4320.50 2114.71 0.67 2.28 323.25 *(0.00) 2.13 (0.99)

4. Performance evaluation metric

We would like to better validate the robustness of the prediction network of VMD-AttGRU, this
work adopts five commonly-used criteria to examine the superiority of the model from the various
perspectives. They are including the mean absolute error (MAE), root mean square error (RMSE),
mean absolute percentage error (MAPE), Theil Inequality Coefficient (TIC) and directional statistic
Dstat, in which the first four indices are employed to measure the level forecasting accuracy and the
Dstat is employed to measure the correctness of predicted direction for a time series in terms of
percentage. They are respectively defined as follows:
1
| | (13)

1
(14)

1
(15)

1
∑
(16)
1 1
∑ ∑

1
100% ∗

(17)
1, if 0, 1
0, otherwise
where expresses the actual value, signifies the forecasting value, N is the length of sample of
forecasting results, the same applies hereinafter. The MAE is used to measure the average absolute
error between the actual series and the predicted series. The RMSE, which is more sensitive to outliers,
is used to measure the deviation between the actual and the predicted series. The MAPE is designed
to compute the average relative errors between the actual series and the predicted series in terms of
percentage, while the directional statistic Dstat is adopted to evaluate the capability of correctly
predicting the moving direction of the time series. In general, the smaller value of the MAE, RMSE,
MAPE and TIC indicates the less difference between the forecasting and the actual values, that is, the

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7160

more accuracy of the prediction of the model. The higher Dstat value corresponds to the better
performance of the model.

5. Empirical results

In this section, the predictive performance of VMD-AttGRU model for stock price indices
forecasting is analyzed. To comprehensively demonstrate the advantages of the proposed hybrid model
and the effectiveness of the attention mechanism in stock price index prediction, single models (LSTM,
GRU, AttGRU) and hybrid model VMD-GRU are considered for comparison. According to the
“decomposition and ensemble” strategy, at first the prices are decomposed by VMD technique, in
which the number of subseries IMFs should be determined first. Table 3 displays the ratio of residual
energy rres in VMD approach under different n for the stock price indices. All rres are below 1%. In
FTSE, the downward tendency of rres tends to be stable when n is larger than 15, while the descending
tendency of rres tend to be stable when n is larger than 16 in IXIC. Therefore, the suitable number of
components in FTSE is set 15, and that in IXIC is set 16.
Taking FTSE as an example, Figure 4 displays the subseries obtained by VMD. They are listed
ranging from high to low frequency, depicting different local oscillations embodied in the data series.
It can be seen intuitively that the decomposed subseries is more regular than the original series, which
helps to reduce the complexity of datasets to be forecasted. Among them, the high frequency
components with relatively small values reflecting the detailed short-term volatilities information of
the original price series, and the low frequency components composed of large values represent the
whole changes of tendency of the daily closing prices.

Table 3. The ratio of residual energy under different n.

n FTSE IXIC
5 0.54% 0.68%
6 0.45% 0.54%
7 0.40% 0.42%
8 0.36% 0.34%
9 0.31% 0.30%
10 0.27% 0.28%
11 0.23% 0.25%
12 0.20% 0.23%
13 0.17% 0.22%
14 0.15% 0.21%
15 0.13% 0.17%
16 0.12% 0.14%
17 0.11% 0.14%
18 0.10% 0.13%

Later, the corresponding AttGRU prediction model is constructed for each composed IMF
subseries. In parameters setting, A historical lag of order 5 is taken to predict the data of the next
period, considering there are 5 trading days per week that can be regarded as a cycle simply. In other
word, the number of input data points is set to 5 and that of outputs is set to 1. After repeated

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7161

experiments, a 5 × 50 × 1 neural network is obtained by setting the number of hidden nodes to 50.
For convenience, set the number of epochs to 300 and the batch size to 64. It should be noted that all
of the processes are implemented in Python 3.x running on a Quad-Core Intel Core i5 processor
operating at 1.40 GHz with an 8 GB installed RAM.

Figure 4. The subseries of FTSE obtained by VMD.

Figure 5. The AttGRU forecasting results for each subseries for FTSE data in the test set.

Figure 5 shows the comparison of the actual value and forecasted value by AttGRU for each
subseries in the FTSE test set. It shows that the predicted curve is very close to the real curve of each
subseries, demonstrating that the AttGRU network can make an accurate prediction of components
with different frequency information.
Figure 6 shows the results for VMD-AttGRU for the two stock price indices test sets along with
the other considered models: LSTM, GRU, AttGRU, and VMD-GRU. Overall, for both stocks, the

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7162

curves are close together, showing that the predicted curve of each model is near the real price curve.
The curve for the VMD-AttGRU model is generally the closest to the actual curve, indicating the best
prediction performance in this comparison. This can be further observed in the inset plots, where a
certain volatile part of the datasets is magnified. So, we can conclude that the VMD-AttGRU model
has the highest accuracy for stock price prediction.

Figure 6. Forecasting results and error of different models for the stock price indices.

Table 4. Predictive performance of different models for FTSE data.

Models MAE RMSE MAPE (%) TIC Dstat Time (s)
LSTM 65.114 91.666 0.943 0.0063 50.00% 42.136
GRU 63.334 90.061 0.918 0.0062 49.70% 45.844
AttGRU 53.185 78.302 0.776 0.0054 49.10% 48.012
VMD-GRU 37.725 46.339 0.551 0.0032 98.19% 687.125
VMD-AttGRU 24.802 39.683 0.375 0.0027 98.04% 744.332

Table 5. Predictive performance of different models for IXIC data.

Models MAE RMSE MAPE (%) TIC Dstat Time (s)

LSTM 113.998 155.707 1.423 0.0100 49.85% 42.887

GRU 91.874 133.213 1.176 0.0085 50.00% 45.878

AttGRU 87.607 131.814 1.132 0.0084 49.10% 49.032

VMD-GRU 83.503 107.762 1.012 0.0069 98.19% 731.371

VMD-AttGRU 65.925 94.245 0.858 0.0060 94.88% 798.146

In order to further analyze the performance of various models, the predictive errors are also
presented in Figure 4. It can be seen that the upper and lower bounds are not much different for single
models. The prediction errors of single models are evidently larger than those of the hybrid model. The
median of the VMD-AttGRU model is closest to 0, and the absolute values of the upper and lower
quartiles are the smallest in the comparison group. The results further show that the relative error of
the target model is relatively smaller and more concentrated, illustrating the better performance of the
proposed model in stock price series data.

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7163

120 200
LSTM LSTM
100 GRU GRU
AttGRU 150 AttGRU
80 VMD-GRU VMD-GRU
VMD-AttGRU VMD-AttGRU

RMSE
MAE

60 100

40
50
20

0 0
FTSE IXIC FTSE IXIC

1 2
LSTM
LSTM
GRU
0.8 GRU
1.5 AttGRU
AttGRU
VMD-GRU
VMD-GRU
0.6 VMD-AttGRU
VMD-AttGRU

MAPE
Dstat

1
0.4

0.5
0.2

0 0
FTSE IXIC FTSE IXIC

0.012
LSTM
0.01 GRU
AttGRU
0.008 VMD-GRU
VMD-AttGRU
TIC

0.006

0.004

0.002

0
FTSE IXIC

Figure 7. Forecasting performance evaluation of different models for FTSE and IXIC data.

To quantitatively measure the predictive performance of each model, the evaluation criteria MAE,
RMSE, MAPE, TIC, Dstat and processing time are calculated in Tables 4 and 5, and the bar graphs are
given in Figure 7. It can be observed that:
1) The hybrid forecasting models following the decomposition-and-ensemble strategy outperform
the single models comprehensively, especially for the directional statistic Dstat, which is approximately
at a level of 50% in single models but is improved by more than 40% after combining with VMD. For
error-type performance measures including MAE, RMSE, MAPE, and TIC, the values of the VMD-
based models are all smaller than single models, which also verifies the superior performance of the
hybrid models in stock price index forecasting.
2) When introducing the attention mechanism to the GRU network, the error-type performance
measures obviously decrease, indicating an improvement of forecasting accuracy. Taking MAPE for
the FTSE data as an example, the MAPE of GRU is 0.918, while that of AttGRU is 0.776, reduced
by 15.46%. VMD-GRU has a MAPE value of 0.551 and VMD-AttGRU has a value of 0.375, reduced
by 51.57%. However, the accuracy measured by Dstat decreases for both FTSE and IXIC data after
adding the attention mechanism. Specifically, the value for AttGRU and VMD-AttGRU is smaller than
that for GRU and VMD-GRU respectively. Considering that the final predicted result is determined by
the linear summation of predicted results of different IMFs and the forecasting quality of each IMF

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7164

affects the final result largely, Figure 8 further exhibits the comparison of Dstat value of each IMF
predicted by AttGRU and GRU for the VMD-based hybrid model, in which the Dstat values of different
IMFs predicted by GRU are generally higher than that predicted by AttGRU for both FTSE and IXIC
series. These all indicate that the introduction of the attention mechanism does not improve the
prediction accuracy in terms of direction.
3) The prediction precision of the proposed VMD-AttGRU model appears to be significantly
higher than other compared models except for Dstat For FTSE and IXIC data, the values of Dstat for
VMD-GRU are both the largest, reaching 98.19%, while those for VMD-AttGRU are 98.04 and
94.88%, respectively, which are 0.15 and 3.31% lower than the largest predicted by VMD-GRU.
4) The processing time of hybrid models is significantly longer than that of single models,
meaning that the process of establishing and training the forecasting models for each IMF takes longer
time. In the comparison of AttGRU with GRU as well as VMD-AttGRU with VMD-GRU, the
introduction of attention mechanism layer also leads to a longer processing time. Compared with the
LSTM, the processing time of GRU is relative shorter for both FTSE and IXIC, indicating the
processing speed by the gates of each hidden layer unit in GRU is slower than that in LSTM.
In brief, following the “Divide-and-Conquer” principle, on the one hand, the proposed hybrid
model VMD-AttGRU can improve the forecasting accuracy in terms of error-type performance
measures. On the other hand, the introduction of attention mechanism weakens the correctness of
predicted direction. Moreover, the “Decomposition-and-Ensemble” framework of the forecasting
model inevitably causes greater data processing, which leads to a higher time cost while improving the
forecasting quality.

100

95
Dstat(%)

90

85 VMD-AttGRU for IXIC

VMD-GRU for IXIC
VMD-AttGRU for FTSE
VMD-GRU for FTSE
80
IMF1 IMF3 IMF5 IMF7 IMF9 IMF11 IMF13 IMF15

Figure 8. The Dstat values of IMFs predicted by AttGRU and GRU for hybrid model.

6. Conclusions

A hybrid model VMD-AttGRU is proposed in this study to forecast the stock price indices of
FTSE and IXIC. Since the price series is non-stationary and non-linear, the VMD approach is applied
to weaken the adverse effect of too much noise in prediction. Moreover, considering that not all
elements in the input series contribute equally to the forecasting tasks, the attention mechanism is
utilized to assign weights to different input elements for the GRU network and achieves a more

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7165

accurate forecasting result. Compared with single models (LSTM, GRU, and AttGRU) and a hybrid
model (VMD-GRU), the proposed VMD-AttGRU model exhibits superiority in improving forecasting
accuracy of stock price indices after analyzing its performance (MAE, RMSE, MAPE, and TIC)
together with trend-type performance (Dstat). The proposed VMD-AttGRU model can provide an
effective paradigm for the prediction of financial time series, which could also be applied to predicting
time series in other fields.

Acknowledgments

The work was partially supported by the Humanities and Social Sciences Foundation of Ministry
of Education of China (No. 18YJCZH134) and the Fundamental Research Funds for the Central
Universities (No. FRF-BR-18-001B).

Conflict of interests

The authors declare there is no conflict of interests.

References

1. C. Zhang, H. Pan, Y. Ma, X. Huang, Analysis of Asia Pacific stock markets with a novel
multiscale model, Phys. A, 534 (2019), 120939.
2. A. L. D. Loureiro, V. L. Miguéis, L. F. M. da Silva, Exploring the use of deep neural networks
for sales forecasting in fashion retail, Decis. Support Syst., 114 (2018), 81–93.
3. J. Wang, J. Wang, Forecasting stock market indexes using principle component analysis and
stochastic time effective neural networks, Neurocomputing, 156 (2015), 68–78.
4. Y. Xu, S. B. Cohen, Stock movement prediction from tweets and historical prices, In Proceedings
of the 56th Annual Meeting of the Association for Computational Linguistics, 2018.
5. D.P. Mandic, J.A. Chambers, Exploiting inherent relationships in RNN architectures, Neural
Networks, 12 (1999), 1341–1345.
6. T. Deng, X. He, Z. Zeng, Recurrent neural network for combined economic and emission dispatch,
Applied Intelligence, Appl. Intell., 48 (2018), 2180–2198.
7. S. Hochreiter, J. Schmidhuber, Long short-term memory, Neural Comput., 9 (1997), 1735–1780.
8. K. Wang, X. Qi, H. Liu, Photovoltaic power forecasting based LSTM-Convolutional Network,
Energy, 189 (2019), 116225.
9. Z. Karevan, J. A. K. Suykens, Transductive LSTM for time-series prediction: An application to
weather forecasting, Neural Networks, 125 (2020), 1–9.
10. B. Zhao, Z. P. Wang, W. J. Ji, X. Gao, X. B. Li, A Short-term Power Load Forecasting Method
Based on Attention Mechanism of CNN-GRU, Power Syst. Technol., 12 (2019).
11. Z. Y. Peng, S. Peng, L. D. Fu, B. C. Lu, J. J. Tang, K. Wang, et al., A novel deep learning ensemble
model with data denoising for short-term wind speed forecasting, Energy Convers. Manage., 207
(2020), 112524.
12. W. Y. Wu, W. L. Liao, J. Miao, G. L. Du, Using Gated Recurrent Unit Network to Forecast Short-
Term Load Considering Impact of Electricity Price, Energy Procedia, 158 (2019) 3369–3374.

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.

 7166

13. J. Zhang, D. Li, Y. Hao, Z. Tan, A hybrid model using signal processing technology, econometric
models and neural network for carbon spot price forecasting, J. Cleaner Prod., 204 (2018), 958–
964.
14. J. Wang, L. Y. Tang, Y. Y. Luo, P. Ge, A weighted EMD-based prediction model based on
TOPSIS and feed forward neural network for noised time series, Knowl. Based Syst., 132 (2017),
167–178.
15. J. Cao, Z. Li, J. Li, Financial time series forecasting model based on CEEMDAN and LSTM,
Phys. A, 519 (2019), 127–139.
16. K. Dragomiretskiy, D. Zosso, Variational mode decomposition, IEEE Trans. Signal Process., 62
(2014), 531–544.
17. S. Lahmiri, Intraday stock price forecasting based on variational mode decomposition, J. Comput.
Sci., 12 (2016), 23–27.
18. S. Lahmiri, A variational mode decomposition approach for analysis and forecasting of economic
and financial time series, Expert Syst. Appl., 55 (2016), 268–273.
19. S. Lahmiri, Comparing variational and empirical mode decomposition in forecasting day-ahead
energy prices, IEEE Syst. J., 11 (2015), 1907–1910.
20. Q. Zhu, F. Zhang, S. Liu, Y. Wu, L. Wang, A hybrid VMD–BiGRU model for rubber futures time
series forecasting, Appl. Soft Comput., 84 (2019), 105739.
21. C. Li, G. Tang, X. Xue, A. Saeed, X. Hu, Short-term wind speed interval prediction based on
ensemble GRU model, IEEE Trans. Sustainable Energy, 11 (2020), 1370–1380.
22. R. Wang, C. Li, W. Fu, G. Tang, Deep learning method based on gated recurrent unit and
variational mode decomposition for short-term wind power interval prediction, IEEE Trans.
Neural Networks Learn. Syst., 31 (2019), 3814–3827.
23. S. Boyd, N. Parikh, E. Chu, B. Peleato. J. Eckstein, Distributed Optimization and Statistical
Learning via the Alternating Direction Method of Multipliers, Now Foundations and Trends, 2011.
24. Y. Liu, C. Yang, K. Huang, W. Cui, Non-ferrous metals price forecasting based on variational
mode decomposition and LSTM network, Knowl. Based Syst., 188 (2020), 105006.
25. J. W. E, J. M. Ye, L. L. He, H. H. Jin, Energy price prediction based on independent component
analysis and gated recurrent unit neural network, Energy, 189 (2019), 116278.
26. S. Chen, L. Ge, Exploring the attention mechanism in LSTM-based Hong Kong stock price
movement prediction, Quant. Finance, 19 (2019), 1507–1515.
27. R. Desimon, J. Duncan, Neural mechanisms of selective visual attention, Annu. Rev. Neurosci.,
18 (1995), 193–222.
28. M. T. Luong, H. Pham, C. D. Manning, Effective approaches to attention-based neural machine
translation, arXiv:1508.04025.
29. L. Li, S. Tang, Y. Zhang, L. Deng, Q. Tian, GLA: global-local attention for image description,
IEEE Trans. Multimedia, 20 (2017), 726–737.
30. S. Wang, X. Wang, S. Wang, D. Wang, Bi-directional long short-term memory method based on
attention mechanism and rolling update for short-term load forecasting, Int. J. Electric. Power
Energy Syst., 109 (2019), 470–479.

©2020 the Author(s), licensee AIMS Press. This is an open access

article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/4.0)

Mathematical Biosciences and Engineering Volume 17, Issue 6, 7151–7166.