International Journal of Intelligence Science, 2012, 2, 176-180
http://dx.doi.org/10.4236/ijis.2012.224023 Published Online October 2012 (http://www.SciRP.org/journal/ijis)
Using Data Mining with Time Series Data in
Short-Term Stocks Prediction: A Literature
José Manuel Azevedo1, Rui Almeida2, Pedro Almeida3
1Department of Mathematics, Instituto Politécnico do Porto, Porto, Portugal
2Department of Mathematics, Faculdade de Ciências, Universidade da Beira Interior, Covilhã, Portugal
3Department of Informatics, Faculdade de Engenharia, Universidade da Beira Interior, Covilhã, Portugal
Email: firstname.lastname@example.org, email@example.com, firstname.lastname@example.org
Received July 1, 2012; revised August 20, 2012; accepted September 1, 2012
Data Mining (DM) methods are being increasingly used in prediction with time series data, in addition to traditional
statistical approaches. This pape r presents a literature review of the use of DM with time series data, focusing on short-
time stocks predictio n. This is an area that has been attracting a great deal of attention from researchers in the field. The
main contribution of this paper is to provide an outlin e of the use of DM with time series data, using mainly examples
related with short-term stocks prediction. This is important to a better understanding of the field. Some of the main
trends and open issues will also be introduced.
Keywords: Data Mining; Time Series; Fundamental Data; Data Frequency; Application Domain; Short-Term Stocks
Data Mining (DM) is a challenging field for research and
has some practical successful application in several dif-
ferent areas. DM methods are being increasingly used in
prediction with time series data, in addition to traditional
statistical approaches [1-3].
DM can be presented as one of the phases of the
Knowledge Discov ery in Databases (KDD ) process [4-6],
and is identified as “the means by which the patterns are
extracted from data” . Nowadays, it can be said that
the two terms, DM and KDD, are indistinctly used.
The OECD Glossary of statistical terms  presents
the following definition: “A ti me series is a set of reg ular
time-ordered observations of a quantitative characteristic
of an individual or collective phenomenon taken at suc-
cessive, in most cases equidistant, periods/points of time”.
There are several application domains of DM with time
series data, being that one important application domain
is short-term stocks prediction. This will be the focus of
this paper. Short-term stocks prediction is a difficult is-
sue and can be considered as an open research issue
[9,10]. Intelligent forecasting models have achieved bet-
ter results than traditional methods, particularly in short-
term forecasts . Although intelligent forecasting me-
thods are better, we can still improve the results in terms
of accuracy in addition to other factors.
The main contribution of this paper is to provide an
outline of the use of DM with time series data, using
mainly examples related with short-term stocks or market
indexes predictions. This is important to a better under-
standing of the field. Some of the main trends and open
issues will also be introduced.
The paper is organized as follows: DM with time se-
ries data is presented in Section 2, the integration of fun-
damental data is explored in Section 3, data frequency
issues are introduced in Section 4. The paper closes in
Section 5, with conc lusion an d f uture research di re ctions.
2. Data Mining with Time Series Data
Since the seminal paper of Fayyad in 1996 , the Data
Mining (DM) area has attracted a great deal of interest
and can nowadays be considered as an established field.
DM applications can be found in a diversified range of
application domains. One important application domain
is that of time series data. “A time-series data set consists
of sequences of numeric values obtained over repeated
measurements of time. The values are typically measured
at equal time intervals (e.g ., every minute, hour, or day)”.
. The referred measures can be taken over one variable
or several variables—univariate or multivariate tim e series.
opyright © 2012 SciRes. IJIS
J. M. AZEVEDO ET AL. 177
2.1. Data Mining with Time Series Data
DM with time series data is popular and many applica-
tions can be found in the literature, for instance, for
earthquake forecasting , characterization of ozone
behavior , or flood prediction . Other application
example is that of financial decision making. A decision
support tool for financial forecasting, named as EDDIE,
is presented in . In , a new architecture that im-
plements a binary neural network, AURA, to produce
discrete probability distribution as forecasts, using high
frequency data sets, is presented. The use of support
vector machines and back propagation neural networks to
predict credit ratings is presented in .
One important application concerns short-term stocks
prediction, which is the main focus of this pap er. In ,
an approach to the paradox of obtaining better results
with long-horizon forecasts than with short-horizon fore-
casts is presented, and it is claimed that the paradox is
solved, since the proposed model obtains promising re-
sults. Nevertheless, there is a great deal of interest from
investors in short-horizon forecasts, thus the authors con-
sider that research focusing on this issue is important,
namely in using data mining with time series for short-
term stocks prediction.
2.2. Data Mining Techniques Used with Time
Series Data for Short-Term Stocks
Several DM techniques are used with time series data in
order to obtain short-term stocks prediction. An interest-
ing approach to portfolio management, using the Gaus-
sian temporal factor analysis technique, is introduced in
. Neural networks are one of the most popular tech-
niques for stocks prediction. [20-25] are some examples.
In  rough sets and classification trees are used, as
well. Rough sets are also used in . Support Vector
Machines are used in .
There were not yet been given strong evidences of
some technique being better than other, but nonlinear
models are more popular.
2.3. Specific Challenges
Using DM with time series data presents several specific
challenges. In [28,29] the authors focus on the issue of
representing time series data in order to effectively and
efficiently apply DM. In , three types of algorithms
are presented and compared, namely, the sliding window
algorithm, the top-down algorithm, and the bottom-up
algorithm, and a new approach, that is claimed to over-
come the inconveniences of these three algorithms, is
introduced. In , a new concept, named as median
strings, is presented as a simple and, at the same time,
powerful representation for time series data.
Another interesting issue is to find out if different time
series, or parts of a time series, have similar behavior.
This issue can be approached through the use of simila-
rity measures and indexing techniques. Interesting re-
views can be found in [30,31].
Over fitting is a common problem across DM applica-
tions and DM with time series data is not an exception.
In , an approach that intends to overcome this pro-
blem is presented.
Other important issue concerns the way to implement
each one of the phases of the KDD process, taking into
account the specificities of time series data. An applica-
tion of DM with time series data for short-term stock
prediction is presented in , analyzing all the phases of
the KDD process. Promising results were achieved, but it
is referred that the inclusion of fundamental data could
help improving the obtained results.
Table 1 presents a resume of the main techniques and
3. Including Fundamental Data
Concerning short-term stocks prediction, a possible ap-
proach is to collect the historical financial data, such as
open price, higher price, lower price, close price, and
volume. These can be used in a daily basis frequency, or
other frequencies considered as appropriate. Several in-
dicators can be derived and used for more adequate
analysis. This approach is named as technical analysis.
Another possible approach is to use statistical data, such
as, macroeconomics indexes, and basic financial indica-
tors of the company. This approach is named as funda-
mental analysis. Table 2 resumes some of the technical
and fundamental features found in the literature. Other
researches, for instance [37-39], present similar indica-
From the literature review it is clear that one of the
main issues in obtaining good pr edictio n s is related to the
first phase of the KDD process, that is to say, the selec-
Table 1. Data mining with time series data: Main techniques
Neural networks [20-25]
Vector machines support 
Rough sets [22,26]
Classification trees 
Gaussian temporal factor analysis 
Data representation [28,29]
Similar behavior [30,31]
Over fitting 
Implementing all KDD phases 
Copyright © 2012 SciRes. IJIS
J. M. AZEVEDO ET AL.
Copyright © 2012 SciRes. IJIS
Table 2. Features for techni cal and fundamental analysis.
Type Features References
ROA(A); EBI Gross margin; g ross margin growth o perating income; operation income growth; net
income; net income growth; continued net income; cash flow ratio; sales growth ratio; current
ratio; ordinary income gro wth; continued income growth; total asset growth; return on total asset;
quick ratio; liabilities ratio; total asset turnover; account receivable turnover; inventory turnover;
fixed asset turnover; days payables outstanding;
And several of others: gross national product; real GDP; unemployment rate; real economic growth;
monetary supply and amou nt ; g ross margin growth ; CCI; personal income; industrial production;
Taiwan export/import volume; operation income growth liabilities; total asset growth fixed asset
turnover; monitoring indicator Export for eign exchange volume; WPI; merchandise trade volume
Tsai and Hsiao (2010) 
Fundamental Demand index; moving average divergence convergence; relative strength index; positive
directional movement index; negative directional mov ement index; moving average; r-squared;
linear regression slope; average true range
Technical Price channel (top); price channel (bottom); price per earning per share; volume; open price; range;
changes; close price
Zarandi, Rezaee, Turksen
and Neshat (2009) 
Average position change; bollinger band %; cutler’s relative Strength index; exponential moving
average; stochastic oscillator; typical price; volume accumulator; volume weighted RSI-MFI;
volume weighted RSI, williams %R; advance decline line; average true range; average position
change; chaikin A/D oscillator; on balance volume; stoch. osc.; typical price
Ince and Trafalis (2007) 
Fundamental Money supply (M1B); governme nt c on sumption level, gross national products, gross domestic
products; consumer price index; whole-sale products index; rate of exchange
Technical Moving average convergence/divergence; price rate of change; stochastic %K; stochastic %D;
relative strength index; stochastic oscillator and directional indicator
Cheng, Chen and Lin
Technical On balance volume; moving average; average stock yield Shen, Guo, Wu and Wu
tion of the adequate feature combination, since the same
methods can yield different results if different features
are selected as inputs. Weekly
Another aspect that arises from the literature review is
that most researchers use only one of the two types of
analysis, technical or fundamental. Thus analyzing com-
binations of both types of i ndicators is yet under-explored.
Figure 1. Time series with different frequencies.
In addition, most studies use macroeconomics vari-
ables, forgetting the important financial indicators of the
companies. Considering the domain application, it is
clear that the evolution of stock prices is influenced by
both types of variables, so considering it could conduct
to good resu lts.
low frequency time series obtained from the collection of
fundamental data. Forecasts should be done in a daily
basis, thus there are some important issues for research.
Some research can be found in the literature approa-
ching the issue of integrating time series features with
different frequencies. Traditional approaches use regres-
sion algorithms such as MIDAS [37,38]. Nevertheless,
this approach does not use DM.
One of the main issues related to the combination of
both types of features is that time series data have dif-
ferent frequencies (Figure 1). Usually technical features
have daily frequencies and fundamental features have
monthly, quarterly, and lower frequencies, presenting
some integration issues. These in tegration issues are very
important and have several implications.
In the literature review, only a few works, use DM
with time series data with different frequencies. [22,34]
are two examples. These studies present promising re-
sults, but the use of neural networks is somehow a limi-
tation. Neural networks, despite usually yielding good
results, functions as a “black box” . This way it is difficult
to understand the mechanism and the generated model.
4. Integrating Features with Different
From the literature review it can be concluded that
these issues needs further research, and it can be useful to
test other methods, and to explore the selection of some
As stated above, interesting results could be obtained
through the integration of time series data with different
frequencies. With short-term stocks predictions, there is
the need to use mainly time series with data collected
daily, yielding high frequency time series, opposed to The application domain is an important issue to con-
J. M. AZEVEDO ET AL. 179
sider when applying DM, thus it should also be consider-
ed in this case. Taking into account the application do-
main will surely bring good in sights and will surely yield
5. Conclusions and Future Research
This paper presents a literature review of the use of data
mining with time series data. This literature review is
very useful, since it brings a better understanding of the
field of study, and this is an important contribution of
From the literature review it can be con cluded that this
subject attracts a great deal of interest by researchers.
Nevertheless, several research issues remain unexplored.
One of the ones that were identified during this research
is related with the combined use of fundamental and
technical indicators. The combined use of both types of
indicators reveals also the issue of integrating time series
with different fr equencie s.
Feature selection, corresponding to the first phase of
the KDD process, is also an issue that requires more re-
search to be done.
Future research directions include the study of ways to
select the best features for DM with time series data. The
existence of features with different frequencies is a concern,
and methods that will help how to envisage this problem
will be planned and implemented.
This work was partially supported by the research pro-
jects: PEst-OE/MAT/UI0212/2011, financed by FEDER
through COMPETE—Programa Operacional Factores de
Competitividade and by FCT—Fundação para a Ciência
e a Tecnologia.
 P. Almeida, “Previsão do Comportamento de Séries
Temporais Financeiras com Apoio de Conhecimento
Sobre o Domínio,” Ph.D. Thesis, Universidade da Beira
Interior, Covilhã, 2003.
 L. Breiman, “Statistical Modeling: The Two Cultures,”
Statistical Science, Vol. 18, No. 3, 2001, pp. 199-231.
 M. A. Ruggiero, “Ciberne tic Tradin g Strategies—Developing
a Profitable Trading System State-of-the-Art Techno-
logies,” John Wiley & Sons, New York, 1977.
 U. M. Fayyad, G. Piatetski-Shapiro and P. Smyth, “From
Data Mining to Knowledge Discovery: An Overview,” In:
U. M. Fayyad, G. Piatetsky-Shapiro, P. Smyth and R.
Uthurusamy, Eds., Advances in Knowledge Discovery
and Data Mining, The MIT Press, Cambridge, 1996, pp.
 J. Han, M. Kamber and J. Pei, “Data Mining: Concepts
and Techniques,” Morgan Kaufman Publishers, Califor-
 D. Hand, H. Mannila and P. Smyth, “Principles of Data
Mining,” The MIT Press, Cambridge, 2011.
 A. Azevedo and M. F. Santos, “KDD, SEMMA, and
CRISP-DM: A Parallel Overview,” Proceedings of the
IADIS European Conference on Data Mining, Amster-
dam, 24-26 July 2008, pp. 182-185.
 OECD, “Time Series,” 2006.
 M. A. Ferreira and P. Santa-Clara, “Forecasting Stock
Market Returns: The Sum of the Parts Is More than the
Whole,” Journal of Financial Economics, Vol. 100, No. 3,
2011, pp. 514-537. doi:10.1016/j.jfineco.2011.02.003
 T. Fu, “A Review on Time Series Data Mining,” En-
gineering Applications of Artificial Intelligence, Vol. 21,
No. 1, 2011, pp. 164-181.
 T. O. Hill, M. Connor and W. Remus, “Neural Network
Models for Time Series Forecasts,” Management Science,
Vol. 42, No. 7, 1996, pp. 1082-1092.
 S. Fong and Z. Nannan, “Towards an Adaptive Fore-
casting of Earthquake Time Series from Decomposable
and Salient Characteritics,” Proceendings of the 3rd In-
ternational Conference on Pervasive Patterns and Appli-
cations, Rome, 25 September 2011, pp. 53-60.
 K. J. Walsh, M. Milligan, M. Woodman and J. Sherwell,
“Data Mining to Characterize Ozone Behavior in Bal-
timore and Washington DC,” Journal of Atmospheric En-
vironment, Vol. 42, No. 18, 2008, pp. 4280-4292.
 C. Damle and A. Yalcin, “Flood Prediction Using Time
Series Data Mining,” Journal of Hidrology, Vol. 333, No.
2-4, 2007, pp. 305-316.
 E. Tsang, P. Yung and J. Li, “EDDIE-Automation, a De-
cision Support Tool for Financial Forecasting,” Decision
Support Systems, Vol. 37, No. 4, 2004, pp. 559-565.
 A. Pasley and J. Austin, “Distribution Forecasting of
High Frequency Time Series,” Decision Support Systems,
Vol. 37, No. 4, 2004, pp. 501-513.
 Z. Huang, H. Chen, C. J. Hsu, W. H. Chen and S. Wu,
“Credit Ratings Analysis with Support Vector Machines
and Neural Networks. A Market Comparative Study,”
Decision Support Systems, Vol. 37, No. 4, 2004, pp.
 H. M. Krolzig and J. Toro, “Multiperiod Forecasting in
Stock Market: A Paradox Solved,” Decision Support Sys-
tems, Vol. 37, No. 4, 2004, pp. 531-542.
 K. C. Chiu and L. Xu, “Arbitrage Pricing Theory-Based
Gaussian Temporal Factor Analysis for Adaptive Port-
folio Management,” Decision Support Systems, Vol. 37,
No. 4, 2004, pp. 485-500.
Copyright © 2012 SciRes. IJIS
J. M. AZEVEDO ET AL.
Copyright © 2012 SciRes. IJIS
 O. Coupelon, “Nneural Network Modeling for Stock
Movement Prediction: A State of the Art,” 2007.
 M. Kordos and A. Cwiok, “A New Approach to Neural
Network Based Stock Trading Strategy,” Proceedings of
the 12th International Conference on Intelligent Data
Engineering and Automated Learning, Norwich, 7-9 Sep-
tember 2011, pp. 429-436.
 J. H. Cheng, H. P. Chen and Y. M. Lin, “A Hybrid Fore-
cast Marketing Timing Model Based on Probabilistic
Neural Network, Rough Set and C 4.5,” Expert Systems
with Applications, Vol. 37, No. 4, 2010, pp. 1814-1820.
 Z. Yudong and W. Lenan, “Stock Market Prediction of S
& P 500 via Combination of Improved BCO Approach
and BP Neural Network,” Expert Systems with Appli-
cations, Vol. 36, No. 5, 2009, pp. 8849-8854.
 X. Lin, Z. Yang and Y. Song, “Short-Term Stock Price
Based on Echo State Networks,” Expert Systems with Ap-
plications, Vol. 36, No. 3, 2009, pp. 7313-7317.
 T. Chang, “A Comparative Study of Artificial Neural
Networks, and Decision Trees for Digital Game Content
Stocks Price Prediction,” Expert Systems with Appli-
cations, Vol. 38, No. 12, 2011, pp. 14846-14851.
 L. Shen and H. T. Loh, “Applying Rough Set to Market
Timing Decisions,” Decision Support System, Vol. 37, No.
4, 2004, pp. 583-597.
 Q. Wen, Z. Yang, Y. Song and P. Jia, “Automatic Stock
Decision Support System Based on Box Theory and SVM
Algorithm,” Expert Systems with Applications, Vol. 37,
No. 2, 2010, pp. 1015-1022.
 E. Keogh, S. Chu, D. Hart and M. Pazzani, “Segmenting
Time Series: A Survey and No vel Approach,” In: M. Last,
A. Kandel and H. Bunke, Eds., Data Mining in Time Se-
ries Databases—Series in Machine Perception Artificial
Intelligence, World Scientific, Singapore, 2004, pp. 1-21.
 X. Jiang, H. Bunke and J. Csirik, “Median Strings: A
Review,” In: M. Last, A. Kandel and H. Bunke, Eds.,
Data Mining in Time Series Databases—Series in Ma-
chine Perception Artificial Intelligence, World Scientific,
Singapore, 2004, pp. 173-192.
 G. Das and D. Gunopulos, “Time Series Similarity and
Indexing,” In: N. Ye, Ed., The Handbook of Data Mining,
Lawrence Erlbaum Associates, London, 2003, pp. 279-
 M. L. Hetland, “A Survey of Recent Methods for Effi-
cient Retrieval of Similar Time Sequences,” In: M. Last,
A. Kandel and H. Bunke, Eds., Data Mining in Time Se-
ries Databases—Series in Machine Perception Artificial
Intelligence, World Scientific, Singapore, 2004, pp. 23-
 K. Mehta and S. Bhattacharya, “Adequacy of Training
Data for Evolutionary Mining of Trading Rules,” Deci-
sion Support Systems, Vol. 37, No. 4, 2004, pp. 461-474.
 C. F. Tsai and Y. C. Hsiao, “Combining Multiple Feature
Selection Methods for Stock Prediction: Union, Inter-
section, and Multi-Intersection Approaches,” Decision
Support Systems, Vol. 50, No. 1, 2010, pp. 258-269.
 M. H. F. Zarandi, B. Rezaee, I. B. Turksen and E. Neshat,
“A Type-2 Fuzzy Rule-Based Expert System Model for
Stock Price Analysis,” Decision Support Systems, Vol. 36,
No. 1, 2009, pp. 139-154.
 H. Ince and T. Trafalis, “Kernel Principal Component
Analysis and Support Vector Machines for Stock Price
Prediction,” IIE Transactions, Vol. 39, No. 6, 2007, pp.
 W. Shen, X. Guo, C. Wu and D. Wu, “Forecasting Stock
Indices Using Radial Basis Function Neural Networks
Optimized by Artificial Fish Swarm Algorithm,” Knowl-
edge-Based Systems, Vol. 24, No. 3, 2011, pp. 378-385.
 B. C. O. Tas, “Private Information of the Fed and Pre-
dictability of Stock Returns,” Applied Economics, Vol. 43,
No. 19, 2011, pp. 2381-2398.
 P. M. Dechow, A. P. Hutton, L. Meulbroek and R. G.
Sloan, “Short-Sellers, Fundamental Analysis, and Stock
Returns,” Journal of Financial Economics, Vol. 61, No. 1,
2001, pp. 77-106. doi:10.1016/S0304-405X(01)00056-3
 M. Lam, “Neural Network Techniques for Financial Per-
formance Prediction: Integrating Fundamental and Tech-
nical Analysis,” Decision Support Systems, Vol. 37, No. 4,
2004, pp. 567-581. doi:10.1016/S0167-9236(03)00088-5
 K. Wohlrabe, “Forecasting with Mixed-Frequency Time
Series Models,” Ph.D. Thesis, Ludwig Maximilians Uni-
versitat, Munchen, 2008.
 E. Andreou, E. Ghysels and A. Kourtellos, “Forecasting
with Mixed-Frequency Data,” Journal of Econometrics,
Vol. 158, No. 2, 2010, pp. 246-261.