Abstract: Analysis of heart rate variability (HRV) has become a
popular non-invasive tool for assessing the activities of autonomic
nervous system. Most of the methods were hired from techniques
used for time series analysis. Currently used methods are time
domain, frequency domain, geometrical and fractal methods. A new
technique, which searches for pattern repeatability in a time series, is
proposed for quantifying heart rate (HR) time series. These set of
indices, which are termed as pattern repeatability measure and
pattern repeatability ratio are able to distinguish HR data clearly
from noise and electroencephalogram (EEG). The results of analysis
using these measures give an insight into the fundamental difference
between the composition of HR time series with respect to EEG and
noise.
Abstract: By systematically applying different engineering
methods, difficult financial problems become approachable. Using a
combination of theory and techniques such as wavelet transform,
time series data mining, Markov chain based discrete stochastic
optimization, and evolutionary algorithms, this work formulated a
strategy to characterize and forecast non-linear time series. It
attempted to extract typical features from the volatility data sets of
S&P100 and S&P500 indices that include abrupt drops, jumps and
other non-linearity. As a result, accuracy of forecasting has reached
an average of over 75% surpassing any other publicly available
results on the forecast of any financial index.
Abstract: The paper evaluates several hundred one-day-ahead
VaR forecasting models in the time period between the years 2004
and 2009 on data from six world stock indices - DJI, GSPC, IXIC,
FTSE, GDAXI and N225. The models model mean using the ARMA
processes with up to two lags and variance with one of GARCH,
EGARCH or TARCH processes with up to two lags. The models are
estimated on the data from the in-sample period and their forecasting
accuracy is evaluated on the out-of-sample data, which are more
volatile. The main aim of the paper is to test whether a model
estimated on data with lower volatility can be used in periods with
higher volatility. The evaluation is based on the conditional coverage
test and is performed on each stock index separately. The primary
result of the paper is that the volatility is best modelled using a
GARCH process and that an ARMA process pattern cannot be found
in analyzed time series.
Abstract: The purpose of this paper is to present a Dynamic
Time Warping technique which reduces significantly the data
processing time and memory size of multi-dimensional time series
sampled by the biometric smart pen device BiSP. The acquisition
device is a novel ballpoint pen equipped with a diversity of sensors
for monitoring the kinematics and dynamics of handwriting
movement. The DTW algorithm has been applied for time series
analysis of five different sensor channels providing pressure,
acceleration and tilt data of the pen generated during handwriting on
a paper pad. But the standard DTW has processing time and memory
space problems which limit its practical use for online handwriting
recognition. To face with this problem the DTW has been applied to
the sum of the five sensor signals after an adequate down-sampling
of the data. Preliminary results have shown that processing time and
memory size could significantly be reduced without deterioration of
performance in single character and word recognition. Further
excellent accuracy in recognition was achieved which is mainly due
to the reduced dynamic time warping RDTW technique and a novel
pen device BiSP.
Abstract: Microarrays technique allows the simultaneous measurements of the expression levels of thousands of mRNAs. By mining this data one can identify the dynamics of the gene expression time series. By recourse of principal component analysis, we uncover the circadian rhythmic patterns underlying the gene expression profiles from Cyanobacterium Synechocystis. We applied PCA to reduce the dimensionality of the data set. Examination of the components also provides insight into the underlying factors measured in the experiments. Our results suggest that all rhythmic content of data can be reduced to three main components.
Abstract: In this paper, we study the application of Extreme
Learning Machine (ELM) algorithm for single layered feedforward
neural networks to non-linear chaotic time series problems. In this
algorithm the input weights and the hidden layer bias are randomly
chosen. The ELM formulation leads to solving a system of linear
equations in terms of the unknown weights connecting the hidden
layer to the output layer. The solution of this general system of
linear equations will be obtained using Moore-Penrose generalized
pseudo inverse. For the study of the application of the method we
consider the time series generated by the Mackey Glass delay
differential equation with different time delays, Santa Fe A and
UCR heart beat rate ECG time series. For the choice of sigmoid,
sin and hardlim activation functions the optimal values for the
memory order and the number of hidden neurons which give the
best prediction performance in terms of root mean square error are
determined. It is observed that the results obtained are in close
agreement with the exact solution of the problems considered
which clearly shows that ELM is a very promising alternative
method for time series prediction.
Abstract: In the last 15 years, a number of methods have been proposed for forecasting based on fuzzy time series. Most of the fuzzy time series methods are presented for forecasting of enrollments at the University of Alabama. However, the forecasting accuracy rates of the existing methods are not good enough. In this paper, we compared our proposed new method of fuzzy time series forecasting with existing methods. Our method is based on frequency density based partitioning of the historical enrollment data. The proposed method belongs to the kth order and time-variant methods. The proposed method can get the best forecasting accuracy rate for forecasting enrollments than the existing methods.
Abstract: This paper develops an unscented grid-based filter
and a smoother for accurate nonlinear modeling and analysis
of time series. The filter uses unscented deterministic sampling
during both the time and measurement updating phases, to approximate
directly the distributions of the latent state variable. A
complementary grid smoother is also made to enable computing
of the likelihood. This helps us to formulate an expectation
maximisation algorithm for maximum likelihood estimation of
the state noise and the observation noise. Empirical investigations
show that the proposed unscented grid filter/smoother compares
favourably to other similar filters on nonlinear estimation tasks.
Abstract: The authors have been developing several models
based on artificial neural networks, linear regression models, Box-
Jenkins methodology and ARIMA models to predict the time series
of tourism. The time series consist in the “Monthly Number of Guest
Nights in the Hotels" of one region. Several comparisons between the
different type models have been experimented as well as the features
used at the entrance of the models. The Artificial Neural Network
(ANN) models have always had their performance at the top of the
best models. Usually the feed-forward architecture was used due to
their huge application and results. In this paper the author made a
comparison between different architectures of the ANNs using
simply the same input. Therefore, the traditional feed-forward
architecture, the cascade forwards, a recurrent Elman architecture and
a radial based architecture were discussed and compared based on the
task of predicting the mentioned time series.
Abstract: This paper presents a procedure for estimating VAR
using Sequential Discounting VAR (SDVAR) algorithm for online
model learning to detect fraudulent acts using the telecommunications
call detailed records (CDR). The volatility of the VAR is observed
allowing for non-linearity, outliers and change points based on the
works of [1]. This paper extends their procedure from univariate
to multivariate time series. A simulation and a case study for
detecting telecommunications fraud using CDR illustrate the use of
the algorithm in the bivariate setting.
Abstract: A concern that researchers usually face in different
applications of Artificial Neural Network (ANN) is determination of
the size of effective domain in time series. In this paper, trial and
error method was used on groundwater depth time series to determine
the size of effective domain in the series in an observation well in
Union County, New Jersey, U.S. different domains of 20, 40, 60, 80,
100, and 120 preceding day were examined and the 80 days was
considered as effective length of the domain. Data sets in different
domains were fed to a Feed Forward Back Propagation ANN with
one hidden layer and the groundwater depths were forecasted. Root
Mean Square Error (RMSE) and the correlation factor (R2) of
estimated and observed groundwater depths for all domains were
determined. In general, groundwater depth forecast improved, as
evidenced by lower RMSEs and higher R2s, when the domain length
increased from 20 to 120. However, 80 days was selected as the
effective domain because the improvement was less than 1% beyond
that. Forecasted ground water depths utilizing measured daily data
(set #1) and data averaged over the effective domain (set #2) were
compared. It was postulated that more accurate nature of measured
daily data was the reason for a better forecast with lower RMSE
(0.1027 m compared to 0.255 m) in set #1. However, the size of input
data in this set was 80 times the size of input data in set #2; a factor
that may increase the computational effort unpredictably. It was
concluded that 80 daily data may be successfully utilized to lower the
size of input data sets considerably, while maintaining the effective
information in the data set.
Abstract: Recognizing behavioral patterns of financial markets
is essential for traders. Japanese candlestick chart is a common tool to
visualize and analyze such patterns in an economic time series. Since
the world was introduced to Japanese candlestick charting, traders
saw how combining this tool with intelligent technical approaches
creates a powerful formula for the savvy investors.
This paper propose a generalization to box counting method of
Grassberger-Procaccia, which is based on computing the correlation
dimension of Japanese candlesticks instead commonly used 'close'
points. The results of this method applied on several foreign
exchange rates vs. IRR (Iranian Rial). Satisfactorily show lower
chaotic dimension of Japanese candlesticks series than regular
Grassberger-Procaccia method applied merely on close points of
these same candles. This means there is some valuable information
inside candlesticks.
Abstract: All the geophysical phenomena including river
networks and flow time series are fractal events inherently and fractal
patterns can be investigated through their behaviors. A non-linear
system like a river basin can well be analyzed by a non-linear
measure such as the fractal analysis. A bilateral study is held on the
fractal properties of the river network and the river flow time series.
A moving window technique is utilized to scan the fractal properties
of them. Results depict both events follow the same strategy
regarding to the fractal properties. Both the river network and the
time series fractal dimension tend to saturate in a distinct value.
Abstract: Time series forecasting is an important and widely
popular topic in the research of system modeling. This paper
describes how to use the hybrid PSO-RLSE neuro-fuzzy learning
approach to the problem of time series forecasting. The PSO
algorithm is used to update the premise parameters of the
proposed prediction system, and the RLSE is used to update the
consequence parameters. Thanks to the hybrid learning (HL)
approach for the neuro-fuzzy system, the prediction performance
is excellent and the speed of learning convergence is much faster
than other compared approaches. In the experiments, we use the
well-known Mackey-Glass chaos time series. According to the
experimental results, the prediction performance and accuracy in
time series forecasting by the proposed approach is much better
than other compared approaches, as shown in Table IV. Excellent
prediction performance by the proposed approach has been
observed.
Abstract: The neural network's performance can be measured by efficiency and accuracy. The major disadvantages of neural network approach are that the generalization capability of neural networks is often significantly low, and it may take a very long time to tune the weights in the net to generate an accurate model for a highly complex and nonlinear systems. This paper presents a novel Neuro-fuzzy architecture based on Extended Kalman filter. To test the performance and applicability of the proposed neuro-fuzzy model, simulation study of nonlinear complex dynamic system is carried out. The proposed method can be applied to an on-line incremental adaptive learning for the prediction of financial time series. A benchmark case studie is used to demonstrate that the proposed model is a superior neuro-fuzzy modeling technique.
Abstract: Data of wave height and wind speed were collected
from three existing oil fields in South China Sea – offshore
Peninsular Malaysia, Sarawak and Sabah regions. Extreme values
and other significant data were employed for analysis. The data were
recorded from 1999 until 2008. The results show that offshore
structures are susceptible to unacceptable motions initiated by wind
and waves with worst structural impacts caused by extreme wave
heights. To protect offshore structures from damage, there is a need
to quantify descriptive statistics and determine spectra envelope of
wind speed and wave height, and to ascertain the frequency content
of each spectrum for offshore structures in the South China Sea
shallow waters using measured time series. The results indicate that
the process is nonstationary; it is converted to stationary process by
first differencing the time series. For descriptive statistical analysis,
both wind speed and wave height have significant influence on the
offshore structure during the northeast monsoon with high mean wind
speed of 13.5195 knots ( = 6.3566 knots) and the high mean wave
height of 2.3597 m ( = 0.8690 m). Through observation of the
spectra, there is no clear dominant peak and the peaks fluctuate
randomly. Each wind speed spectrum and wave height spectrum has
its individual identifiable pattern. The wind speed spectrum tends to
grow gradually at the lower frequency range and increasing till it
doubles at the higher frequency range with the mean peak frequency
range of 0.4104 Hz to 0.4721 Hz, while the wave height tends to
grow drastically at the low frequency range, which then fluctuates
and decreases slightly at the high frequency range with the mean
peak frequency range of 0.2911 Hz to 0.3425 Hz.
Abstract: Quantitative characterization of nonlinear directional
couplings between stochastic oscillators from data is considered. We
suggest coupling characteristics readily interpreted from a physical
viewpoint and their estimators. An expression for a statistical
significance level is derived analytically that allows reliable coupling
detection from a relatively short time series. Performance of the
technique is demonstrated in numerical experiments.
Abstract: Daily production of information and importance of the sequence of produced data in forecasting future performance of market causes analysis of data behavior to become a problem of analyzing time series. But time series that are very complicated, usually are random and as a result their changes considered being unpredictable. While these series might be products of a deterministic dynamical and nonlinear process (chaotic) and as a result be predictable. Point of Chaotic theory view, complicated systems have only chaotically face and as a result they seem to be unregulated and random, but it is possible that they abide by a specified math formula. In this article, with regard to test of strange attractor and biggest Lyapunov exponent probability of chaos on several foreign exchange rates vs. IRR (Iranian Rial) has been investigated. Results show that data in this market have complex chaotic behavior with big degree of freedom.
Abstract: Weather systems use enormously complex
combinations of numerical tools for study and forecasting.
Unfortunately, due to phenomena in the world climate, such
as the greenhouse effect, classical models may become
insufficient mostly because they lack adaptation. Therefore,
the weather forecast problem is matched for heuristic
approaches, such as Evolutionary Algorithms.
Experimentation with heuristic methods like Particle Swarm
Optimization (PSO) algorithm can lead to the development of
new insights or promising models that can be fine tuned with
more focused techniques. This paper describes a PSO
approach for analysis and prediction of data and provides
experimental results of the aforementioned method on realworld
meteorological time series.
Abstract: It is well known that during the developments in the
economic sector and through the financial crises occur everywhere in
the whole world, volatility measurement is the most important
concept in financial time series. Therefore in this paper we discuss
the volatility for Amman stocks market (Jordan) for certain period of
time. Since wavelet transform is one of the most famous filtering
methods and grows up very quickly in the last decade, we compare
this method with the traditional technique, Fast Fourier transform to
decide the best method for analyzing the volatility. The comparison
will be done on some of the statistical properties by using Matlab
program.