Abstract: In this paper real money demand function is analyzed
within multivariate time-series framework. Cointegration approach is
used (Johansen procedure) assuming interdependence between
money demand determinants, which are nonstationary variables. This
will help us to understand the behavior of money demand in Croatia,
revealing the significant influence between endogenous variables in
vector autoregrression system (VAR), i.e. vector error correction
model (VECM). Exogeneity of the explanatory variables is tested.
Long-run money demand function is estimated indicating slow speed
of adjustment of removing the disequilibrium. Empirical results
provide the evidence that real industrial production and exchange
rate explains the most variations of money demand in the long-run,
while interest rate is significant only in short-run.
Abstract: This paper proposes a novel architecture for developing decision support systems. Unlike conventional decision support systems, the proposed architecture endeavors to reveal the decision-making process such that humans' subjectivity can be incorporated into a computerized system and, at the same time, to preserve the capability of the computerized system in processing information objectively. A number of techniques used in developing the decision support system are elaborated to make the decisionmarking process transparent. These include procedures for high dimensional data visualization, pattern classification, prediction, and evolutionary computational search. An artificial data set is first employed to compare the proposed approach with other methods. A simulated handwritten data set and a real data set on liver disease diagnosis are then employed to evaluate the efficacy of the proposed approach. The results are analyzed and discussed. The potentials of the proposed architecture as a useful decision support system are demonstrated.
Abstract: The fault detection and diagnosis of complicated
production processes is one of essential tasks needed to run the process
safely with good final product quality. Unexpected events occurred in
the process may have a serious impact on the process. In this work,
triangular representation of process measurement data obtained in an
on-line basis is evaluated using simulation process. The effect of using
linear and nonlinear reduced spaces is also tested. Their diagnosis
performance was demonstrated using multivariate fault data. It has
shown that the nonlinear technique based diagnosis method produced
more reliable results and outperforms linear method. The use of
appropriate reduced space yielded better diagnosis performance. The
presented diagnosis framework is different from existing ones in that it
attempts to extract the fault pattern in the reduced space, not in the
original process variable space. The use of reduced model space helps
to mitigate the sensitivity of the fault pattern to noise.
Abstract: In recent years, most of the regions in the world are
exposed to degradation and erosion caused by increasing
population and over use of land resources. The understanding of
the most important factors on soil erosion and sediment yield are
the main keys for decision making and planning. In this study, the
sediment yield and soil erosion were estimated and the priority of
different soil erosion factors used in the MPSIAC method of soil
erosion estimation is evaluated in AliAbad watershed in southwest
of Isfahan Province, Iran. Different information layers of the
parameters were created using a GIS technique. Then, a
multivariate procedure was applied to estimate sediment yield and
to find the most important factors of soil erosion in the model. The
results showed that land use, geology, land and soil cover are the
most important factors describing the soil erosion estimated by
MPSIAC model.
Abstract: Geographical Information Systems are an integral part
of planning in modern technical systems. Nowadays referred to as
Spatial Decision Support Systems, as they allow synergy database
management systems and models within a single user interface
machine and they are important tools in spatial design for
evaluating policies and programs at all levels of administration.
This work refers to the creation of a Geographical Information
System in the context of a broader research in the area of influence
of an under construction station of the new metro in the Greek
city of Thessaloniki, which included statistical and multivariate
data analysis and diagrammatic representation, mapping and
interpretation of the results.
Abstract: This paper deals with the application of Principal Component Analysis (PCA) and the Hotelling-s T2 Chart, using data collected from a drinking water treatment process. PCA is applied primarily for the dimensional reduction of the collected data. The Hotelling-s T2 control chart was used for the fault detection of the process. The data was taken from a United Utilities Multistage Water Treatment Works downloaded from an Integrated Program Management (IPM) dashboard system. The analysis of the results show that Multivariate Statistical Process Control (MSPC) techniques such as PCA, and control charts such as Hotelling-s T2, can be effectively applied for the early fault detection of continuous multivariable processes such as Drinking Water Treatment. The software package SIMCA-P was used to develop the MSPC models and Hotelling-s T2 Chart from the collected data.
Abstract: Sensitive and predictive DILI (Drug Induced Liver
Injury) biomarkers are needed in drug R&D to improve early
detection of hepatotoxicity. The discovery of DILI biomarkers that
demonstrate the predictive power to identify individuals at risk to
DILI would represent a major advance in the development of
personalized healthcare approaches. In this healthy volunteer
acetaminophen study (4g/day for 7 days, with 3 monitored nontreatment
days before and 4 after), 450 serum samples from 32
subjects were analyzed using protein profiling by antibody
suspension bead arrays. Multiparallel protein profiles were generated
using a DILI target protein array with 300 antibodies, where the
antibodies were selected based on previous literature findings of
putative DILI biomarkers and a screening process using pre dose
samples from the same cohort. Of the 32 subjects, 16 were found to
develop an elevated ALT value (2Xbaseline, responders). Using the
plasma profiling approach together with multivariate statistical
analysis some novel findings linked to lipid metabolism were found
and more important, endogenous protein profiles in baseline samples
(prior to treatment) with predictive power for ALT elevations were
identified.
Abstract: For the electrical metrics that describe photovoltaic
cell performance are inherently multivariate in nature, use of a
univariate, or one variable, statistical process control chart can have
important limitations. Development of a comprehensive process
control strategy is known to be significantly beneficial to reducing
process variability that ultimately drives up the manufacturing cost
photovoltaic cells. The multivariate moving average or MMA chart,
is applied to the electrical metrics of photovoltaic cells to illustrate
the improved sensitivity on process variability this method of control
charting offers. The result show the ability of the MMA chart to
expand to as any variables as needed, suggests an application
with multiple photovoltaic electrical metrics being used in
concert to determine the processes state of control.
Abstract: Through 1980s, management accounting researchers
described the increasing irrelevance of traditional control and
performance measurement systems. The Balanced Scorecard (BSC)
is a critical business tool for a lot of organizations. It is a
performance measurement system which translates mission and
strategy into objectives. Strategy map approach is a development
variant of BSC in which some necessary causal relations must be
established. To recognize these relations, experts usually use
experience. It is also possible to utilize regression for the same
purpose. Structural Equation Modeling (SEM), which is one of the
most powerful methods of multivariate data analysis, obtains more
appropriate results than traditional methods such as regression. In the
present paper, we propose SEM for the first time to identify the
relations between objectives in the strategy map, and a test to
measure the importance of relations. In SEM, factor analysis and test
of hypotheses are done in the same analysis. SEM is known to be
better than other techniques at supporting analysis and reporting. Our
approach provides a framework which permits the experts to design
the strategy map by applying a comprehensive and scientific method
together with their experience. Therefore this scheme is a more
reliable method in comparison with the previously established
methods.
Abstract: Multivariate quality control charts show some advantages to monitor several variables in comparison with the simultaneous use of univariate charts, nevertheless, there are some disadvantages. The main problem is how to interpret the out-ofcontrol signal of a multivariate chart. For example, in the case of control charts designed to monitor the mean vector, the chart signals showing that it must be accepted that there is a shift in the vector, but no indication is given about the variables that have produced this shift. The MEWMA quality control chart is a very powerful scheme to detect small shifts in the mean vector. There are no previous specific works about the interpretation of the out-of-control signal of this chart. In this paper neural networks are designed to interpret the out-of-control signal of the MEWMA chart, and the percentage of correct classifications is studied for different cases.
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: Some quality control tools use non metric subjective information coming from experts, who qualify the intensity of relations existing inside processes, but without quantifying them. In this paper we have developed a quality control analytic tool, measuring the impact or strength of the relationship between process operations and product characteristics. The tool includes two models: a qualitative model, allowing relationships description and analysis; and a formal quantitative model, by means of which relationship quantification is achieved. In the first one, concepts from the Graphs Theory were applied to identify those process elements which can be sources of variation, that is, those quality characteristics or operations that have some sort of prelacy over the others and that should become control items. Also the most dependent elements can be identified, that is those elements receiving the effects of elements identified as variation sources. If controls are focused in those dependent elements, efficiency of control is compromised by the fact that we are controlling effects, not causes. The second model applied adapts the multivariate statistical technique of Covariance Structural Analysis. This approach allowed us to quantify the relationships. The computer package LISREL was used to obtain statistics and to validate the model.
Abstract: With increasing complexity in electronic systems
there is a need for system level anomaly detection and fault isolation.
Anomaly detection based on vector similarity to a training set is used
in this paper through two approaches, one the preserves the original
information, Mahalanobis Distance (MD), and the other that
compresses the data into its principal components, Projection Pursuit
Analysis. These methods have been used to detect deviations in
system performance from normal operation and for critical parameter
isolation in multivariate environments. The study evaluates the
detection capability of each approach on a set of test data with known
faults against a baseline set of data representative of such “healthy"
systems.
Abstract: In this paper we introduce a novel method for
the characterization of synchronziation and coupling effects
in multivariate time series that can be used for the analysis
of EEG or ECoG signals recorded during epileptic seizures.
The method allows to visualize the spatio-temporal evolution
of synchronization and coupling effects that are characteristic
for epileptic seizures. Similar to other methods proposed for
this purpose our method is based on a regression analysis.
However, a more general definition of the regression together
with an effective channel selection procedure allows to use the
method even for time series that are highly correlated, which
is commonly the case in EEG/ECoG recordings with large
numbers of electrodes. The method was experimentally tested
on ECoG recordings of epileptic seizures from patients with
temporal lobe epilepsies. A comparision with the results from
a independent visual inspection by clinical experts showed
an excellent agreement with the patterns obtained with the
proposed method.
Abstract: In manufacturing industries, development of measurement leads to increase the number of monitoring variables and eventually the importance of multivariate control comes to the fore. Statistical process control (SPC) is one of the most widely used as multivariate control chart. Nevertheless, SPC is restricted to apply in processes because its assumption of data as following specific distribution. Unfortunately, process data are composed by the mixture of several processes and it is hard to estimate as one certain distribution. To alternative conventional SPC, therefore, nonparametric control chart come into the picture because of the strength of nonparametric control chart, the absence of parameter estimation. SVDD based control chart is one of the nonparametric control charts having the advantage of flexible control boundary. However,basic concept of SVDD has been an oversight to the important of data characteristic, density distribution. Therefore, we proposed DW-SVDD (Density Weighted SVDD) to cover up the weakness of conventional SVDD. DW-SVDD makes a new attempt to consider dense of data as introducing the notion of density Weight. We extend as control chart using new proposed SVDD and a simulation study of various distributional data is conducted to demonstrate the improvement of performance.
Abstract: The early diagnostic decision making in industrial processes is absolutely necessary to produce high quality final products. It helps to provide early warning for a special event in a process, and finding its assignable cause can be obtained. This work presents a hybrid diagnostic schmes for batch processes. Nonlinear representation of raw process data is combined with classification tree techniques. The nonlinear kernel-based dimension reduction is executed for nonlinear classification decision boundaries for fault classes. In order to enhance diagnosis performance for batch processes, filtering of the data is performed to get rid of the irrelevant information of the process data. For the diagnosis performance of several representation, filtering, and future observation estimation methods, four diagnostic schemes are evaluated. In this work, the performance of the presented diagnosis schemes is demonstrated using batch process data.
Abstract: This study employs a bivariate asymmetric GARCH
model to reveal the hidden dynamics price changes and volatility
among the emerging markets of Thailand and Malaysian after the
Asian financial crisis from January 2001 to December 2008. Our
results indicated that the equity markets are sharing the common
information (shock) that transmitted among each others. These
empirical findings are used to demonstrate the importance of shock
and volatility dynamic transmissions in the cross-market hedging and
market risk.
Abstract: A system for market identification (SMI) is presented.
The resulting representations are multivariable dynamic demand
models. The market specifics are analyzed. Appropriate models and
identification techniques are chosen. Multivariate static and dynamic
models are used to represent the market behavior. The steps of the
first stage of SMI, named data preprocessing, are mentioned. Next,
the second stage, which is the model estimation, is considered in more
details. Stepwise linear regression (SWR) is used to determine the
significant cross-effects and the orders of the model polynomials. The
estimates of the model parameters are obtained by a numerically stable
estimator. Real market data is used to analyze SMI performance.
The main conclusion is related to the applicability of multivariate
dynamic models for representation of market systems.
Abstract: In the automotive industry test drives are being conducted
during the development of new vehicle models or as a part of
quality assurance of series-production vehicles. The communication
on the in-vehicle network, data from external sensors, or internal
data from the electronic control units is recorded by automotive
data loggers during the test drives. The recordings are used for fault
analysis. Since the resulting data volume is tremendous, manually
analysing each recording in great detail is not feasible.
This paper proposes to use machine learning to support domainexperts
by preventing them from contemplating irrelevant data and
rather pointing them to the relevant parts in the recordings. The
underlying idea is to learn the normal behaviour from available
recordings, i.e. a training set, and then to autonomously detect
unexpected deviations and report them as anomalies.
The one-class support vector machine “support vector data description”
is utilised to calculate distances of feature vectors. SVDDSUBSEQ
is proposed as a novel approach, allowing to classify subsequences
in multivariate time series data. The approach allows to
detect unexpected faults without modelling effort as is shown with
experimental results on recordings from test drives.
Abstract: Nevertheless the widespread application of finite
mixture models in segmentation, finite mixture model selection is
still an important issue. In fact, the selection of an adequate number
of segments is a key issue in deriving latent segments structures and
it is desirable that the selection criteria used for this end are effective.
In order to select among several information criteria, which may
support the selection of the correct number of segments we conduct a
simulation study. In particular, this study is intended to determine
which information criteria are more appropriate for mixture model
selection when considering data sets with only categorical
segmentation base variables. The generation of mixtures of
multinomial data supports the proposed analysis. As a result, we
establish a relationship between the level of measurement of
segmentation variables and some (eleven) information criteria-s
performance. The criterion AIC3 shows better performance (it
indicates the correct number of the simulated segments- structure
more often) when referring to mixtures of multinomial segmentation
base variables.