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 current study begins with an awareness that
today-s media environment is characterized by technological
development and a new way of reading caused by the introduction of
the Internet. The researcher conducted a meta analysis framed within
Technological Determinism to investigate the process of hypertext
reading, its differences from linear reading and the effects such
differences can have on people-s ways of mentally structuring their
world. The relationship between literacy and the comprehension
achieved by reading hypertexts is also investigated. The results show
hypertexts are not always user friendly. People experience hyperlinks
as interruptions that distract their attention generating comprehension
and disorientation. On one hand hypertextual jumping reading
generates interruptions that finally make people lose their
concentration. On the other hand hypertexts fascinate people who
would rather read a document in such a format even though the
outcome is often frustrating and affects their ability to elaborate and
retain information.
Abstract: Multi-loop (De-centralized) Proportional-Integral-
Derivative (PID) controllers have been used extensively in process
industries due to their simple structure for control of multivariable
processes. The objective of this work is to design multiple-model
adaptive multi-loop PID strategy (Multiple Model Adaptive-PID)
and neural network based multi-loop PID strategy (Neural Net
Adaptive-PID) for the control of multivariable system. The first
method combines the output of multiple linear PID controllers,
each describing process dynamics at a specific level of operation.
The global output is an interpolation of the individual multi-loop
PID controller outputs weighted based on the current value of the
measured process variable. In the second method, neural network
is used to calculate the PID controller parameters based on the
scheduling variable that corresponds to major shift in the process
dynamics. The proposed control schemes are simple in structure with
less computational complexity. The effectiveness of the proposed
control schemes have been demonstrated on the CSTR process,
which exhibits dynamic non-linearity.