Mathematics,
Год журнала:
2023,
Номер
11(18), С. 3832 - 3832
Опубликована: Сен. 7, 2023
This
study
investigates
the
use
of
a
novel
market
graph
model
for
equity
markets.
Our
is
built
on
distance
correlation
instead
traditional
Pearson
correlation.
We
apply
it
to
S&P500
stocks
from
January
2015
December
2022.
also
compare
our
graphs
in
literature,
those
using
To
further
comparison,
we
build
Spearman
rank
comparisons
reveal
that
non-linear
relationships
stock
returns
are
not
captured
by
either
or
observe
robust
measure
detecting
complex
returns.
Networks
networks,
shown
be
more
responsive
conditions
during
turbulent
periods
such
as
COVID
crash
period.
Information Fusion,
Год журнала:
2024,
Номер
113, С. 102616 - 102616
Опубликована: Авг. 5, 2024
As
a
core
branch
of
financial
forecasting,
stock
forecasting
plays
crucial
role
for
analysts,
investors,
and
policymakers
in
managing
risks
optimizing
investment
strategies,
significantly
enhancing
the
efficiency
effectiveness
economic
decision-making.
With
rapid
development
information
technology
computer
science,
data-driven
neural
network
technologies
have
increasingly
become
mainstream
method
forecasting.
Although
recent
review
studies
provided
basic
introduction
to
deep
learning
methods,
they
still
lack
detailed
discussion
on
architecture
design
innovative
details.
Additionally,
latest
research
emerging
large
language
models
structures
has
yet
be
included
existing
literature.
In
light
this,
this
paper
comprehensively
reviews
literature
networks
field
from
2015
2023,
discussing
various
classic
structures,
including
Recurrent
Neural
Networks
(RNNs),
Convolutional
(CNNs),
Transformers,
Graph
(GNNs),
Generative
Adversarial
(GANs),
Large
Language
Models
(LLMs).
It
analyzes
application
achievements
these
market
Moreover,
article
also
outlines
commonly
used
datasets
evaluation
metrics
further
exploring
unresolved
issues
potential
future
directions,
aiming
provide
clear
guidance
reference
researchers
Journal of Forecasting,
Год журнала:
2025,
Номер
unknown
Опубликована: Март 16, 2025
ABSTRACT
This
study
proposes
a
novel
deep
auto‐optimized
architecture
for
stock
price
forecasting
that
integrates
sectoral
behavior
with
individual
sentiment
to
improve
predictive
accuracy.
Traditional
prediction
models
often
focus
solely
on
behavior,
overlooking
the
impact
of
broader
trends.
The
proposed
approach
utilizes
advanced
learning
models,
including
gated
recurrent
units
(GRU),
bidirectional
GRU,
long
short‐term
memory
(LSTM),
and
LSTM,
their
hybrid
ensembles.
These
are
built
using
Keras
functional
API
auto
ML
network
search
technology.
current
multimodal
framework
incorporates
significantly
improving
performance
metrics.
research
highlights
critical
role
integrating
in
models.
