European journal of management, economics and business.,
Год журнала:
2024,
Номер
1(3), С. 84 - 99
Опубликована: Ноя. 1, 2024
This
paper
presents
a
detailed
analysis
of
the
Holt-Winters-GRU
hybrid
model
for
predicting
global
rice
prices,
an
essential
agricultural
commodity.
The
benefits
traditional
statistical
approaches
are
combined
with
deep
learning
power,
and
results
have
been
found
to
outperform
standalone
GRU.
produced
test
RMSE
27.7532
almost
no
difference
between
training
testing
errors,
thus
showing
robust
generalization
ability.
Detailed
scrutiny
weight
heat
map
GRU
layer
reflects
intricacies
while
depicting
both
seasonal
patterns
intricate
nonlinear
relationships
present
in
price
time
series.
findings
from
study
reveal
that
is
usable
forecasting
movements
policymakers,
traders,
market
analysts,
considering
its
ability
handle
systematic
trends
shocks.
Recommendations
implementation,
enhancement,
risk
management,
policy
applications,
future
research
provided
extend
further
utility
this
approach
commodity
markets.
Energies,
Год журнала:
2024,
Номер
17(16), С. 4142 - 4142
Опубликована: Авг. 20, 2024
Accurate
power
load
forecasting
can
provide
crucial
insights
for
system
scheduling
and
energy
planning.
In
this
paper,
to
address
the
problem
of
low
accuracy
prediction,
we
propose
a
method
that
combines
secondary
data
cleaning
adaptive
variational
mode
decomposition
(VMD),
convolutional
neural
networks
(CNN),
bi-directional
long
short-term
memory
(BILSTM),
adding
attention
mechanism
(AM).
The
Inner
Mongolia
electricity
were
first
cleaned
use
K-means
algorithm,
then
further
refined
with
density-based
spatial
clustering
applications
noise
(DBSCAN)
algorithm.
Subsequently,
parameters
VMD
algorithm
optimized
using
multi-strategy
Cubic-T
dung
beetle
optimization
(CTDBO),
after
which
was
employed
decompose
twice-cleaned
sequences
into
number
intrinsic
functions
(IMFs)
different
frequencies.
These
IMFs
used
as
inputs
CNN-BILSTM-Attention
model.
model,
CNN
is
feature
extraction,
BILSTM
extracting
information
from
sequence,
AM
assigning
weights
features
optimize
prediction
results.
It
proved
experimentally
model
proposed
in
paper
achieves
highest
robustness
compared
other
models
exhibits
high
stability
across
time
periods.
IEEE Access,
Год журнала:
2024,
Номер
12, С. 114760 - 114785
Опубликована: Янв. 1, 2024
Financial
market
prediction
has
shown
considerable
potential
in
the
past
few
years
from
combination
of
contemporary
Deep
Learning
(DL)
techniques
and
traditional
time
series
forecasting
methodologies.
To
predict
stock
prices
three
distinct
companies
General
Electric
(GE),
Microsoft
(MSFT),
Amazon
(AMZN)
datasets.
This
study
presents
a
novel
hybrid
model
that
combines
Double
Exponential
Smoothing
(DES)
method
with
Dual
Attention
Encoder-Decoder
based
Bi-directional
GRU,
optimized
using
Bayesian
Optimization
(DA-ED-Bi-GRU-BO).
By
combining
best
features
old
methods,
seeks
to
efficiently
identify
patterns
trends
data.
When
handling
data,
DES
offers
reliable
flexible
mechanism
considers
seasonality
The
DA-ED-Bi-GRU
added
deep
learning
further
improves
its
comprehension
intricate
found
parameters
are
adjusted
optimization
(BO)
maximize
model's
performance.
Several
performance
indicators,
such
as
Mean
Absolute
Error
(MAE),
Squared
(MSE),
Root
(RMSE),
R-Square
(
R
2
),
Theil's
U-Statistics
(TUS),
used
assess
effectiveness
model.
These
measures
offer
thorough
insights
into
precision,
dependability,
accuracy
predictions.
experimental
findings
show
proposed
ability
GE,
MSFT,
AMZN
values
reasonable
accuracy.
Along
framework,
DL
conventional
smoothing
approaches
combine
provide
potent
tool
may
help
traders
investors
make
wise
judgments.
