Index
A
Activation function
Adam() optimizer
add() method
Applications in economics and finance
Arbitrary model
Autoencoder loss function
Autoencoder model
actual and OLS-predicted GDP growth
architecture
dimensionality reduction
regression setting, latent state
generative machine learning
noise reduction
functions
latent state time series
loss function, minimizing
neural network
predict() method
reconstructed series for US GDP growth
train, Keras API
Automatic differentiation
Autoregressive coefficient
Autoregressive model
B
Bag-of-words (BoW) model
CountVectorizer()
document-term matrix
fit_transform()
inverse document frequencies
sklearn.feature_extraction
submodules
term-frequency inverse-document frequency (tf-idf) metric
Batch matrix multiplication
Batch normalization
Batch size
Bayesian regression methods
Big Data
Binary_crossentropy loss
Binary cross-entropy loss function
BoostedTreesClassifier
Boosted trees regressor
C
Cake-eating problem
Central index key (CIK)
Chain rule
Closed-form expressions
Confidence intervals
Confusion_matrix
Constant tensors
Convolutional neural networks (CNNs)
convolutional layer
image data, training
model architecture
sequential model
training
CounterVectorizer()
CountVectorizer()
Custom loss function
Custom reinforcement learning environment
D
Data collection
Data generation functions
Data preparation
BeautifulSoup
characters to lowercase, convertion
convert to lowercase
cleaning process
installing NLTK
join paragraphs
non-word elements
remove special characters
remove stop words and rare words
replace words
6-K filing
stem/lemmatize
text into sentences
urlopen submodule
Decision tree
feature engineering
Gini impurity
HMDA mortgage application data
information gain
maximum depth
model, HMDA data
nodes
recursive sample splitting
training
Decoder function
Deep learning
Deep neural network classifiers
Deep neural network (DNN)
Deep Q-learning
Deep reinforcement learning
Denoisers
Dense neural networks
forecast of inflation
generated sequences
model architecture
np.array()/tf.constant() objects
overlapping sequences
sequence generator for inflation
summary() method
time series
Derivatives of polynomials
Dictionary-based methods
DataFrame
Loughran-McDonald measure
net positivity index
positive word counts
read_excel submodule
stemmed LM dictionary
text analysis in economics
Differential calculus
automatic differentiation
common derivatives of polynomials
computing derivatives
in economics
first and second derivatives
in machine learning
multidimensional derivatives
gradients
Hessian
Jacobian
transcendental functions
Dimensionality reduction in economics
noise reduction
PCA
SeePrincipal component analysis (PCA)
PLS
sklearn and tensorflow
Dirichlet distribution
Discriminator model
Discriminator vs. generator models
DNNClassifier
Document-feature matrix
Document-term (DT) matrix
Dot product of vectors
Dropout
Dynamic embedded topic model (D-ETM)
E
Economic policy uncertainty (EPU)
Economics
active research and predictions
machine learning impact
off-the-shelf routines
policy analysis
traditional econometric methods
Economists
EDGAR search interface, company filings
Efficient Markets Hypothesis (EMH)
Elastic net regression
Elementwise tensor multiplication
Empirical analysis
Encoder function
Epsilon-greedy policy
Estimators API
Estimators approach
Euler equation
Exponential smoothed autoregressive models (ESTAR)
F
Factor analysis (FA)
Factor-augmented vector autoregressions (FAVAR)
Feature extraction
First-order condition (FOC)
fit() method
fit_generator() method
Forward propagation
Functional API
G
Generative adversarial networks (GANs)
adversarial model
discriminator model
discriminator networks
equilibrium condition, image generation
GDP growth data
generative model
generator component
Generative machine learning
get_feature_names()
Gini impurity
Global minimum
Google Colaboratory (Colab) notebook
gradient() method
Gradient boosted trees
classification trees
model parameters
model residual
and random forest
regression trees
tree and prediction function
Gradients
GradientTape()
Graphics processing units (GPUs)
H
Hessian matrix
Hidden layer
Home Mortgage Disclosure Act (HMDA)
I
Image data
k-tensor of pixel intensities
matplotlib.pyplot
numpy arrays
RGB image
ships in satellite imagery
use in TensorFlow
Image datasets
imshow() function
Information entropy
Information gain
J
Jacobian
K
Keras
class weights
compile and train, model
confusion matrix
evaluate model
functional API
LSTM model
model summary, print
neural network
Sequential API
k-fold cross-validation
k regressors
Kullback-Leibler (KL) divergence
L
lassoLoss()
LASSO regression
Latent Dirichlet allocation (LDA) model
assumptions
document corpus
issues
limitations
parameters
6-K filing text data
theoretical overview of static
time series contexts
Learning decay
Learning method
Least absolute deviations (LAD) regression
alphaHat and betaHat
closed-form algebraic expression
input data for linear regression
loss function
minimize() method
Monte Carlo experiment
optimization
parameter estimate counts
parameter training histories
parameter values
stddev parameter
tf.reduce_mean()
true values, model parameters
variables initialize
Least absolute errors (LAE)
Least absolute shrinkage and selection parameter (LASSO) model
Linear activation function
Linear Algebra
broadcasting
batch matrix multiplication
scalar-tensor addition and multiplication
scalar addition and multiplication
tensor addition
tensor multiplication
dot product
elementwise
matrix
Linear function, slope
Linear model
regressor
Linear regression
LAD
SeeLeast absolute deviations (LAD) regression
loss functions
OLS
overview
Linear regression model
Logistic regression
Long short-term memory (LSTM)
Loss functions
continuous dependent variables
discrete dependent variables
submodules of TensorFlow
tf.losses submodule
Loughran-McDonald (LM) dictionary
M
Machine learning in economics
distributed training
extensions
flexibility
high-quality documentation
macroeconometric analysis
production quality
traditional problems
Macroeconometric analysis
Markov chain Monte Carlo (MCMC)
Matrix addition
Matrix multiplication
Mean absolute error (MAE)
Mean absolute error (MAE) loss
Mean absolute percentage error (MAPE)
Mean squared error loss
Mean squared logarithmic error (MSLE)
Model fine-tuning
model.predict_generator(generator)
Model selection
Model tuning
Monte Carlo simulation
Multidimensional derivatives
Multivariate forecasts
gradient boosted trees
load and preview inflation forecast data
LSTM
N
Natural language processing (NLP)
Natural language toolkit (NLTK)
Neoclassical business cycle model
Neural networks
forward propagation
in Keras
layers
linear regression model
modification
reshape images
Noise reduction
Non-linear regression
autoregressive model
exchange rates
load data
loss function
minimization of loss function
model, define
optimization
random walk model, nominal exchange rate
TAR model
TAR model, USD-GBP exchange rate
train TAR model
Non-linear text regression
np.array() format
Numerical differentiation
numpy() method
numpy arrays
O
Optimization algorithms
Optimizers
instantiate
modern extensions
SGD optimizer
Ordinary least squares (OLS)
P
Partial least squares (PLS)
Partially linear models
alphaHat and betaHat
arbitrary model
construction and training
data generation
econometric challenges
epoch of training
initialize variables
linear regression model
loss function, defining
minimize method
Monte Carlo experiment
non-linear function
non-linear model
parameters
parameter values
Penalized linear regression
Penalized regression
Penalty function
Poisson distribution of word counts
Policy analysis
Policy function
Polynomials, differentiation rules
predict() method
Prediction policy problems
Pretrained models
feature extraction
model fine-tuning
Principal component analysis (PCA)
actual and PCR-predicted GDP growth
association strengths
dimensionality of dataset
elbow method
fit() method
library import, sklearn
minimization problem
PCR
plot of variance
regression
sklearn perform
TensorFlow perform
variables, defining
Principal component regression (PCR)
Python
Q
Q-learning
R
Ragged tensor
Random forest model
Random tensors
Random walk model
Reconstruction loss
Recurrent neural networks (RNNs)
architecture in Keras model
blue nodes
cells
compile and train
fit_generator() method
inflation array
in Keras
linear activation function
multiplication step
np.expand_dims()
one-quarter-ahead forecast of inflation
output value
pink nodes
sequence generator for inflation
sequential data
SimpleRNN layer
summary() method
Regression
linear
SeeLinear regression
logistic
Regression trees
Regular expression
Regularized regression
reset() method
Ridge regression
S
Scalar addition and multiplication
Scalar-tensor addition
Scalar-tensor multiplication
Second-order condition (SOC)
Sequential() model
Sequential API
Sequential models
dense neural networks
SeeDense neural networks
intermediate hidden states
LSTM
RNNs
Sequential vs. parallel training
SIC classification codes
Sigmoid function
6-K financial filing, metal mining company
sklearn
Smooth transition autoregressive models (STAR)
Sparse categorical cross-entropy loss function
Standard industrial classification (SIC) code
Standard normal distribution
Stochastic gradient descent (SGD)
algorithm
optimizer
summary() method
Support vector machine (SVM) models
T
Tensor addition
TensorFlow
automatic differentiation
computational graph for OLS
computing derivatives
constants and variables
differential calculus
SeeDifferential calculus
documentation
economics and finance
machine learning
theoretical models
estimators library
installation
linear algebra
SeeLinear Algebra
loading data
logs, TensorBoard visualization
OLS model with tf.estimator()
OLS model with tf.keras()
OLS predictions with static graphs
OLS regression
open source library, machine learning
predictions
print tensors
static computational graph
Tensor multiplication
Tensor processing units (TPUs)
Tensors
Term-document matrix
Text analysis
Text as data
applications
representation
statistical methods
Text-based regression
Text classification
Text data notation
Text regression
predict() method
tf.Variable()
Text regression
compile and train, Keras model
continuous dependent variable
deep learning model
document-term matrix
Keras model, architecture
LAD regression
LASSO regression
loss function
minimization problem, penalized estimator
penalized regression
perform optimization
predicted values
regression model
train LASSO model
tf.add()
tf.constant()
tf.estimator.DNNClassifier()
tf.estimator.DNNRegressor()
TfidfVectorizer()
tf.keras.Input() method
tf.keras.layers.Concatenate() operation
tf.keras.layers.Dense()
tf.keras.Sequential()
tf.optimizers.SGD()
tf.Session()
tf.Variable()
Theoretical models
cake-eating problem
neoclassical business cycle model
Threshold autoregressive (TAR) models
TimeseriesGenerator()
toarray() method
Topic modeling
assign topic probabilities to sentences
assumptions
distribution
document corpus
framework
“generative”
LDA
SeeLatent Dirichlet allocation (LDA) model
“probabilistic”
6-K filing text data
topic proportions by sentence
transform() method
vector of weights
Traditional econometric methods
confidence intervals
empirical analysis
model selection
train_test_split
Transcendental functions
transform() method
Tree-based models
U
Unsupervised method, embeddings
US Securities and Exchange Commission (SEC) filings
V
Vanishing gradient problem
Variational autoencoders (VAEs)
architecture
decoder model
define function, sampling task
encoder model
GDP growth data, preparation
generated time series
generative tasks
implementation in TensorFlow
KL divergence
latent states
latent states and time series
limitations
loss, define
loss function
mean and lvar parameters
model architecture, defining
predict() method
sampling function
Vector addition
W, X, Y, Z
Wasserstein GANs
wordDist
Word embeddings
Word sequences
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.