Quint Wiersma's repositories
StateSpace.jl
Provides methods for a linear Gaussian State Space model such as filtering (Kalman filter), smoothing (Kalman smoother), forecasting, likelihood evaluation, and estimation of hyperparameters (Maximum Likelihood, Expectation-Maximization (EM), and Expectation-Conditional Maximization (ECM), w/ and w/o penalization)
DynamicFactorModels.jl
Provides methods dynamic factor models, such as estimation, w/ and w/o penalization, forecasting and filtering
GradientDescent.jl
Provides gradient descent methods, such as Newton and quasi-Newton methods, for differentiable objective functions.
ProximalMethods.jl
Provides proximal operator evaluation routines and proximal optimization algorithms, such as (accelerated) proximal gradient methods and alternating direction method of multipliers (ADMM), for non-smooth/non-differentiable objective functions.
TensorAutoregressions.jl
Provides methods for tensor-valued autoregressive modelling, such as estimation, forecasting and impulse response function (IRF) analysis
Hyperopt.jl
Hyperparameter optimization in Julia.
LineSearch.jl
Provides line search methods, such as backtracking, to determine the optimal step size in optimization and root-finding algorithms.