| zoomgrid-package | R Documentation |
The zoomgrid package implements a grid search algorithm with optional zoom-in layers. The method is intended for objective functions with multiple local optima, where standard optimisation routines may be sensitive to initial values.
Core idea. A standard grid search proceeds as follows:
Construct a grid of candidate points over a bounded parameter domain.
Evaluate the objective function at each grid point.
Return the best point(s) found on the grid (for minimisation or maximisation).
A sufficiently fine grid may yield a good approximation to the global optimum, but the number of grid points grows quickly with dimension, making exhaustive evaluation expensive.
Zoom-in layers. To reduce computational cost, zoomgrid provides an optional
zoom-in mechanism. After an initial grid search, the algorithm selects the best n
points (controlled by num) and constructs smaller subgrids around those points.
This procedure is repeated for k additional layers (controlled by zoom).
At each layer, subgrid sizes shrink according to the decay rate decay, which allows
the algorithm to concentrate evaluations in promising neighbourhoods while keeping the total
number of function evaluations manageable.
In typical use, the method serves one of the following purposes:
provide suitable initial values for subsequent optimisation,
or approximate the optimiser when a very fine one-shot grid is impractical.
Parallel evaluation. Function evaluations can be parallelised by setting
parallel = TRUE. Parallel execution is implemented via the future framework
using a cross-platform backend. The packages future and future.apply are listed
under ‘Suggests’ and are required only when parallel execution is requested.
Yukai Yang
Department of Statistics, Uppsala University
Yang, Y. (2012). Modelling Nonlinear Vector Economic Time Series. PhD thesis, Aarhus University, Department of Economics and Business and CREATES. See Section 1.5.4.
build_grid: construct a grid over the parameter domain.
grid_search: perform grid search with optional zoom-in layers.
grid_search_check: estimate runtime under given settings.
Maintainer: Yukai Yang yukai.yang@statistik.uu.se (ORCID)
Useful links:
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.