v0.2.2

First release with own optimization algorithms.

• parallel Nelder-Mead by Jacek Barszczewski
• parallel Pounders by Sebastian Gsell

The algorithms are still work in progress and will get more convergence criteria, support for bounds, etc. in the future.

v0.2.1

Mainly improvements to the documentation.

v0.2.0

Release 0.2.0

Add a lot of new functionality with a few minor breaking changes. We have more
optimizers, better error handling, bootstrap and inference for method of simulated
moments.

Breaking changes

• logging is disabled by default during optimization.
• the log_option "if_exists" was renamed to "if_table_exists"
• The comparison plot function is removed.
• first_derivative now returns a dictionary, independent of arguments.
• structure of the logging database has changed
• there is an additional boolean flag named scaling in minimize and maximize

New features

• Optimizer benchmarking
• Multistart optimization
• More optimizers (fides, ipopt, nlopt, pygmo)
• estimate_msm and estimate_ml functions
• More diagnostic tools for numerical derivatives

v0.1.3

• Improve numerical stability of inference functions
• Better documentation

v0.1.2

• Small fixes to documentation
• Now compatible with newest pandas
• By default not all inputs to maximize/minimize are saved in database

v0.1.1

Complete overhaul of estimagic.

• Restructured documentation
• New internal optimizer interface
• Better handling of derivatives
• Harmonized algo_options

v0.0.31

Release v0.0.31 and relax some version constraints. (#162)

v0.0.30

• Allow custom gradient in maximize and minimize

v0.0.29

• Make estimagic compatible with pandas 1.0

v0.0.28

• New command line interface to dashboard
• Estimagic becomes a noarch package