This is a test framework for goodness-of-fit statistic tests.
Framework consists of 5 modules
- Core module - provides distributions, cdf, pdf etc.
- Persistence module - provides different stores to store data.
- Experiment module - provides pipeline for experiment and default components for pipeline.
- Expert system module - provides expert system for goodness-of-fit testing.
- Tests module - provides different goodness-of-fit tests.
CriticalValueSqLiteStore - store critical values and target distributions in SQLite.
CriticalValueFileStore - store critical values and target distributions in JSON and CSV.
RvsSqLiteStore - store generated rvs in SQLite.
RvsFileStore - store generated rvs in CSV.
PowerResultSqLiteStore - store PowerCalculationWorker result in SQLite
PowerCalculationWorker - calculates goodness-of-fit test power
| № | Test | Status |
|---|---|---|
| 1 | Anderson–Darling | Done |
| 2 | Chi square | Done |
| 3 | Kolmogorov–Smirnov | Done |
| 4 | Lilliefors | Done |
| 5 | Cramér–von Mises | Done |
| 6 | Min-Toshiyuki | Done |
| 7 | Smith and Brian | Done |
| 8 | Ozturk and Korukoglu | Done |
| 9 | Tiku-Singh | Done |
| 10 | Lockhart-O'Reilly-Stephens | Done |
| 11 | Mann-Scheuer-Fertig | Done |
| 12 | Evans, Johnson and Green | Done |
| 13 | Skewness | Done |
| 14 | Kurtosis | Done |
| 15 | Statistic based on stabilized probability plot | Done |
| 16 | Test statistic of Shapiro Wilk | Done |
| Test | Second Header |
|---|---|
| Ozturk and Korukoglu | Content Cell |
| Jackson | Content Cell |
| Lewis | Content Cell |
| Test | Second Header |
|---|---|
| Anderson–Darling | Content Cell |
| Kolmogorov–Smirnov | Content Cell |
| Chi square | Content Cell |
| skewness | Content Cell |
| kurtosis | Content Cell |
git clone https://github.com/PySATL/pysatl-experiment
cd pysatl-experiment
git submodule add https://github.com/PySATL/pysatl-criterion.git pysatl_criterion
git submodule update --init --recursiveInstall dependencies:
poetry installInstall requirements
poetry install --with devInstall pre-commit hooks:
poetry run pre-commit installStarting manually:
poetry run pre-commit run --all-files --color always --verbose --show-diff-on-failure- Creating of experiment.
poetry run experiment create NAME- Set the experiment type value. Experiment types: critical_value, power, time_complexity.
poetry run experiment configure NAME experiment-type critical_value - Setting the hypothesis value. Experiment types: normal, exponential, weibull.
poetry run experiment configure NAME hypothesis normal- Set the sample size value. (min = 10)
poetry run experiment configure NAME sample-sizes 23- Setting the value of the Monte Carlo number. (min = 100)
poetry run experiment configure NAME monte-carlo-count 154- Setting the significance levels.
poetry run experiment configure NAME significance-levels 0.05 0.01- Setting the criteria.
poetry run experiment configure NAME criteria KS- Setting the file name for connecting the storage.
poetry run experiment configure NAME storage-connection FILENAME - Running the experiment.
poetry run experiment build-and-run NAME Parameters experiment-type, hypothesis, sample-sizes, monte-carlo-count, significance-levels, storage-connection required to set values.