Metadata-Version: 2.4
Name: tsorder
Version: 0.1.0
Classifier: Programming Language :: Rust
Classifier: Programming Language :: Python :: 3
Classifier: Topic :: Scientific/Engineering :: Mathematics
Summary: Order selection for time series models: information criteria, Fine-Hwang threshold estimators, McMillan-degree estimators, ARMA identification and integration order, in a dependency-free Rust core
Author: Paulo Renato Alves Firmino
Author-email: André Leite <leite@castlab.org>, Raydonal Ospina <raydonal@castlab.org>, Cristiano Ferraz <cferraz@castlab.org>, Hugo Vasconcelos <hugo.vasconcelos@ufpe.br>, Diogo Bezerra <diogo.bezerra@ufpe.br>
License-Expression: MIT
Requires-Python: >=3.9
Description-Content-Type: text/markdown; charset=UTF-8; variant=GFM
Project-URL: Homepage, https://castlaboratory.github.io/tsorder/
Project-URL: Repository, https://github.com/castlaboratory/tsorder

# tsorder (Python)

Thin PyO3 shell over the dependency-free Rust core `tsorder-core` (no
statistical logic lives here). Accepts lists, numpy arrays or pandas
Series; returns plain dicts. Bit-identical to the R package by
construction — both replay the same golden fixture in their test suites.

```python
# from the repository checkout (needs Rust >= 1.81):
#   pip install maturin && maturin develop --release -m bindings/python/Cargo.toml
import tsorder

fit = tsorder.fh_order(x)                 # PACF-clip, the default
fit["order"], fit["tau"]

tsorder.ic_order(x, criterion="bic")      # classical criteria
tsorder.svc_order(x)                      # McMillan degree
tsorder.integration_order(x)              # how many differences
tsorder.hannan_rissanen(x)                # ARMA (p, q)
```

See the [project page](https://castlaboratory.github.io/tsorder/) and the
[R package site](https://castlaboratory.github.io/tsorder/r/).

