fue — Exact Maximum Likelihood for Univariate Time Series

fue is a Python implementation of the FUE/FUF estimation engine originally written in C by Arthur B. Treadway and David E. Guerrero, based on the algorithms designed and coded by José Alberto Mauricio. It fits ARMAX models with linear transfer-function interventions by exact maximum likelihood using the Ansley (1979) innovations form and the Mauricio (1997, AS 311) / Mauricio (1995, JASA) algorithms.

Features

Exact ML estimation of ARIMA/SARIMA/ARMAX models (Box–Jenkins family)
Seasonal and non-seasonal AR/MA operators; regular and seasonal differencing
Linear transfer function interventions: impulse, step, ramp, Fourier harmonics, alternator, and arbitrary user-defined regressors
Fixed-frequency AR/MA factors (AR(2) with constrained spectral peak)
Box–Cox transformation with automatic back-transformation of forecasts
Multi-step forecasts with asymptotic prediction intervals
ASCII .out and .pre report output compatible with the C FUE binary
HTML forecast reports (requires jinja2)
CLI tools fue and fuf mirroring the C binaries

Installation

pip install fue                  # pure Python (numpy + scipy + matplotlib)
pip install "fue[report]"        # + HTML forecast reports (jinja2)
pip install "fue[pdf]"           # + PDF export (weasyprint)

Python 3.10+ required.

Quick start

import numpy as np
import fue

# Build a time series
data = np.array([...])   # monthly observations
ts = fue.TimeSeries(data, freq=12, start=(2002, 1), name="CPI")

# Specify an ARIMA(1,1,0)(1,0,0)_12 model with a log transform
m = fue.Model(ts,
              ar=[[0.3]],          # AR(1), initial value 0.3
              sar=[[0.2]],         # seasonal AR(1) at lag 12
              d=1,                 # one regular difference
              boxlam=0.0,          # log transform (Box–Cox λ=0)
              refactor=100.0)

result = m.fit()
print(result.sigma2, result.aic)
m.plot_residuals()

Load from a `.inp` file (fue format)

ts, m = fue.load("model.inp")
result = m.fit()
m.write_out("model.out")
m.write_pre("model.pre")

Forecast from a pre-estimated `.inp` file (fuf format)

ts, m = fue.load_fuf("forecast_model.inp")
fr = m.forecast_fuf(horizon=24)
print(fr.level)          # point forecasts in original units
print(fr.level_std)      # forecast standard deviations

Command-line interface

# Estimate a model and write .out / .pre
fue model_name [eml|aml] [chk|nochk] [-f horizon]

# Generate forecasts from a pre-estimated .inp file
fuf forecast_model

Numerical methods

Algorithm	Reference	Used for
Ansley (1979) innovations form	Mauricio (1997) AS 311	Exact log-likelihood (`elf_scalar`)
Kalman filter (quick recursions)	Mélard (1984) AS 197	Inner BFGS loop (`flikam_scalar`)
BFGS with Cholesky factor update	Dennis & Schnabel (1983) ch. 9	Optimization (`raxopt`)
Scaled objective Π(x)/Π₀	Mauricio (1995) JASA §3	Numerical conditioning

Bug tracking

Bugs are tracked in-repo under bugs/ — one Markdown file per report (BUG-NNNN-slug.md) with a small frontmatter schema. A fix references the id in its commit, e.g. fix(forecast): BUG-0001 ….

fue-bug list                       # list reports (open marked with *)
fue-bug show BUG-0001              # print a report
fue-bug new "title" --component forecast   # file a new report
fue-bug check                      # validate all reports (runs in CI, tests/test_bugs.py)
fue-bug index                      # regenerate bugs/README.md

Authors and licence

fue is developed by Arthur B. Treadway and David E. Guerrero, based on the algorithms designed and coded by José Alberto Mauricio.

Released under the GNU General Public Licence v2.0 or later (GPL-2.0-or-later). See COPYING for the full licence text.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

fue-0.1.9.tar.gz (198.6 kB view details)

Uploaded Aug 5, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

fue-0.1.9-py3-none-any.whl (92.1 kB view details)

Uploaded Aug 5, 2026 Python 3

fue-0.1.9-cp313-cp313-win_amd64.whl (147.9 kB view details)

Uploaded Aug 5, 2026 CPython 3.13Windows x86-64

fue-0.1.9-cp313-cp313-musllinux_1_2_x86_64.whl (1.2 MB view details)

Uploaded Aug 5, 2026 CPython 3.13musllinux: musl 1.2+ x86-64

fue-0.1.9-cp313-cp313-musllinux_1_2_aarch64.whl (1.1 MB view details)

Uploaded Aug 5, 2026 CPython 3.13musllinux: musl 1.2+ ARM64

fue-0.1.9-cp313-cp313-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (1.4 MB view details)

Uploaded Aug 5, 2026 CPython 3.13manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

fue-0.1.9-cp313-cp313-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (1.3 MB view details)

Uploaded Aug 5, 2026 CPython 3.13manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

fue-0.1.9-cp313-cp313-macosx_14_0_arm64.whl (1.0 MB view details)

Uploaded Aug 5, 2026 CPython 3.13macOS 14.0+ ARM64

fue-0.1.9-cp312-cp312-win_amd64.whl (147.9 kB view details)

Uploaded Aug 5, 2026 CPython 3.12Windows x86-64

fue-0.1.9-cp312-cp312-musllinux_1_2_x86_64.whl (1.2 MB view details)

Uploaded Aug 5, 2026 CPython 3.12musllinux: musl 1.2+ x86-64

fue-0.1.9-cp312-cp312-musllinux_1_2_aarch64.whl (1.1 MB view details)

Uploaded Aug 5, 2026 CPython 3.12musllinux: musl 1.2+ ARM64

fue-0.1.9-cp312-cp312-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (1.4 MB view details)

Uploaded Aug 5, 2026 CPython 3.12manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

fue-0.1.9-cp312-cp312-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (1.3 MB view details)

Uploaded Aug 5, 2026 CPython 3.12manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

fue-0.1.9-cp312-cp312-macosx_14_0_arm64.whl (1.0 MB view details)

Uploaded Aug 5, 2026 CPython 3.12macOS 14.0+ ARM64

fue-0.1.9-cp311-cp311-win_amd64.whl (147.9 kB view details)

Uploaded Aug 5, 2026 CPython 3.11Windows x86-64

fue-0.1.9-cp311-cp311-musllinux_1_2_x86_64.whl (1.2 MB view details)

Uploaded Aug 5, 2026 CPython 3.11musllinux: musl 1.2+ x86-64

fue-0.1.9-cp311-cp311-musllinux_1_2_aarch64.whl (1.1 MB view details)

Uploaded Aug 5, 2026 CPython 3.11musllinux: musl 1.2+ ARM64

fue-0.1.9-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (1.4 MB view details)

Uploaded Aug 5, 2026 CPython 3.11manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

fue-0.1.9-cp311-cp311-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (1.3 MB view details)

Uploaded Aug 5, 2026 CPython 3.11manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

fue-0.1.9-cp311-cp311-macosx_14_0_arm64.whl (1.0 MB view details)

Uploaded Aug 5, 2026 CPython 3.11macOS 14.0+ ARM64

fue-0.1.9-cp310-cp310-win_amd64.whl (147.9 kB view details)

Uploaded Aug 5, 2026 CPython 3.10Windows x86-64

fue-0.1.9-cp310-cp310-musllinux_1_2_x86_64.whl (1.2 MB view details)

Uploaded Aug 5, 2026 CPython 3.10musllinux: musl 1.2+ x86-64

fue-0.1.9-cp310-cp310-musllinux_1_2_aarch64.whl (1.1 MB view details)

Uploaded Aug 5, 2026 CPython 3.10musllinux: musl 1.2+ ARM64

fue-0.1.9-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (1.4 MB view details)

Uploaded Aug 5, 2026 CPython 3.10manylinux: glibc 2.17+ x86-64manylinux: glibc 2.28+ x86-64

fue-0.1.9-cp310-cp310-manylinux2014_aarch64.manylinux_2_17_aarch64.manylinux_2_28_aarch64.whl (1.3 MB view details)

Uploaded Aug 5, 2026 CPython 3.10manylinux: glibc 2.17+ ARM64manylinux: glibc 2.28+ ARM64

fue-0.1.9-cp310-cp310-macosx_14_0_arm64.whl (1.0 MB view details)

Uploaded Aug 5, 2026 CPython 3.10macOS 14.0+ ARM64

File details

Details for the file fue-0.1.9.tar.gz.

File metadata

Download URL: fue-0.1.9.tar.gz
Upload date: Aug 5, 2026
Size: 198.6 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/7.0.0 CPython/3.13.14

File hashes

Hashes for fue-0.1.9.tar.gz
Algorithm	Hash digest
SHA256	`0aaa7dcbb4b009bdb366d84b3a225b31656833141c68fc9858b28f317a8d08dc`
MD5	`376138358825a93e12a862c84f2ef0e5`
BLAKE2b-256	`27df3aa224b6c3aec1e986bd030051a771d43adc04c29bd0c6a1a88e031f977e`

Algorithm	Hash digest
SHA256	`a35133838ca28c945bf98602cb8c1ba1fb76d06062521e2ff19e3cd5fb7368dd`
MD5	`252135f0229bcd85b664d76a229de7bf`
BLAKE2b-256	`8506b5b0d16bb54a64982adee02f87a7ef168b56cc7f5b978f92df07e68b2082`

Algorithm	Hash digest
SHA256	`bf961d234dc123810376cb661e3c7e3abf1500de01b589de4b1af3844ae77788`
MD5	`22ef19bab81ec4d93bf2c34e333b5b12`
BLAKE2b-256	`ee072b91c5a619d7bfd21496724e0364ff2b5bec483d3d0ebda8d89b4d20cf9c`

Algorithm	Hash digest
SHA256	`d3b08960ae66c0d992644505506ef891e5453e5b9b0d58fcc89ebede48d152b5`
MD5	`d9495e3db8a672b385a403fc640c0940`
BLAKE2b-256	`9d5befe61627cff40e1534f96fa246d1002ee2601406393de9dc2f013aeb2b4b`

Algorithm	Hash digest
SHA256	`206bd0f5795a3eeb31986add911f973213bc22be080854516e486fb86da5527e`
MD5	`2164573e01682bf440045c13c604ebc7`
BLAKE2b-256	`626f2d1011f64e7164a7f07b87ce659f4af74dd0a4ee0bb3b673ae0b41baa347`

Algorithm	Hash digest
SHA256	`fb56cfb3684ce98a7321cee34172854acdbf1d55d7239a8a61ffdf95bf327e52`
MD5	`ee2d00a4e6b0a36aa1ea855dfac9f7d9`
BLAKE2b-256	`acc275119d71317da4b90cf49b7416143cb477b5afb715663a139a7f21e404b3`

Algorithm	Hash digest
SHA256	`f5ffd4c49fd0e83b31673d07fefb401708a1d981c77bca40b3dc56902879cbe3`
MD5	`b2d886267ba11acb249f1967e6bfce62`
BLAKE2b-256	`5fc163b12c5508b71ef644f25f9718a9398ba0d138ca80d8a4585671c21c26cc`

Algorithm	Hash digest
SHA256	`b8f7e5767fd207478e1886a4eb169f3b58b726bc2bc6dff35d40b5526822c40d`
MD5	`d54b797802b28cda558e458bbd55e488`
BLAKE2b-256	`dbd1d1c96363c84569fbf97d2d3c0fcb71690f53b1727b3caec64c953cd49257`

Algorithm	Hash digest
SHA256	`5cae42d8614b983cfad1a773b4ecdab6cfad7f0588b07bcbf12cd186d1552300`
MD5	`2d81c027c8075d2bbfbea2d2154ab6ed`
BLAKE2b-256	`2edaf8fb5ae7587ec8869b6fb4d32a47cace566d62c99e251852613f41779309`

Algorithm	Hash digest
SHA256	`244583ca3b247b56b767df9beda277942ad553ccdea14414dff98ba08679d0f7`
MD5	`7c65a6dfb1a86253d14def70b7e94a1e`
BLAKE2b-256	`d59e512c82c4e319eae509273f4e28aa88830037576701ef140743ee101dbae0`

Algorithm	Hash digest
SHA256	`322287a07cfab4f38f0392bdabaceb714feadb53b7173b2fc21184459a61661d`
MD5	`1688437d52c6f29830e60d49a969bcef`
BLAKE2b-256	`522ea4a44a07f1c852297de3fccd5ed363b86fb004b3ede63c4e367f7ca9a47c`

Algorithm	Hash digest
SHA256	`061e84202eb758301e85eefeec241c22c957873c4dbf5bf2ed31626c4f61d1c0`
MD5	`015911bec2d6dfc60a6ce82c0e2cba30`
BLAKE2b-256	`5364eab977dd40d2fa65d9b5ef28d693e4f9dfb8ea5091810e224496e0559865`

Algorithm	Hash digest
SHA256	`1b306a0ec8e61608e746235ffaed2d3fcc87c5977fd9df03a724ce911962fd24`
MD5	`f6c7a30d744b91520be258c7409a76c0`
BLAKE2b-256	`aa4d01d9f93d7f5341dc45f1b18350c7685568b35bcd0fdac2ed2b72374d6c7f`

Algorithm	Hash digest
SHA256	`c8c22e7da1c7cf4c32a222382d976ce8736477f918c5fd376c7597c34f493aed`
MD5	`9e4886b135caf3055ce81f502e824cf2`
BLAKE2b-256	`e9dfcfe3ef65dc587b345ea671c560167142031e9e1d8c526b05495e1335d0a6`

Algorithm	Hash digest
SHA256	`bd833c371c02d3e6a0664e69b82046f38b3a2d7ab3140010fd14a3bf8db94903`
MD5	`5dba50fda2731209c03b1ffb5d6a889d`
BLAKE2b-256	`477115231ed36f3e625664972c30b9eedb1b3aba31527b2d8c805428212f492f`

Algorithm	Hash digest
SHA256	`e85aa350589437e554f3780b55db754fe05b518378a8a679c98fd3c8effd4326`
MD5	`2fc913d900fd2c87b6eef9ef237945f3`
BLAKE2b-256	`076dab8740ee14210399ca230b94d0d0d553ee048cb5c269a7e49514e82847b5`

Algorithm	Hash digest
SHA256	`933f62f6312c11c24c7634e7e41b3d7778aa20bc27c8582c7482c1f41bd1946a`
MD5	`501f43efc20fbd887247957a7a74c616`
BLAKE2b-256	`66a8d9189cb59d608b2d19111fe448bab927e21bf1c7ab40bee21c3d6db4f7b2`

Algorithm	Hash digest
SHA256	`3d1c2184d13980a66592f3fd06ef0fd72a5f1aec96bf34cc1e9f4b858c97039c`
MD5	`672ea887673947a11b5e1ece23620b0a`
BLAKE2b-256	`be90604fdc5360781e66bd80e73e48d974ac56db46eb0cfaa651d494064c2d60`

Algorithm	Hash digest
SHA256	`ccf4024d9a7b5f9dbdb6f7143965ea1a1ac9b2fe64200d00db9804c3b77956a3`
MD5	`932a8c2d144ef8d7b09839b268738b78`
BLAKE2b-256	`764dcc6e4ede590829d12b2d2180aa284333b19b8d1f4274e0b1d4f90be87e20`

Algorithm	Hash digest
SHA256	`2fb80d013fdf3b80369cc3fca00d0dc0aa5f7ac12593f4e54d9b0340ac6afee7`
MD5	`fe23b358e9146ca2e87b5bf20e7098fa`
BLAKE2b-256	`4114a969cbd375b61c9b45ce62c6c2b0ee51005e675a132da3621d10fb4ec535`

Algorithm	Hash digest
SHA256	`b3bea1474bcfef59bedb32ff8089aabd69ef577aab36f233b750a01cb411436b`
MD5	`50fc86d0ec6c439f6431ddb13b192146`
BLAKE2b-256	`17a1db29d97bf00f1591590af00c9f52cd06eafbf199c5efbcd681d27a51dc06`

fue 0.1.9

fue — Exact Maximum Likelihood for Univariate Time Series

Features

Installation

Quick start

Load from a .inp file (fue format)

Forecast from a pre-estimated .inp file (fuf format)

Command-line interface

Numerical methods

Bug tracking

Authors and licence

Key dates

1 maintainer

Credits

License

Requires

Provides Extra

Tags

Classifiers

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

Load from a `.inp` file (fue format)

Forecast from a pre-estimated `.inp` file (fuf format)

Algorithm	Hash digest
SHA256	`ed5f9b220a7331c27641ceb75e62e74ef31eb2e20a8f4d4f62ec09075659518d`
MD5	`a81dc6b662b9e2bfc054d404db82bb48`
BLAKE2b-256	`a0010d2a7f96dc4cb64ba7a51d313d8c6093562479dc5a651b30d0f428a28cab`

Algorithm	Hash digest
SHA256	`8bddcbf8b42a36dce5c0040b1cbe316fd48cfd88033e7e4bcfa7d7564a5363d4`
MD5	`faef756431a8a7a63e929ef32d865066`
BLAKE2b-256	`7577c3ec10bba68a77e4b337ced5a12ee245525a029c4f5b29586ce9f5b13ae4`

Algorithm	Hash digest
SHA256	`852e7004ad669f220d697322922f5258d629e67d485fd706d0594c5e7ee531ae`
MD5	`70051382d19a98e49acec23a2281ecff`
BLAKE2b-256	`34572312120215cd0f51662cf45509e5089be876f4afda48cb28ea6656b1634d`

Algorithm	Hash digest
SHA256	`f9c48b0dbb42620799f000cde38c9beeed058efecce1de12a8566dac6a68d70d`
MD5	`9c1c8184e4a35a098eaa7a38c36eaf41`
BLAKE2b-256	`6406be6a22a3a78eaaaadd59922f2947db169851775c0f4960d3205e88cb2a7e`

Algorithm	Hash digest
SHA256	`dcffab5de632af2ba73434e9b26a9314abada2a9a312d15d7ca33ed74433d2b2`
MD5	`bde4b7c3a7d6b2256a6ee60c063cf42a`
BLAKE2b-256	`386443e2b28eda2e934c18f0dd37198788a5ab53bc72edd5d480e69ba5f671b2`