demetrapy

demetrapy provides Python and command-line access to the seasonal-adjustment procedures in JDemetra+. It runs the JDemetra+ X13 and TRAMO/SEATS implementations through JPype; it does not reimplement either procedure in Python.

The package is intended for empirical work in which adjustment specifications must be recorded, repeated, and applied to several series. It accepts regular monthly, quarterly, half-yearly, and yearly observations. Calendar effects, intervention variables, outliers, ARIMA specifications, forecasts, and the principal decomposition options can be set in code or in a JSON file.

Seasonal adjustment is an inferential procedure, not merely a filter applied to a column of numbers. Results depend on the transformation, regression effects, ARIMA model, decomposition method, and span of the sample. Published series should therefore be accompanied by their specification and revision policy. demetrapy exposes JDemetra+ diagnostics and processing messages for this purpose, but it does not decide whether a specification is economically appropriate.

Installation

demetrapy requires Python 3.9 or later and Java 8 or later.

python -m pip install demetrapy

For development from a clone:

python -m venv .venv
source .venv/bin/activate
python -m pip install -e .

The first calculation downloads the pinned demetra-tstoolkit 2.2.6 JAR from Maven Central and stores it in ~/.cache/demetrapy. Set DEMETRAPY_JAR to the path of a local copy when automatic download is not suitable. The Windows guide covers Command Prompt, proxy-restricted, and offline installations.

Python interface

For a pandas object with a regular DatetimeIndex, adjust_dataframe() infers the observation frequency and adjusts each column separately:

import pandas as pd

from demetrapy import adjust_dataframe

data = pd.DataFrame(
	{"production": observations},
	index=pd.date_range("2015-01-01", periods=len(observations), freq="MS"),
)

result = adjust_dataframe(data, method="x13", spec="RSA4")
adjusted = result[("production", "sa")]

The lower-level adjust() function accepts one regular sequence and an explicit starting period:

from demetrapy import adjust

result = adjust(
	values,
	frequency="Quarterly",
	start_year=2005,
	start_period=1,
	method="tramoseats",
	spec="RSA4",
)
adjusted = result["sa"]

By default, both functions return the five compact components used in routine work:

Set detailed=True when the calculation must retain the full JDemetra+ result dictionary, diagnostics, processing messages, forecasts, backcasts, and fitted ARIMA model:

detailed = adjust(
	values,
	frequency="Monthly",
	start_year=2015,
	forecast_horizon=12,
	detailed=True,
)

print(detailed.arima_model.notation)
forecast = detailed.series["final.sa_f"]

Command line

The command-line interface reads a regular CSV file with date and value columns:

date,value
2019-01-01,101.2
2019-02-01,103.8

The default calculation is monthly X13 with the RSA4 preset:

demetrapy input.csv --output adjusted.csv

A JSON file records a fuller specification:

demetrapy \
  --data input.csv \
  --config examples/full_config.json \
  --output adjusted.csv

The output contains y, sa, t, s, and i, aligned with the input dates. See the usage guide for the complete command-line and Python API.

Specifications and regressors

The package constructs an isolated JDemetra+ processing context for each calculation. Preset defaults remain those of JDemetra+ unless an option is overridden explicitly.

UserDefined trading-day variables follow the distinction made in the JDemetra+ graphical interface: a calendar pool may contain several registered series while each target selects only the variables relevant to its own equation. The target observations and calendar pool may be supplied as separate DataFrames, provided their frequencies agree and the calendar domain covers the estimation sample.

The configuration reference documents processing order, valid option groups, preset behavior, and result semantics.

Inspection

A static summary plot can be produced from the command line:

demetrapy --data input.csv --plot-output adjustment.png

The local dashboard is included in the standard installation:

demetrapy-dashboard

It accepts observation, configuration, and calendar-pool files and reports the adjusted series together with diagnostics, model information, processing messages, and downloadable results. The dashboard is a convenient inspection tool; it uses the same calculation path as the Python and command-line interfaces.

Examples

Reproducibility and compatibility

The compact result schema is versioned, and the supported Python, Java, and JDemetra+ combinations are stated in COMPATIBILITY.md. Tests use both processing engines and include synthetic seasonal and calendar effects with known structure. CI runs on Linux, Windows, and macOS.

python -m unittest discover -s tests

demetrapy is an independent interface to JDemetra+ and is not an official publication of the JDemetra+ project.