timeseries-eda 0.1.3

Why tseda?

Existing libraries solve individual problems. No single package provides all of:

• Comprehensive EDA & data auditing

• Forecastability assessment

• Automated diagnostics (stationarity, seasonality, anomalies)

• Structural break / changepoint detection

• Feature engineering

• Model recommendations

• Interactive reports

tseda fills that gap, using only numpy, pandas, scipy, and matplotlib as core dependencies.

Installation

pip install timeseries-eda

For stationarity tests that use statsmodels (ADF, KPSS, Phillips-Perron):

pip install timeseries-eda[stats]

For building the documentation:

pip install timeseries-eda[docs]

Quick Start

import numpy as np
import pandas as pd
from tseda import TimeSeries

# Build a TimeSeries object
idx = pd.date_range("2020-01-01", periods=365, freq="D")
ts  = TimeSeries(
    np.cumsum(np.random.randn(365)),
    index=idx,
    name="stock_price",
    unit="USD",
)
print(ts)

# Data quality
from tseda.quality import MissingValueAnalyzer, OutlierDetector
missing = MissingValueAnalyzer().analyze(ts)
outliers = OutlierDetector().mad(ts)

# Statistics
from tseda.statistics import DescriptiveAnalyzer, StationarityTester
stats = DescriptiveAnalyzer().analyze(ts)
adf   = StationarityTester().adf(ts)
print(adf.summary() if hasattr(adf, "summary") else adf)

# Decomposition
from tseda.decomposition import STLDecomposer
dec = STLDecomposer().decompose(ts, period=7)
print(dec.summary())

# Seasonality
from tseda.seasonality import SeasonalityDetector
season = SeasonalityDetector().detect(ts)
print(f"Dominant period: {season.dominant_period}")

Modules

Dependencies

Core (always installed):

• numpy >= 1.23

• pandas >= 1.5

• scipy >= 1.9

• matplotlib >= 3.6

Optional:

• statsmodels >= 0.14 — ADF, KPSS, Phillips-Perron, STL (pip install timeseries-eda[stats])

Documentation

https://amir-jafari.github.io/Time-Series-EDA

Contributing

Contributions are welcome! Please open an issue or pull request at https://github.com/amir-jafari/Time-Series-EDA.

License

MIT © 2026 Amirhossein Jafari