quantresearch-thd 0.1.26

Ensemble framework for detecting outliers in grouped time-series data

Quantresearch_thd

Quantresearch_thd is an ensemble framework for detecting outliers in grouped time-series data. It automates the entire workflow from data cleaning and calendar interpolation to running 8 different detection algorithms and generating visual diagnostic reports.

Key Capabilities

• Ensemble Scoring: Combines 8 models (Statistical + ML) to provide a robust Anomaly_Score and a final is_Anomaly consensus.
• Hierarchical Processing: Natively handles grouped data (e.g., detecting anomalies per Region, Product, or Channel).
• Automated Preprocessing: Handles missing dates via linear interpolation and filters out "low-quality" unique_ids automatically.
• Parallel Execution: Leverages joblib for multi-core processing of large datasets.
• Visual Analytics: Generates pie charts, stacked bar plots, and detailed group-level time-series breakdowns.

Included Models

The pipeline utilizes an ensemble of the following methodologies:

• Statistical: Percentile (5th/95th), Standard Deviation (SD), Median Absolute Deviation (MAD), and Interquartile Range (IQR).

• Time-Series Specific: EWMA (Exponentially Weighted Moving Average) and FB Prophet (Walk-forward validation).

• Machine Learning: Isolation Forest (General & Time-series optimized) and DBSCAN.

Detailed Functionality

• Robust Input Validation: Clear error messaging for missing parameters or incorrect data types.

• Quality Control: Automatically generates a Success Report

• Visual Suite: Automated rendering of Pie Charts (Summary), Stacked Bars (Distribution), and Top-5 Anomaly Heatmaps.

🚀 Quick Start

!pip install quantresearch_thd
import pandas as pd
from quantresearch_thd import timeseries_anomaly_detection

 # Load your data
df = pd.read_csv("your_data.csv")

 # Run the pipeline
anomaly_df, success_report, exclusion_report = timeseries_anomaly_detection( master_data=df,
                                                                             unique_ids=['category', 'region'],
                                                                             variable='sales',
                                                                             date_column='timestamp',
                                                                             freq='W-MON',
                                                                             eval_period=1  # Evaluate the most recent recor
                                                                             )

📊 Visualizing Results & Deep Dives

Inspecting a Specific Group, if a specific group shows a high anomaly rate, use the evaluation_info tool to render detailed diagnostic plots.

from quantresearch_thd import evaluation_info

# Filter the specific group you want to inspect. Define the group values (must match the order in unique_ids)
group_values = ['appliances', 'TX'] 

# Filter the results for this group
mask = anomaly_df[unique_ids].eq(group_values).all(axis=1)
group_df = anomaly_df[mask]

# Generate detailed diagnostic plots
evaluation_info(group_df,
                unique_ids,
                variable,
                date_column,
                eval_period=1
                )

The Evaluation Dashboard provides:

• Model Breakdown: Individual charts for FB Prophet, EWMA, and Isolation Forest with confidence intervals.

• Ensemble View: A summary highlighting where multiple models overlap.

• Statistical Thresholds: Visual markers for IQR, MAD, percentile and SD limits.

Input_data:

Mandatory

• master_data (pd.DataFrame) : Name of your dataframe containing inputs to be evaluated for anomalies include variables, dates, and unique_ids.
• unique_ids (list[str]) : List of columns used to segment data ['SKU', 'channel', "store_id"].
• variable (str) : The numerical target column name to analyze for presence of anomalies.
• date_column (str) : The datetime column representing the time dimension ["date","week","month"].

Default

• freq (str) : Frequency of the date column. Default: 'W-MON'. Accepts 'D', or 'MS'.
• eval_period (int) : Number of trailing records or periods to evaluate for anomalies. Default: 1.
• max_records (int) : Max history to consider starting from the most recent date. Default: all history
• imputation_method : Technique to fill missing time units. Default: 'linear'. Acceptable values are : 'mean', 'mode', 'zero', 'linear'
• mad_threshold (int) : MAD parameter, controls Median Absolute Deviation sensitivity. Default: 2.
• mad_scale_factor (int) : MAD parameter, The scaling constant used to normalize the MAD. Default: 0.6745.
• alpha (float) : EWMA parameter, controls the smoothing factor for EWMA trend. Default: 0.3.
• sigma (float) : EWMA parameter, determines the standard deviation multiplier for upper and lower bounds. Default: 1.5.
• prophet_CI (float) : Prophet parameter, determines the confidence interval. Range 0 to 1, Default: 0.9.
• contamination (float) : Isolation Forest parameter, expected % of outliers (0 to 0.5). Default: 0.03.
• random_state (int) : Seed for model reproducibility. Default: 42.

📤 RETURNS

tuple [pd.DataFrame, pd.DataFrame, pd.DataFrame]:

• final_results : The main output, a dataframe that identifies anomalies with Anomaly_Votes and is_Anomaly.
• evaluation_report : Summary of interpolation %, record counts, and anomaly rates.

---

Output columns of final_results : All the output values are at "unique_ids" level.

MIN_value The minimum historical "variable" values. For train data the value is fixed. For test data varies. It is the min_value up to t-1.

---

MAX_value The maximum historical "variable" values. For train data the value is fixed. For test data varies. It is the max_value up to t-1.

---

Percentile_low / Percentile_high The 5th and 95th percentile "variable" values Used to detect unusually low or unusually high "variable" values. Fixed for train data. Varies for test data. Takes the stats by considering historical data upto t-1.

---

Percentile_anomaly Flags based on percentile limits: • Low → value < Percentile_low • High → value > Percentile_high • None → within the range

---

Mean / SD (Standard Deviation) The average "variable"and its standard deviation based on historical data.Fixed for train data. Varies for test data. Takes the stats by considering historical data upto t-1.

---

SD2_low / SD2_high Two-standard-deviation control limits: • SD2_low = mean − 2×SD (floored at 0) • SD2_high = mean + 2×SD

---

SD_anomaly Flags based on SD2 limits: • Low → value < SD2_low • High → value > SD2_high • None → within the range

---

Median / MAD (Median Absolute Deviation) Median of "variable" and the median of absolute deviations from the median.Fixed for train data. Varies for test data. Takes the stats by considering historical data upto t-1. Used for robust anomaly detection when data contains outliers.

---

MAD_low / MAD_high MAD-based limits: • MAD_low = median − 2 × MAD / 0.6745 (floored at 0) • MAD_high = median + 2 × MAD / 0.6745

---

MAD_anomaly Flags based on MAD limits: • Low → value < MAD_low • High → value > MAD_high • None → within the range

---

Q1 / Q3 / IQR (Interquartile Range) • Q1: 25th percentile • Q3: 75th percentile • IQR = Q3 − Q1 Used to detect unusually low or high "variable" values.

---

IQR_low / IQR_high IQR-based limits: • IQR_low = Q1 − 1.5 × IQR (floored at 0) • IQR_high = Q3 + 1.5 × IQR

---

IQR_anomaly Flags based on IQR limits: • Low → value < IQR_low • High → value > IQR_high • None → within the range

---

is_Percentile_anomaly / is_SD_anomaly / is_MAD_anomaly / is_IQR_anomaly Boolean indicators stating whether each method classified the value as an anomaly (low or high).

---

Alpha Smoothing factor used in EWMA. Higher values give more weight to recent observations.

---

EWMA_forecast Expected value estimated using the EWMA model.

---

EWMA_STD Rolling standard deviation of residuals around the EWMA forecast.

---

EWMA_high Upper anomaly threshold (EWMA_forecast + sigma × EWMA_STD).

---

EWMA_low lower anomaly threshold (EWMA_forecast - sigma × EWMA_STD).

---

Is_EWMA_anomaly Boolean flag indicating whether the observed value falls outside the EWMA bounds.

---

FB_forecast Expected value estimated using the EWMA model.

---

FB_low Lower confidence interval of the Prophet forecast

---

FB_high Upper confidence interval of the Prophet forecast.

---

FB_residual Difference between observed value and Prophet forecast.

---

FB_anomaly Raw anomaly indicator based on Prophet confidence bounds.

---

Is_FB_anomaly Boolean flag indicating a Prophet-detected anomaly.

---

isolation_forest_score Score from the Isolation Forest model indicating anomaly severity. Typical range: –0.5 to +0.5 • Higher scores = more normal • Lower scores = more anomalous

---

is_IsoForest_anomaly Boolean flag based on Isolation Forest model output: • True → model predicts anomaly (prediction = –1) • False → model predicts normal (prediction = 1)

---

dbscan_score Cluster label or distance score produced by DBSCAN (-1 indicates noise/anomaly).

---

is_DBSCAN_anomaly Boolean flag indicating DBSCAN-detected anomaly.

---

Anomaly_Votes Count of anomaly-detection methods that agree a point is anomalous. Ranges from 0 to 8.

---

is_Anomaly Final ensemble decision: • True → value flagged anomalous by 4 or more methods • False → fewer than 4 methods indicate anomaly