Forecasting
at Scale using
"Photon"

End-to-End Time Series Forecasting with Event & Market data.

Features included in photon

Preproceesing of Time Series Data. Removing outliters and addressing missing values. Generating Hierarchical Time Series

Performing Statistical Tests on Time Series Data - Auto correlation, ADF, KPSS

Integrating Event Data, Market data. Genearting complex features using time series signals.

Forecasting using Deep learning & classical Time Series techniques. Peforming accuracy metrics, sequential training & backtesting

Anicca's Photon enables time series forecasting
with strong mathematical insights for businesses to act

01

Photon's Data
Preprocessing Layer

Photon automatically ingests time series data using various integration points. A fully automated data quality check is carried on the TS Data. Any missing values are imputeded with advanced statistical methods. It then generates hierarchical time series depending on the configuration.

02

Statistical
Analysis

Photon runs advanced statistical tests for analyzing stationarity, seasonality & cyclic nature of Time Series Data. It provides inference to support model decision making process.

03

Advanced Feature
Engineering

Photon uses external data feeds such as event data, market data depending on the domain of forecasting. Additionally, It combines unstructured & structured information to generate features for model training.

04

Time Series
Modeling

Modeling using deep learning, machine learning & classical techniques for forecasting. The recipe combining different models include intelligent decision graphs.

05

Consumption
Layer

Output from the model include Forecast, uncertainity, accuracy measurement & backtest results. API's are provided for consuming the output & meta information

01

Photon's Data
Preprocessing Layer

Photon automatically ingests time series data using various integration points. A fully automated data quality check is carried on the TS Data. Any missing values are imputeded with advanced statistical methods. It then generates hierarchical time series depending on the configuration.

02

Statistical
Analysis

Photon runs advanced statistical tests for analyzing stationarity, seasonality & cyclic nature of Time Series Data. It provides inference to support model decision making process.

03

Advanced Feature
Engineering

Photon uses external data feeds such as event data, market data depending on the domain of forecasting. Additionally, It combines unstructured & structured information to generate features for model training.

04

Time Series
Modeling

Modeling using deep learning, machine learning & classical techniques for forecasting. The recipe combining different models include intelligent decision graphs.

05

Consumption
Layer

Output from the model include Forecast, uncertainity, accuracy measurement & backtest results. API's are provided for consuming the output & meta information

Explore how Photon helps businesses from different domains

Artificial General Intelligence Journey

Anicca Data addresses your data driven decision making needs using Deep Learning, Machine Learning, and Optimization Techniques.
Our science and engineering experts are eager to speak to you.