A Guide on Dynamic Parameter Estimation for Causal Forecasting
Executives now-a-days rely on forecasting approcahes in virtually any decision making. The use cases are ubiquitous in business and technology domains ranging from Demand/Sales forecasting in Supply Chain Management, Hiring/Attrition rate forecasting in HR operations, Predicting the cell traffic or netwrok health parameters for a Telecom business, just to name a few. The challenges faced in forecasting in general, are mainly because of inherent non-linearity and non-stationarity in the time series. In this talk, I will try to walk through a systematic approach for parameter estimation in forecasting of time series data with causal factors available in the light of statistical estimation theory.
- Forecasting Overview : Time Series based and Causal Forecasting
- Linear Estimation :
i. Deterministic Estimation : Best Linear Unbiased Estimator and Sequential LSE
ii. Bayesian Estimation : Sequential LMMSE and Kalman Filter
- Nonlinear Estimation : Extended Kalman Filter and Sequential Neural Network
I did my PhD in Electrical Engineering from University of Texas at Arlington and am currently working as a Data Scientist in Ericsson Global AI team.