MUMBAI, India, Oct. 26 -- Intellectual Property India has published a patent application (202541083761 A) filed by Manipal Academy Of Higher Education on December 17, 2025, for A System And Method For Forecasting A Stock Index Using Rolling Data Segment.
Inventors include Satyaban Sahoo; and Deepti Singh.
The application for the patent was published on October 23, 2026, under issue no. 43/2026.
Abstract: A stock index forecasting system (100) comprises one or more sensors (209) receiving, from an external data source, time series data of a stock index, a data acquisition unit (207) acquiring the data, a memory (203) storing the data and a forecast value, a communication interface (205) transmitting the forecast value to a remote device (103), and a processor (201) coupled thereto. The processor (201) extracts a fixed-length rolling data segment, updating it by replacing an oldest observation with a newest observation, and calibrates a segment-specific differencing order using three stationarity tests, accepting the order upon majority agreement or upon reaching a maximum order regardless of outcome. The processor (201) selects a parameter configuration having a minimum information criterion value, generates a forecast value, and repeats these operations iteratively while maintaining a constant memory (203) footprint independent of the time series data's total length. (FIG.2)
Disclaimer: Curated by HT Syndication.
