Snowflake inc. (20240346386). TIME SERIES FORECASTING USING UNIVARIATE ENSEMBLE MODEL simplified abstract

From WikiPatents
Jump to navigation Jump to search

TIME SERIES FORECASTING USING UNIVARIATE ENSEMBLE MODEL

Organization Name

snowflake inc.

Inventor(s)

Michel Adar of Campbell CA (US)

Boxin Jiang of Sunnyvale CA (US)

Anh Quynh Kieu of Bellevue WA (US)

Boyu Wang of Menlo Park CA (US)

TIME SERIES FORECASTING USING UNIVARIATE ENSEMBLE MODEL - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240346386 titled 'TIME SERIES FORECASTING USING UNIVARIATE ENSEMBLE MODEL

    • Simplified Explanation:**

The patent application describes a time series forecasting algorithm that does not require hyperparameter tuning. It involves analyzing time series data using a quadratic function to predict a quadratic trend, removing the trend to generate detrended data, determining a moving median, descaling the data, and generating a forecast based on seasonal and trend predictions.

    • Key Features and Innovation:**
  • Fast and accurate time series forecasting algorithm
  • Eliminates the need for hyperparameter tuning
  • Utilizes a quadratic function to predict a quadratic trend in the data
  • Removes the trend to generate detrended data
  • Determines a moving median of the time series data
  • Descales the data using an amplitude scaling factor
  • Generates a forecast based on seasonal and trend predictions
    • Potential Applications:**
  • Financial forecasting
  • Demand forecasting
  • Stock market analysis
  • Weather prediction
  • Sales forecasting
    • Problems Solved:**
  • Eliminates the need for hyperparameter tuning in time series forecasting
  • Provides a fast and accurate forecasting algorithm
  • Improves the accuracy of predictions by considering trends and seasonality in the data
    • Benefits:**
  • Improved accuracy in time series forecasting
  • Faster forecasting process
  • Elimination of the need for manual hyperparameter tuning
  • Enhanced decision-making based on more reliable forecasts
    • Commercial Applications:**
  • "Fast and Accurate Time Series Forecasting Algorithm for Financial Markets"
  • This technology can be used by financial institutions, retail companies, and other industries that rely on forecasting for decision-making.
    • Questions about Time Series Forecasting:**

1. How does this algorithm compare to traditional time series forecasting methods? 2. What are the potential limitations of this algorithm in real-world applications?


Original Abstract Submitted

disclosed is a fast and accurate time series forecasting algorithm that eliminates the need for hyperparameter tuning. time series data may be analyzed using a quadratic function to determine a quadratic trend prediction, which is removed from the time series data to generate first detrended time series data. a moving median of the time series data is determined and the moving median is removed from the time series data to generate second detrended time series data. an amplitude scaling factor is determined based on the second detrended time series data and the first detrended time series data is descaled using the amplitude scaling factor to generate descaled time series data. the descaled time series data is analyzed to determine a seasonal prediction and a time series forecast is generated based on the seasonal prediction, the quadratic trend prediction, and the amplitude scaling factor.

Retrieved from "http://wikipatents.org/index.php?title=Snowflake_inc._(20240346386)._TIME_SERIES_FORECASTING_USING_UNIVARIATE_ENSEMBLE_MODEL_simplified_abstract&oldid=248963"