Holiday Demand Forecasting using XGBoost

How DoorDash makes precision predictions with decision trees and spiky holidays
tech
data science
machine learning
time series
forecasting
internet article bookmark
Date

Friday February 2, 2024

Topics
tech
data science
machine learning
time series
forecasting
internet article bookmark

How DoorDash Improves Holiday Predictions via Cascade ML Approach - DoorDash Engineering Blog

  1. Transform your raw data, removing spikes
  2. Fit a trend like to the transformed data
  3. Predict tomorrow using transformed predicfions
  4. reverse The transformation to get expected numbers.

It’s kinda of the principle behind differencing: it’s making more stationary a time series.

Hi

_________________________

Bryan lives somewhere at the intersection of faith, fatherhood, and futurism and writes about tech, books, Christianity, gratitude, and whatever’s on his mind. If you liked reading, perhaps you’ll also like subscribing: