Tag: time series
-
Europe May Be Quietly Winning a Different AI Race: Forecasting the Future
While the world watches the chatbot race, European teams are building powerful AI models for a less glamorous and remarkably important problem: forecasting what happens next.
-
Your Demand Chart Hid a 132-Unit Problem. The FT’s Chart Rules Catch It
Four datasets can share the same summary statistics and still tell four different stories. So can your SKU history. Here is how to name the relationship in your data, pick the chart that fits it, and spot the trend break your eye keeps missing.
-
Six Times They Tried to Prove Sophistication Wins
Forty years of open forecasting competitions, and the winner’s margin over a dumb average stayed thin. Then M6 asked whether accuracy makes money. The correlation was 0.04.
-
I Raced Amazon’s Forecasting AI Against a 66-Year-Old Formula. The Formula Won.
I put Amazon’s zero-shot forecasting AI against five classical models on 148 real retail series. A formula from 1960 beat a 205-million-parameter model by 14 to 16%.
-
Taking the Engine Apart: Time Series Decomposition for Supply Chain Forecasters
Every time series is a cocktail of trend, seasonality, and noise. Decomposition is how you separate the ingredients — and once you can see each one, choosing the right forecast model stops being a guessing game.
-
Your Line Chart Is Hiding 8 Patterns: How to Find Them with fpp3
Four datasets. Identical statistics. Completely different shapes. If you’re not plotting your demand data before forecasting it, you’re flying blind — and fpp3 gives you the visual toolkit to see what your spreadsheet hides.
-
Stop Forecasting in Excel: Why R Is the Only Serious Tool for Supply Chain Demand Planning
Excel can’t do seasonality, model comparison, or prediction intervals without heroic effort — R does all three in six lines of code. Here’s why April is the month you finally make the switch.
-
FPP3: Stop Guessing Which Forecast Model Works — Measure It
Rob Hyndman’s fpp3 ecosystem lets you fit, compare, and evaluate multiple forecasting models in three lines of R code — here’s why supply chain teams should stop fighting Excel and start using a real framework.