# Automatic Model Building in GEFCom 2017 Qualifying Match

## TL;DR

This paper presents an automatic model building approach for time series forecasting, successfully applied in GEFCom 2017, demonstrating that automated decisions on trend variables can achieve competitive results.

## Contribution

It introduces the Tangent Information Modeller (TIM), an automated modeling strategy that includes trend decision automation, outperforming manual methods in a forecasting competition.

## Key findings

- TIM with temperature shuffling won GEFCom 2017
- Automated trend decision by TIM also achieved winning results
- The approach demonstrates effective automation in time series modeling

## Abstract

The Tangent Works team participated in GEFCom 2017 to test its automatic model building strategy for time series known as Tangent Information Modeller (TIM). Model building using TIM combined with historical temperature shuffling resulted in winning the competition. This strategy involved one remaining degree of freedom, a decision on using a trend variable. This paper describes our modelling efforts in the competition, and furthermore outlines a fully automated scenario where the decision on using the trend variable is handled by TIM. The results show that such a setup would also win the competition.