0

I know that there are two ways to do backtesting, sliding window and expanding window.

In practice, which method is better? What are the pros and cons of each method?

In my opinion, I guess if the time series pattern is more related to the current event, then the sliding method is better.

Sliding window as below figure

enter image description here

Expanding window as below figure

enter image description here

Source: https://www.kaggle.com/cworsnup/backtesting-cross-validation-for-timeseries/notebook

Cena
  • 3,316
  • 2
  • 17
  • 34
Luke
  • 15
  • 2
  • 7

1 Answers1

1

Which is better? Simplify answer is its depend on your Time Series data. Sliding Window and Expanding Window both have their use cases.

When you come up to testing high frequency data such as daily and hourly time series its better to use Sliding Window backtesting approach. Otherwise, if your historical points of Time Series are limited such as weekly, monthly or quarterly the Expanding Window form better.

Reference: Omphalos

Roy
  • 11
  • 1
  • 2