Custom forecast frequencies now supported in Amazon Forecast

Today, [Amazon Forecast](/forecast/) is excited to announce the ability to customize the forecast frequency, enabling you to more closely align forecast intervals with your demand cycles. Amazon Forecast is a fully managed service that uses machine learning (ML) algorithms to deliver highly accurate time series forecasts. Previously, forecasting frequencies were fixed at specific (yearly, monthly, daily, hourly) intervals or minute (30, 15, 10, 5, 1) increments. These frequencies were not customizable - restricting you from forecasting at user-defined frequencies such as every two hours or every three weeks - and resulted in additional effort to manually aggregate forecasts to desired increments, increasing costs. With the launch of custom forecast frequencies, you can now specify your desired forecasting frequency and Amazon Forecast will automatically aggregate your data. This feature simplifies forecasting at intervals such quarterly (3-months) or bi-weekly frequencies, and allows you create forecasts that are better aligned with your business cycles. To customize your frequencies, follow the steps outlined in our [documentation](https://docs.aws.amazon.com/forecast/latest/dg/howitworks-predictor.html#creating-predictors), or our [GitHub notebook](https://github.com/aws-samples/amazon-forecast-samples/blob/main/notebooks/advanced/Custom%5FFrequency%5FForecast/Custom%5FFrequency%5FForecast.ipynb). This feature is available in all Regions where Forecast is publicly available. For more information about Region availability, see [AWS Regional Services](/about-aws/global-infrastructure/regional-product-services/).