Collaborating Authors

Forecast intervals for aggregates


A common problem is to forecast the aggregate of several time periods of data, using a model fitted to the disaggregated data. For example, you may have monthly data but wish to forecast the total for the next year. Or you may have weekly data, and want to forecast the total for the next four weeks. If the point forecasts are means, then adding them up will give a good estimate of the total. But prediction intervals are more tricky due to the correlations between forecast errors.

Hurricane path forecasts have much improved. Are they as good as they can get?

National Geographic

For something more complicated, like a tropical cyclone, the limit of predictability is much shorter. Tiny errors eventually grow so big that a forecast quickly becomes useless, no better at predicting the future than chance. Time-wise, forecasts stretching farther out than five or six days were long considered pie-in-the-sky goals. And the shorter-term forecasts can only get better if the initial errors get smaller.

Let shipping forecast compilations lull you to sleep


Hard Refresh is a soothing weekly column where we try to reset your brain and cleanse it of whatever terrible thing you just witnessed on Twitter. You've laughed through Vine compilations. You've cringed through TikTok compilations. Now, try sleeping through Shipping Forecast compilations. The Shipping Forecast is a nautical weather report on the BBC.

German think tanks trim 2016 growth forecast to 1.6 percent

U.S. News

Leading German economic think tanks have trimmed their 2016 growth forecast for the country's economy to 1.6 percent, pointing to slower expansion in China and elsewhere. Thursday's forecast by four economic institutes was down from the 1.8 percent they predicted in October. For 2017, they are forecasting growth of 1.5 percent. The German economy, Europe's biggest, expanded by 1.7 percent last year. It is traditionally export-heavy but lately has been fueled increasingly by domestic demand.