Let's say I make a prediction for a number of clouds per day which pass over an area. It comes with 95% confidence intervals [cloud low, cloud high]
Then let's say I make a prediction for the mean amount of water per day which passes over that area in the clouds. It comes with 95% confidence intervals [mean low, mean high]
To get the total water per day, I have to multiply the mean per day * the number of days.
Should I then multiply the confidence intervals [mean low * cloud low, mean high * cloud high], to get the confidence interval of the product?