Nans values in time series

Hello...i have a time series data set with 2688 points....but i have 7 convective nans values. I want to do forecast with this data set....do i have to fill the gap or this is not necessary because the nan values will not affect my predictions? Thank you in advance

Answer 1

[Star Strider]

You must decide that.

There are methods to interpolate the NaN values if you want to do that, one of which is the fillmissing function (R2016b and later).

It would likely be best to do the forecast with and without the NaN values (some functions omit them automatically) and see if interpolating them affects the forecast.