Interpolate Nan values in timetable

Data from the meteo with every day from 01/01/2022 to 31/12/2022 within 10mn of interval. What I'm trying to is to interpolate my missing values from meteo_full_data but based on the mean of previous {20,40,60 } days before and after the missing period. But I can't figure out how, this is what I got so far :

% start and end dates for the missing 20 days
start_date = datetime(2022, 6, 19, 21, 50, 0);
end_date = datetime(2022, 7, 8, 15, 50, 0);
% Calculate the means for the before and after periods
mean_before_60 = mean(data_before_60{:,:}, 'omitnan');
mean_before_40 = mean(data_before_40{:,:}, 'omitnan');
mean_before_20 = mean(data_before_20{:,:}, 'omitnan');
mean_after_20 = mean(data_after_20{:,:}, 'omitnan');
mean_after_40 = mean(data_after_40{:,:}, 'omitnan');
mean_after_60 = mean(data_after_60{:,:}, 'omitnan');

cols_to_interp = {'AR_Hum_', 'AR_Temp_C', 'GlobalRadiation5DegressKWh', 'GlobalRadiation45DegressKWh'};

mean_before = [mean_before_60; mean_before_40; mean_before_20];
mean_after = [mean_after_20; mean_after_40; mean_after_60];
mean_table= [mean_before;mean_after];
%mean_table = mean(mean_table, 'all')
data_to_interp = meteo_full_data(meteo_full_data.TimeStamp >= start_date & meteo_full_data.TimeStamp <= end_date, cols_to_interp)   

And I tried these two methods, no good :

inter_data = fillmissing(data_to_interp, 'constant', mean_table, 'DataVariables', cols_to_interp);
%inter_data = fillmissing(data_to_interp,"linear","DataVariables",cols_to_interp) 

and this one

%missing_rows = any(ismissing(meteo_full_data(:, cols_to_interp)), 2);
% Replace the missing values with the mean values
meteo_full_data(missing_rows, cols_to_interp) = num2cell(repmat(mean_table, sum(missing_rows), length(cols_to_interp)))

Answer 1

[Matt J]

You could use movmean, like in this example,

https://www.mathworks.com/help/matlab/ref/movmean.html#bu6jxup