Hi everyone,

I’m working on a time series analysis project, and I’m facing a challenge with missing data. My dataset has several gaps due to missing records, but unlike simple random missingness, these gaps are not uniform and occur at different time intervals.

I’m unsure how to handle this situation. Should I:

Impute missing values based on neighboring values (e.g., linear interpolation, forward/backward filling)? Remove rows with missing data, even though they represent important time periods? Or, should I explore advanced techniques like multiple imputation or using models like ARIMA to predict missing values? I'm looking for advice on how to deal with these missing values in a way that won’t skew the time series trends, especially for forecasting. Also, are there any specific Python libraries or methods you’d recommend for this task? level devil Thanks in advance for your help!