Which filling method is applied to fill missing values between the last recorded data point and the global end date of a dataset?

Back filling is the method that is applied to fill missing values between the last recorded data point and the global end date of a dataset. This approach involves using the last known value to fill in the subsequent missing values moving backward in time. It assumes that the most recent data point is the best estimate of what the missing values could be, hence maintaining a continuity in the data series.

In the context of time series data, when there's a gap starting from the last observed point and extending to the end of the dataset, using back filling ensures that the dataset remains usable for analysis or forecasting tasks. It effectively allows analysts and data scientists to continue their work without being hindered by gaps in the dataset that could otherwise lead to biases or inaccuracies in modeling.

Other filling methods, such as middle filling, are not specifically designed for this kind of scenario since they may involve averaging or using midpoints that could distort the observed trends. Future filling, which implies predicting future values based on existing data, is also not a suitable approach for filling up gaps past the last recorded data point, as it can introduce uncertainty regarding the validity of those predictions when the actual data has not been observed. Extrapolation would involve estimating unknown values by extending the trend beyond the last known data