What is the difference between forecasting and backcasting. What is the algorithm used for backcasting?
Backasting and Forecasting, both are used to estimate the future but the way both approaches to it is very different. In Backcasting, we manage to minimize the risk associated with future estimation.
Let’s look at the difference between these two:
Forecasting: We look at the past trend and present conditions then extrapolate the future values. In Forecasting, we simply do the following steps:
- Collect past and present data
- Taking future possibilities/ events in account for estimation
- Do extrapolation/ estimation for future
Backcasting: We move almost opposite to forecasting. In this technique we start with the point where we want to be in the future and then try to adjust everything in the present according to the target we want to achieve. If we summarize the steps in which backcasting is done then these can be as follows:
- Define and establish the targets for future
- Collect data from past and present condition/ events
- Analyze the collected data to finalize what changes and amendments are required to the present conditions so that ultimately it comes inline with the vision established.
- Establish a policy and strategy to implement the changes required
Hope this helps!
Regards,
Imran
1 Like
yes this is really helpful Imran, thanks a lot for this. Can you also add something on calculation side, I mean how am I suppose to improve my estimation using backcasting.
Well you would use the same algorithms as for forecasting.
Forecasting: You use observations 1-100 to predict 101
Backcasting: You use observations 100-2 to predict 1
Same algorithms just using the time series in reverse order.
Got it thanks a lot:)