Definition
The National Hurricane Center’s forecasting of Hurricane Katrina in 2005. Their predictions, made days in advance, allowed for evacuation orders that saved many lives, despite the devastating impact of the storm.
Example
The National Hurricane Center’s forecasting of Hurricane Katrina in 2005. Their predictions, made days in advance, allowed for evacuation orders that saved many lives, despite the devastating impact of the storm.
Ask yourself
- What data and trends are relevant to the situation I’m trying to forecast?
- How far into the future am I attempting to forecast, and how does this affect my methodology?
- What assumptions am I making in my forecast, and how might they be challenged?
- How can I be sure that I am not being overly reductive or simplistic in my assessment?
- How will I measure the accuracy of my forecast over time?
Tools
- Time series analysis
- Delphi method
- Scenario planning
- Trend extrapolation
- Cross-impact analysis
- Systems dynamics modeling
- Machine learning and AI predictive models
Forecasting vs. Prediction
- Forecasting: Typically deals with specific variables or trends, often in a quantifiable manner.
- Prediction: Can be broader, sometimes claiming to foresee exact future states or events.
- Forecasting: Explicitly acknowledges uncertainty, often providing ranges or probabilities.
- Prediction: May imply a higher degree of certainty about future outcomes.
- Forecasting: Often focuses on shorter to medium-term futures.
- Prediction: Can span any timeframe, including long-term futures.
- Forecasting: Uses structured methods and models based on data and trends.
- Prediction: Can be based on various methods, including intuition or speculation.
- Forecasting: Regularly updated as new information becomes available.
- Prediction: Often presented as a fixed statement about the future.
Other posts in this series
