Marian's Reviews > Thinking, Fast and Slow

Fantastic book, while is not easy to read it is quite rewarding in its true insights, scientific research and beautiful examples and great references to other books. Also, most important, the book is more valuable if time is allowed to think of practical examples

For example, Chapter 23 The Outside View, DK introduces the forecasting method of Flyvberg in overcoming the planning fallacy, with reference to infrastructure projects - in parenthesis my comments:
1. Identifying an appropriate reference class (is not complicated, doesn't need too much qualification)

2. Obtain statistics of the reference class
(this can be very complicated, very time consuming. Could be solved via improved data management and availability of statistics (specific providers).
I expect this exercise in public management to be very unlikely.
Some subjective obstacles (could work blended):
o Corruption: optimistic estimation (regarding costs and time needed) in order to obtain an approval. The decision factors could also be corrupt.
o Arrogance: can come from project manager and decision factors who want to push a project in order to show their worth, in order to achieve some material success for different soft reasons: reelection, keeping the job, social recognition, previous perceived success.
Solution: Usage of reference classes can be mandatory by law. Certain institution or department can be allocated the responsibility of providing quality reference class statistics.)

2'.Use the information to generate a baseline prediction
(mildly an obstacle. Beside using simple averaging, more sophisticated and potentially more precise methods could be avoided to be employed due to lack of knowledge and again time consuming by calling for expert help.
Solution: specific institution/department with responsibilities and knowledge. Baseline prediction)

3. Use specific information about the case to adjust baseline prediction
(Could be tricky if corruption or lack of willingness is present)