The accounting risk (AccR) is influenced by the reasonable ranges and how well they are calibrated. What are the steps involved to achieve their calibration?

Question

What will be an ideal response?

Answer 1

Calibration is achieved through repetition and feedback, which improves the quality of expert's estimates.
If the environment is predictable and an expert has learned through training to identify the situations
where their intuition misleads them, then the feedback will have corrected their intuitions. This the basis
of calibration training for assessing reasonable ranges of accounting estimates in new or changing
conditions. According to Hubbard (2009), calibration can be summarized in the following steps:
1. Repetition and feedback. Take several tests in succession, assessing how well you did after each one
and attempting to improve your performance in the next one. Continue to track performance after training
is complete. For example, an expert can be asked to estimate the return on the TSX over a stated period of
time. Two estimates can be asked: a value they are 95% sure is too high, and one they are 95% sure is too
low. The range between these two estimates is the "90% confidence interval" and outcomes outside this
interval can be viewed as "surprises". A well-calibrated expert who assesses such confidence intervals
multiple times can expect 10% of the outcomes to be surprises. As Kahneman notes however, managers
can be grossly overconfident in their ability to predict markets or even their own company's performance.
Actual surprises might be three times higher than estimated. This indicates that perhaps auditors should be
encouraged to develop their own reasonable ranges and get their ranges better calibrated through
estimation exercises involving repetition and feedback. For example, answer Trivial Pursuit type questions
involving ranges, or imagine an equivalent bet or critical-thinking exercise.
2. Equivalent bets. For each estimate, set up the equivalent bet to test whether that range or probability
really reflects your uncertainty.
3. Consider two pros and cons. Think of at least two reasons why you should be confident in your
assessment and two ways you could be wrong.
4. Avoid anchoring. Think of range questions as two separate binary questions of the form "Are you 95%
certain that the true value is over/under (pick one) the lower/upper (pick one) bound?"