Session 5 : The Rational War : When can you trust your intuitions? and your reasoning? Flashcards
What is the main critique of traditional rational choice models in decision-making ?
Rational choice models, which focus on maximizing expected utility, are unrealistic in situations of uncertainty because individuals often don’t have access to complete information about alternatives or probabilities.
Example: The 2008 financial crisis showed the failure of rational choice models, as investors couldn’t predict market behavior in unpredictable conditions.
N.B : In uncertain environments, decision-makers need alternative strategies to rational choice models, which are not practical without full knowledge of outcomes and probabilities.
What is the difference between risk and uncertainty?
Risk occurs when probabilities are known, while uncertainty arises when probabilities are unknown or unknowable.
Example: Investing in a well-established company involves risk (known probabilities based on data), while investing in a startup involves uncertainty (no clear data or probabilities).
N.B : Most real-world decisions are made under uncertainty, not risk. Optimization models, which work well for risk, often fail in uncertain situations.
What are heuristics and why are they useful in decision-making under uncertainty ?
Heuristics are simple mental shortcuts or rules of thumb that help people make decisions efficiently, especially when time or information is limited.
Example: Doctors in emergency rooms use quick decision-making rules (heuristics) like asking key questions to diagnose a heart attack without analyzing every detail.
N.B : Heuristics are adaptive and often outperform complex models in uncertain environments by focusing on the most relevant information and filtering out noise.
Why is simplicity often better than complexity in decision-making ?
Simpler models are often more robust because they generalize better to new situations, while complex models risk overfitting past data and failing in new conditions.
Example: The “buy and hold” investment strategy outperforms more complex trading strategies because it focuses on long-term gains without frequent adjustments.
N.B : Simple solutions can handle uncertainty better by being less prone to overfitting and are often more practical and reliable.
What is the bias-variance tradeoff in decision-making models ?
Simpler models may introduce bias (they don’t fit the past data perfectly), but they have lower variance (they are less sensitive to changes and perform better in new situations). Complex models may fit the data better but can overfit and fail to generalize.
Example: In machine learning, simpler algorithms like linear regression often outperform deep neural networks in real-world applications, despite being less accurate in the training phase.
N.B : In uncertain environments, simpler models with low variance often provide more stable and reliable predictions than overly complex models.
How can heuristics be applied effectively in fields like medicine, finance, and policy-making ?
Heuristics can guide decision-makers in fields like medicine (emergency diagnostics), finance (portfolio management with the 1/N heuristic), and policy-making (quick decisions during crises like COVID-19).
Example: In finance, the 1/N heuristic, where money is split evenly across N assets, performs as well as more complex models that try to predict returns and risks.
N.B : In high-stakes environments where data is incomplete or time is limited, using simple heuristics can lead to better outcomes than complex, data-heavy models.
How did the 2008 financial crisis expose the limits of traditional decision-making models ?
Traditional economic models failed to predict or manage the crisis because they relied on assumptions of known risks, but the reality involved deep uncertainty, where probabilities and outcomes were unknown.
N.B : The crisis demonstrated the need for decision-making approaches that work under uncertainty, where complete knowledge is not available.
How are heuristics used in medical decision-making ?
Doctors use heuristics in emergency situations to make quick decisions. For example, they follow simple decision trees to diagnose heart attacks based on key symptoms rather than analyzing all possible causes.
N.B : Heuristics help doctors save time and lives when full analysis isn’t possible, proving the effectiveness of simple, fast rules in uncertain, high-stakes situations.
What is an example of simplicity outperforming complexity in finance ?
The “buy and hold” investment strategy is a simple approach where investors buy stocks and hold them for the long term, outperforming more complex trading strategies that involve frequent buying and selling.
N.B : Simpler strategies like “buy and hold” work well because they avoid the pitfalls of overcomplicating decisions, especially in unpredictable markets.
How did heuristics help in public policy during the COVID-19 pandemic ?
Governments used simple rules, such as “lockdown now, reassess later” or “prioritize healthcare workers for vaccines,” to make rapid decisions under uncertainty when data was incomplete or constantly changing.
N.B : Heuristics allowed policymakers to act quickly and effectively in an uncertain crisis where waiting for complete data could have had disastrous consequences.
Why is the bias-variance tradeoff important in making predictions ?
Models with high variance are sensitive to new data and tend to overfit past patterns, while biased models (like simpler heuristics) are more stable and perform better in real-world situations where new and unexpected variables come into play.
Example: A complex stock-predicting algorithm might perform well on past data but fail in future market conditions, whereas a simple heuristic-based approach would yield more consistent results.
N.B : When making predictions, it’s essential to find the right balance between bias and variance to ensure models are adaptable to new data and conditions.
What is the main takeaway from Gerd Gigerenzer’s views on decision-making under uncertainty ?
In uncertain environments, simpler, heuristic-based decision-making processes often outperform complex optimization models, providing more robust, reliable results in real-world situations.
N.B : Gigerenzer’s focus on simplicity and the use of heuristics challenges traditional economic and decision-making models, which assume that more complexity equals better outcomes.
