How to Lie with Statistics

preview_player
Показать описание
Speaker- David Sigerson, Konstantin Greger We all love data, but sometimes it needs a little…help. Or so some people think. This session will delve into how data and visualizations are misused and how to detect when it is happening to you. Knowing the weaknesses in human perception allows you to amplify your results, making them seem more authoritative whilst removing alternative interpretations. In this session we’ll be discussing random correlations, bias in samples, ways to present data to support your conclusions, but will you use your new found knowledge for good or evil?
Рекомендации по теме
Комментарии
Автор

Fama's critique, for those interested:

" In Daniel Kahneman’s book...he states that our brains have two sides:
One is rational, and one is impulsive and irrational. What behavior can’t be explained by that model?"

This seems fairly fundamental. Rational choice theory, for all its problems, at least provides a formal structure for analyzing economic phenomena.

Behavioural econ, and the heuristics and biases program, doesn't do this. It basically offers post hoc explanations of deviations from rational choice. And that's all.

The main heuristics (availability, representativeness, anchoring and adjustment) are so broadly specified as to have basically no predictive power or explanatory power.

charlesshaw
Автор

Q sent me, in Bill gates video he had on his desk titled : How to lie with Statistics

imnaked