Poly%206-7.ppt - Poly Lecture 6 Data Mining Ethics Luck v/s...

Info iconThis preview shows pages 1–13. Sign up to view the full content.

View Full Document Right Arrow Icon
Poly Lecture 6 Data Mining Ethics
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
2 Luck v/s Skills: All So Easy
Background image of page 2
3 The Round Trip Fallacy Mistaking the necessary and the causal Successful Wear a suit Wear a suit Successful
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Nonsense?
Background image of page 4
How data can be bad for you Year 1: 1 million Year 2: 500,000 Year 3: 250,000 Year 4: 125,000 Year 5: 62,500 Year 6: 31,250 Year 11 about 1000 geniuses …just by luck (gamblers/investors/traders/monkeys/predictors)
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Ioannidis
Background image of page 6
Fitting explanations "He works long hours“ “He works short (but intense) hours” "He worked with X" "She comes early to the office" "She works late. ..The other day I called her at the office at 8 PM; she was there!" “He does yoga” "He follows a strict low-carb vegetarian diet"
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Artificial population attributes People in the experiment were successful by pure luck If the base population is large enough it is near impossible to NOT have a superstar The problem here is with the size of the base population This is called (among other names) the survivorship bias Little to do with the odds
Background image of page 8
Hell; It’s the Others (Sartre) The trader’s nightmare is not at all her own P/L It is one’s bad luck It is the good luck of the others Elementary simulation: under some conditions (ex ante “Sharpe” 1, maximum drawdown -20%) What is the 5-year survival rate? How does it compare to known diseases?
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Results Year 1 79% Year 2 60% Year 3 45% Year 4 35% Year 5 26%
Background image of page 10
Distribution You observe the Max {x1,…,xn} This matters more…
Background image of page 11

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Random Matrices 0.08 MAX . 36
Background image of page 12
Image of page 13
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 40

Poly%206-7.ppt - Poly Lecture 6 Data Mining Ethics Luck v/s...

This preview shows document pages 1 - 13. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online