4 ms·
When I got my statistics degree, a significant amount of time was spent in early classes demonstrating how non-intuitive statistical thinking can be. This, of c
by madrox 6y ago
When I got my statistics degree, a significant amount of time was spent in early classes demonstrating how non-intuitive statistical thinking can be. This, of course, is probably the most memorable example. Another fun one was a TED talk about Peter Donnelly on how juries get fooled that included a bunch of similar examples: https://www.ted.com/talks/peter_donnelly_how_juries_are_fooled_by_statistics?language=en https://www.ted.com/talks/peter_donnelly_how_juries_are_fool...
- atum47 6y agowell, I made this experiment cause I didn't believed that you would have more chance switching doors. Couldn't wrap my head around the idea...
- madrox 6y agoTotally! Simulation is a great way of learning this stuff. I spent a lot of time in R doing similar things over the years.
- Frost1x 6y agoThe Monty Hall problem supposedly even confused Paul Erdos (famous mathematician) who even rejected it after given a statistical proof. He didn't accept the answer until stepping through a computer simulation so if it seems unintuitive, you're in good company. This problem is, in lore, claimed to trip up more nobel laureates then most problems because it's so counter intuitive to not notice the conditional probability and injected information from Monty.
- atum47 6y agothat's really interesting
- abiogenesis 6y agoTry increasing the number of doors and your brain will (slowly) accept it. "One million doors" as described in another comment [1] is a good way to visualise it. [1] https://news.ycombinator.com/item?id=23161993 https://news.ycombinator.com/item?id=23161993