@@phuongdh No. The prediction is probabilistic. You can't be wrong when you give a non-zero probability of something that happens. And to 538's credit, the probability they assigned to the actual outcome was considerably higher than many other polling and evidence based models, suggesting their model might be better, although not necessarily as it is only one data point.
I love these recorded lectures. A real breath of fresh air after 7 classes of Udacity :)
lecture starts at 2:00
Dyna at 42:00
king
2016 US election prediction at 25:33
lol how wrong they were
@@phuongdh No. The prediction is probabilistic. You can't be wrong when you give a non-zero probability of something that happens. And to 538's credit, the probability they assigned to the actual outcome was considerably higher than many other polling and evidence based models, suggesting their model might be better, although not necessarily as it is only one data point.
before the dark times, before the empire
I particularly looked for comments like this one 😂
@@andrewp3123 That just means they didn't do a thorough job with their model with such a terrible back testing result
Can you change the category to Education.