Thursday, November 15, 2018

p-value

The probability of getting data that shows a relationship  in an experiment when there is in fact no relationship.
Example. Coin a is not fixed=> the null hypothesis=no relationship between two measured phenomena. After flipping the coin 100 times, 66 heads show up. What is the probability of getting this data in a fair coin?
  \binom nk = \frac{n(n-1)\ldots(n-k+1)}{k(k-1)\dots1},
\frac{n!}{k!(n-k)!}
=> 100!/66!(34!)=580717429720889409486981450
p^n(1-p)100-n=0.5^66(0.5)^34=.5^100 since a^ma^n=a^m+n=0.0000000000000000000000000000007889
580717429720889409486981450 X 0.0000000000000000000000000000007889 =0.00045812798
OR the p-value is given the hypothesis is true, what is the probability that we would see a graph as such. If it is less than 5% chance over an assumed many repetitions in many drug and social science studies, we reject the hypothesis.

extinctions per year

There are two main lists used by scientists to keep track of the facts of extinction. -http://creo.amnh.org/creodata.html or https://www.iucnredlist.org/about/citationinfo Less than one species extinction per year. -https://wattsupwiththat.files.w
ordpress.com/2010/01/extinctions_birds_mammals_historical.jpg