I've read the article about the number of observations necessary to detect a favorable number (bias)
the article says that
"Based on the work of Ethier (1982), Keren and Lewis (1994) report that the number of observations necessary to detect a favorable number (bias) is generally quite large. For example, on a wheel with 37 numbers it would be necessary to view 30,195 spins in order to detect a bias of 1/33 with a 90% level of certainty. "
How to calculate it to get 30,195 times. I have no idea. Can someone explain it for me in details? Thank you
the article says that
"Based on the work of Ethier (1982), Keren and Lewis (1994) report that the number of observations necessary to detect a favorable number (bias) is generally quite large. For example, on a wheel with 37 numbers it would be necessary to view 30,195 spins in order to detect a bias of 1/33 with a 90% level of certainty. "
How to calculate it to get 30,195 times. I have no idea. Can someone explain it for me in details? Thank you