Hypothesis Tests

 
 

Let’s recall some important stuff again, before we get into testing hypothesis part. I write this again and again in every pages because this is the part most students forget and just don’t get it when it comes to type I and II errors.


Another way to get it is do it my way if you want to, errors talk about null hypothesis, then, I take it if the hypothesis is true, you will keep it, and if you throw it away, you are making a mistake. That’s a type I. If it is not true, you can’t use it so you should not keep it, but if you stick to it, that’s type II.



α = P(rejecting H0 when it is true) = P(Type I Error)

α = Level of Significance = producer’s risk



β = P(not rejecting H0 when it is false) = P(Type II Error)

β = consumer’s risk

( 1 - β ) = the complement of β = the power of the test

When the value of β  is lower, the power of test is higher, and it is a better test.


It means that the less you make type II mistakes, the correct your decision will be and the test is great.

throw away there is no use

About H0

TURE

Not TRUE

it is true so throwing it away is a mistake Type I

it is not true so there is no use and you must throw it away

it is not true but you keep it then you are making a mistake Type II

you must keep it when it is true because it is useful

keep it, it is useful

DNA Pot (c) 2009 - Current

Review Concepts before testing: α, β, and power of test