Communiquez avec les autres et partagez vos connaissances professionnelles

Inscrivez-vous ou connectez-vous pour rejoindre votre communauté professionnelle.

Suivre

How we can control type Ist and type 2nd errors?

user-image
Question ajoutée par QasiM sayyed , student , swatcom
Date de publication: 2015/04/12
ghulam abbas ghazi gajani
par ghulam abbas ghazi gajani , Research Assistant, Instructor Statistics, Economics and Mathematics , Lesbela University

Type I error is the margin of error you can bear (if you are going to purchase the spare parts used in aeroplane you need to assure these should be almost99% accurate the margin of error is reduced to1% rather than the parts used in agricultural machinery say you need to be sure that these should be95% accurate the margin of error is increased to5%). the Type II error is the power of the test that is related to the sample size as the sample size increase the estimates are close to the parametric values and hence are more reliable. the small sample size increases the type II error.

sagar uparkar
par sagar uparkar , Manager , Icici Bank

Hi QasiM,

Good afternoon.  In my opinion, Type I error is more related to level of significance. If you'll change it from .05 to .01 then there'll be less probability of getting Type I error. Type II error is more related to sample size. If you'll increase the sample size then Type II error will get reduce. To decide the appropriate sample size you should use different tests like, t-test, z-test etc. as per your data requirement. Apart from this, there is one more possibility is wrong recording of data or poor performance of subject. So after studying the data carefully, you can come to final conclusion about your action plan. Whether you should change significance level, sample size or rectify the data or train subject to improve accuracy. I hope I'd explained well. If you do have different opinion kindly share. 

More Questions Like This