PDA

View Full Version : Hypothesis

jaben
12-01-2005, 01:21 PM
HI,
I have a question about null and alternate hypothesis. So basically the test is to see if we can reject the null (HO). I am not going to test this assignment since I don't have all the data. I need to write a hypothesis with two population means for online clothing sales and retail clothing sales. In the future, with more data, I am going to test this to see if people shop more at online clothing stores or at retail clothing stores. So for now I just need to come up with the basics.
Am i on the right track? And this is a two tailed test?
HO:u=retail clothing stores
H1:u<=retail clothing stores

HO:u=online clothing stores
H1:u<online clothing stores

Ok I am really confused, as always, so I could be completely off.
Thanks for your help.

JohnM
12-01-2005, 01:28 PM
Since you're interested in a general comparison of retail vs online, and you don't necessarily suspect that one will be higher than the other, it is a non-directional (two-tailed) hypothesis test

Ho: mu(retail) = mu(online)
Ha: mu(retail) <> mu(online)

jaben
12-01-2005, 01:39 PM
Thanks for your help. Oh and why do some use HA instead of H1?
Thanks,

JohnM
12-01-2005, 01:47 PM
Just convention, I suppose - really no difference....

...although in some really advanced courses there may be coverage of situations with "multiple" alternative hypotheses, which would favor using Ho, H1, H2, ..., Hn