Page 2 of 2
Posted: Wed Jul 07, 2004 2:39 pm
by bpp
Hi John R.,
Maybe it would be helpful if you could write down exactly what questions you are trying to answer here. For example, do you have a null hypothesis about what the SWR is? I don't...
Bpp
Posted: Wed Jul 07, 2004 4:31 pm
by JWR1945
For Bpp:
I am much happier presenting a two-sided 90% confidence level along with a footnote that explains that the confidence level is 95% when using a strictly one-sided interpretation than I am when I give the results from a one-sided test and then having to explain away the (apparent) contradiction to the average person.
You are correct in the sense that a discussion strictly limited to Safe Withdrawal Rates would identify the confidence level as 95%.
The difficulty is that all three rates are important: the Calculate Rate, the Safe Withdrawal Rate and the High Risk Rate. This suggests that I am really interested in knowing whether the data are consistent with the Calculated Rate, not whether a rate is safe.
Have fun,
John R.
Posted: Wed Jul 07, 2004 6:37 pm
by bpp
Fair enough.
Bpp