« first day (2344 days earlier)      last day (2647 days later) » 

12:07 AM
@whuber Thank you very much for pointing out the now obvious flaw in the plotting on my post today. It made a huge difference! I had always assumed that the asymmetry was due to plotting "imperfections", but it ended up diminishing the illustration of the point. I hope you didn't mind my asking you to check on the code - at the time I wasn't sure I was going to be able to fix it, and I figured it was best to capture your attention while the issue was recent. Best wishes, and thank you!
 
 
4 hours later…
3:58 AM
I see Frank Harrell has a blog ... fharrell.com
 
 
2 hours later…
5:49 AM
Thanks for mentioning that @Glen_b. I like Frank's work but I didn't know about his blog.
 
6:12 AM
Only been going a very short time
 
 
2 hours later…
8:34 AM
Yay
38.5 hours of self-studying statistics got me there.
 
9:29 AM
So does Xi'an - xianblog.wordpress.com - for anyone who hasn't seen it yet.
 
 
3 hours later…
12:50 PM
The lower limit could be 0 but it is not clear that a two-sided interval would make 0 the lower cutoff. It is possible that this is for a one-sided interval. the number you get and the number in the book differ by only the second decimal place. So other possibilities are round off error or or differences in interpolation from a t table. Maybe you can clear up thee issues and then focus on a question that could be answered. — Michael Chernick 4 mins ago
I wish I understood something from this comment
 
 
4 hours later…
4:48 PM
Am I just grumpy or is question quality declining? stats.stackexchange.com/questions/256527/… and stats.stackexchange.com/questions/256495/technique-and-model and stats.stackexchange.com/questions/256453/… seem to indicate that today is a bad day for question quality -- is it part of a trend?
 
5:40 PM
@Sycorax average question quality has been declining for a long while. Those are typical low-quality questions, but we get more of them as Stackexchange becomes more popular (it shows up in a lot of google searches now for example, which brings in a lot of new people every day)
@CowperKettle An actually correct answer to that question you post above is "It's impossible to tell from the information given".
There's no basis there to justify using a normal distribution
The mere fact that the sample mean is calculated on a sample of size 25 is certainly not enough.
No doubt the text has made a false claim somewhere, or you wouldn't have been able to pick the answer it thought was correct
 
6:03 PM
@CowperKettle Well, I found Michael's comment confusing too. Hopefully fcop's following comment made more sense to you
 
Yes, it made more sense. I'll be translating some pharmaceutical stuff for several days probably, I hope some statistics stays in my head. O_O At least today I read some and did some simple problems.
 
 
4 hours later…
9:59 PM
@Sycorax It's you. :-) I've been growing steadily grumpier for several years, too... .
Seriously, though, the number of closed questions as a proportion of the total has been rising exponentially for a long time. The number of closed questions has increased 48% per year for over six years, while the number of questions asked has increased only 33% per year. During the last three years the rates have been 45% and 25%, respectively.
That means that in another 7.5 years, all questions will be closed!
 
10:21 PM
@whuber Well, at least then the mods will get some well-deserved rest (because we can just script to auto-close all questions)
 

« first day (2344 days earlier)      last day (2647 days later) »