[Zapmeister]

Originally Posted by Trumpet63

But what's the R squared?
Did we accept or reject the null hypothesis?

ok i'm not sure if you're being serious with this post, or (as i suspect) whether you've just finished taking a statistics class and decided to fling words around on here to make yourself look smart, but just in case:

if you want the r^2 value then put the numbers into the online calculator i linked and press the calculate button and scroll down.

linear regression is the wrong test to use for hypothesis testing here. that's because the number of accounts has got to go up with time, which is a causal factor that you're ignoring when you do the test. what you should be doing is testing whether the new accounts are generated at a constant rate, and if you want to do that using only the data i've given you, you do a chi-squared goodness-of-fit. but in this case there's no point in doing that because if you stare at the numbers and don't blink for a few seconds, you'll notice that hell will freeze over and leonardo dicaprio will win an oscar before a poisson process would generate numbers like this. obviously a fixed rate poisson process is far too strong a condition to use here, but there's not much that can be done except assume linearity and go for the trend line.

Originally Posted by _Zenith_

With people like Preggers, we'll hit that milestone easily in the next year, I won't even give it a full year to be honest

haha yeah he was talking to me about this in prochat a while back. xelnya (i think) said give it 2 months and he'll get there :P