forked from statsmodels/statsmodels
-
Notifications
You must be signed in to change notification settings - Fork 0
test page for experimenting with the wiki
josef-pkt edited this page Apr 25, 2012
·
2 revisions
[ P(E) = {n choose k} p^k (1-p)^{ n-k} ]
\\frac{x^5}{2*x + 1}
x^5
The CDF is defined as:
\\exp\left(-\\lambda\\right)\\sum_{i=0}^{y}\\frac{\\lambda^{i}}{i!}
where \\lambda ( lambda ) assumes the loglinear model. I.e.,
\\ln\\lambda_{i}=X\\beta
The parameter X is X\\beta in the above formula.
math without rst markup see `https://github.com/github/gollum`_
[ y = f(X beta) + epsilon ]
[ E(X_t, epsilon_t) = 0 ]
[ E(epsilon_t, epsilon_{t'}) = 0 hspace{3} textbf{for all} hspace{3} t neq t' ]
If we additionally assume that the errors are i.i.d. normal, then ...