Loading…
revised
Measuring departure between the posterior predictive distribution and the true data generating distribution
added 38 characters in body
Loading…
comment
Measuring departure between the posterior predictive distribution and the true data generating distribution
My bad, it should be a half-Cauchy prior which means that you are truncating the negative part of Cauchy and its supported over $(0,\infty)$.
Loading…
comment
Co-ordinate ascent update for $B$
Thank you @greg for the detailed explanation!
accepted
revised
Co-ordinate ascent update for $B$
deleted 296 characters in body
Loading…
Loading…
awarded
revised
Bounding an exponential integral by finding a function
deleted 372 characters in body
Loading…
comment
Bounding an exponential integral by finding a function
I see, but can we actually give an upper bound to the integral, independent of $t, A$?
revised
Bounding an exponential integral by finding a function
added 61 characters in body
Loading…
revised
Bounding an exponential integral by finding a function
added 13 characters in body
Loading…
Loading…
comment
Show that $\lim_{n \rightarrow \infty} \frac{N_n(i,j)}{N_n(i)}=P_{ij}$
Thanks for your answer. I was actually skeptical about writing $$\frac{1}{n} \sum_{m=0}^{n-1} I_{\{X_m=i, X_{m+1}=j\}} \rightarrow \pi_i.P_{ij}$$. Do you think that is correct?
revised
Loading…
revised
Show that $\lim_{n \rightarrow \infty} \frac{N_n(i,j)}{N_n(i)}=P_{ij}$
added 385 characters in body; edited tags
Loading…
revised
Loading…