About
I am a professor of Statistics at both Université Paris Dauphine, Paris, France, and University of Warwick, Coventry, United Kingdom, and an ERC Synergy grant holder over the period 2023-2029, with a definitely unhealthy (but so far not fatal) fascination for mountains and (easy) climbing, in particular for Scotland in Winter, an almost daily run, and a reading list mainly centred on fantasy books… Plus an addiction to bloggin’ since 2008! Hence the categories on this blog (or ‘og, because ‘log and b’og did not sound good). The Statistics posts do mainly focus on computational and Bayesian topics, on papers or preprints I find of interest (or worth criticising), and on the (not so) occasional trip abroad to a research centre or to a conference. Societal, political, and philosophical entries are only reflecting my personal opinions, that I do not ask ‘Og’s readers to share.
Needless to say (?), this blog is not approved by, supported by, or in any other way affiliated with the Université Paris Dauphine, CREST-INSEE, University of Warwick, or any other organization, and it only reflects my opinions. This is also one of the reasons why it is posted on wordpress rather than on my University webpage, another one being that wordpress provides a handy (if sometimes slow) tool for editing blogs…
Warning: the links found on this blog to Amazon.com or more rarely to Amazon.fr entries are actually susceptible to earn me a monetary gain for qualifying purchases. And to feed further my book addiction.
Xi'an's Og 
March 23, 2023 at 2:37 pm
Hello ,
I am looking into using RJMCMC to select the number and location of knot locations for a spatial model. At each step of the MCMC algorithm, we propose adding, deleting, or moving a knot location with equal probability 1/3. I wish to find out if I’d be breaking any rules (e.g., violating detailed balance) by letting these probabilities differ from 1/3 by adapting to the data. For example, if there is high variability, we may favor adding a knot. The literature on RJMCMC is difficult to grasp so I was hoping I could make a post on stack exchange to provide more details and get your thoughts. I’m sure you’re very busy and I completely understanding if you do not have time. Thank you!
March 23, 2023 at 5:28 pm
Changing the probabilities is akin to picking another proposal than the prior (?), hence I see no issue provided detailed balance / reversibility is preserved
March 23, 2023 at 5:36 pm
I have provided more details here
https://stats.stackexchange.com/questions/610486/reversible-jump-mcmc-for-knot-selection
if you’re interested. My intuition says it shouldn’t be an issue but I’m not sure how to justify this.
March 24, 2023 at 5:43 pm
I’m curious: What’s the process you are trying to model? There are a large number of physical/biological/chemical processes it might describe and it’s hard for me to differentiate between them without these details. Your description is too abstract.
August 31, 2021 at 3:25 pm
[…] who inspired me to do this. First is the statistician Christian Robert, whose blog can be found here. In addition to running a great blog, I have a high opinion of Christian Robert’s character […]
May 18, 2021 at 9:22 pm
Hello,
I am trying to write winbugs codes for a latent variable w.
My outcome y follows Bernoulli distribution and I want to use the latent one (w) in my analysis.
y=1 if w>0 and y=0 if =<0
W is continuous
Could you please help me about the codes?
May 19, 2021 at 8:34 am
I suggest you ask the question on Cross Validated (with more details as the question cannot be answered as such).