https://www.jiscmail.ac.uk/cgi-bin/webadmin?RSS&L=BUGS&v=1.0BUGS List
https://www.jiscmail.ac.uk/cgi-bin/webadmin?A0=BUGS
BUGS List Archives2016-08-15T09:35:52ZPowered by L-Soft's LISTSERV mailing list manager
http://www.lsoft.com/products/listserv-powered.asp
http://www.lsoft.com/images/listserv_small.gifCall for Papers- 3rd International Conference on Commmunications and Network Engineering Limerick City, Ireland December 7-12, 2016. Due September 30th.
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;63b68128.1608
Dear all<br>Apologies for Cross Posting- please see here http://html.asr.org/iccnecfp2016/ Call for Papers for the forthcoming International Conference on Communications and Networking Engineering, Limerick city, Ireland, 7-12th December 2016. Submissions due September 30th..<br><br>Best Regards<br><br>Padraig Kirby MSc (LIS) BA (Hons) HdipLIS<br>Senior Library Assistant<br>The Library<br>Limerick Institute of Technology,<br>Moylish park<br>Limerick<br>Ireland<br>00353 61 293516<br>Padraig.Kirby@lit.ie<mailto:Padraig.Kirby@lit.ie>
2016-08-12T12:57:22+00:00Padraig.Kirbyhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;63b68128.1608Winbugs trap error 66
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;6d24620c.1608
Hello,<br><br>I am trying to conduct a network MA and I keep getting an error in winbugs14 called trap error 66.<br><br>From what I've read online, all I need to do is change the prior on the standard deviation from dunif(0, 100) to dunif(0, 10). Unfortunately, I have no idea how to do this. Can you please provide assistance? [...]
2016-08-08T15:15:08+00:00Mark Gichuruhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;6d24620c.1608error variable is not defined in two-level SEM with categorical variables
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;cc2b1db3.1608
Hello,<br><br>I'm building a two-level SEM in openBUGS with a mixture of ordered<br>categorical, dichotomous and continuous variables. Some of these<br>variables contain missing values at random. OpenBUGS is able to handle<br>missing values in the continuous variables but crashes if there are<br>missing values in the ordered categorical or dichotomous variables.<br><br>I used the latent variable approach described in Lee 2007 (Structural<br>equation models a Bayesian approach, ch 6 and 7) to handle the<br>categorical and dichotomous variables. Each categorical variable is<br>assumed to come from a underlying continuous normal variable. In order<br>to link the order categorical variable [...]
2016-08-08T10:40:59+02:00michelahttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;cc2b1db3.1608Re: Variable TT is not defined: Error in OpenBUGS
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;6436c340.1608
On 04/08/16 16:00, subrata sarker wrote:<br>> Hello,<br>><br>> I am trying to run a model in OpenBUGS. But I am getting an error<br>> "Variable TT is not defined". Here is my BUGS code:<br>><br>That's a R issue: you haven't given any of the variables of sp.data a name:<br><br>> Data=structure(.Data=c(0.1, 0.2, 0.5, 0.2, 0.2, 0.8, 0.2, 0.4, 0.3,<br>0.2, 0.1, 0.6, 0.1, 0.5, 0.6), .Dim=c(5,3))<br>> TT=nrow(Data)<br>> N=ncol(Data)<br>> y=as.matrix(Data)<br>> sp.data = list(y, N, TT)<br>> str(sp.data)<br>List of 3<br>$ : num [1:5, 1:3] 0.1 0.2 0.5 0.2 0.2 0.8 0.2 0.4 0.3 [...]
2016-08-04T16:37:08+02:00Bob O'Harahttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;6436c340.1608Variable TT is not defined: Error in OpenBUGS
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;b0194ba6.1608
Hello,<br><br>I am trying to run a model in OpenBUGS. But I am getting an error "Variable<br>TT is not defined". Here is my BUGS code:<br><br>model{<br>for(t in 2:TT){<br>for(i in 1:N){<br>y[t,i] ~ dmnorm(y1[t,i], tau[i])<br>}<br>y1[t,1] <- (p[1,1] * y[t-1,1]) + (p[1,2] *y[t-1,2]) + (p[1,3] *y[t-1,3])<br>y1[t,2] <- (p[2,1] * y[t-1,1]) + (p[2,2] *y[t-1,2]) + (p[2,3] *y[t-1,3])<br>y1[t,3] <- (p[3,1] * y[t-1,1]) + (p[3,2] *y[t-1,2]) + (p[3,3] *y[t-1,3])<br>}<br>for (s1 in 1:N) {<br>for (s2 in 1:N) {<br>p[s1, s2]~ dnorm(1, 1)<br>}<br>}<br>for(i in 1:N){<br>tau[i] ~ dgamma(0.001, 0.001)<br>}<br>} [...]
2016-08-04T16:00:07+02:00subrata sarkerhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;b0194ba6.1608Re: Winbugs: "this initial value does not correspond to a stochastic node"
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;7479244e.1608
Hi,<br><br>I suggest to add the right loop the of the node "for(t in 2:(n-1)){"<br>at the end of the line alpha[t]<-exp(eta[t]).<br><br>try to<br>Quoting Bob O'Hara <bohara@senckenberg.de>:<br><br>> On 01/08/16 11:41, Lecomte Jean-Baptiste wrote:<br>>> Hi,<br>>><br>>> I think you have to give an initial value for every stochastic<br>>> nodes. For example, you have to give initial values for epsilon,<br>>> delta, gamma ....<br>>><br>> No you don't: any stochastic node that doesn't have an initial value<br>> given will have one generated from its prior.<br>><br>> I can't even compile the code (multiple definitions [...]
2016-08-02T08:33:22+03:00Elisabet AGGELIDOUhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;7479244e.1608Re: Winbugs: "this initial value does not correspond to a stochastic node"
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;a14dd41d.1608
On 01/08/16 11:41, Lecomte Jean-Baptiste wrote:<br>> Hi,<br>><br>> I think you have to give an initial value for every stochastic nodes.<br>> For example, you have to give initial values for epsilon, delta, gamma<br>> ....<br>><br>No you don't: any stochastic node that doesn't have an initial value<br>given will have one generated from its prior. [...]
2016-08-01T13:58:20+02:00Bob O'Harahttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;a14dd41d.1608Re: Winbugs: "this initial value does not correspond to a stochastic node"
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;951c7ffc.1608
Hi,<br><br>I think you have to give an initial value for every stochastic nodes. For<br>example, you have to give initial values for epsilon, delta, gamma ....<br><br>Jean-Baptiste<br><br>2016-07-28 10:57 GMT+02:00 Abongo <abongovin@gmail.com>:<br><br>> Hello, List!<br>> This message came up when I tried to load initial values. This seems to be<br>> an uncommon error since my search of Internet did not get any hits. Any<br>> comments and suggests would be very much appreciated. Thank you!<br>><br>> model{<br>> c<-10000<br>> for(t in 2:(n-1)){<br>> zero[t]<-0<br>> zero[t]~dpois(mean.zero[t])<br>> mean.zero[t]<- -l[t]+c<br>><br>> l[t]<-0.5*(log(lambda.error)-log(2*22/7)-3*log(epsilon[t]))-0.5*lambda.error*pow((epsilon[t]-mu[t])/mu[t],2)/epsilon[t]<br>> }<br>> [...]
2016-08-01T11:41:28+02:00Lecomte Jean-Baptistehttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;951c7ffc.1608Single season- Two species occupancy model
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;a61fd237.1607
Dears<br><br>I am estimating some occupancy parameters (Occupancy model) in Presence.<br>The model used was "Single season- Two species" and I compared between<br>these parametrization and I selected the model with the AIC<br><br>1. ψA, ψB, φ, *p*A, *p*B, *r*A, *r*B, λ<br>2. ψA, ψBA, ψBa, *p*A, *p*B, *r*A, *r*BA, rBa<br>3. ψA, ψBa ,*N*, pA, *p*B, *r*A, *r*Ba, *r* [...]
2016-07-28T23:59:32-05:00Néstor Javier Roncancio Duquehttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;a61fd237.1607Winbugs: "this initial value does not correspond to a stochastic node"
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;fdca4129.1607
Hello, List!<br>This message came up when I tried to load initial values. This seems to be an uncommon error since my search of Internet did not get any hits. Any comments and suggests would be very much appreciated. Thank you!<br><br>model{<br>c<-10000<br>for(t in 2:(n-1)){<br>zero[t]<-0<br>zero[t]~dpois(mean.zero[t])<br>mean.zero[t]<- -l[t]+c<br>l[t]<-0.5*(log(lambda.error)-log(2*22/7)-3*log(epsilon[t]))-0.5*lambda.error*pow((epsilon[t]-mu[t])/mu[t],2)/epsilon[t]<br>}<br>for(t in 2:(n-1)){<br>y[t]~dpois(lambda[t])<br>lambda[t]<-alpha[t]+lambda_1*y[t-1]+epsilon[t]<br>alpha[t]<-exp(eta[t])<br><br>for(i in 1:s){<br>eta[t]<-alpha_1+beta*t+(gamma[i]*sin(omega[i]*t)+delta[i]*cos(omega[i]*t))<br>omega[i]<-i*3.14/26<br>}<br>}<br>for(t in 2:(n-1)){mu[t]~dnorm(0.0,0.001) epsilon[t]~dgamma(0.01,0.01)}<br>for(i in 1:s){gamma[i]~dnorm(0.0,0.0001) delta[i]~dnorm(0.0,0.0001)}<br>lambda.error~dgamma(0.01,0.01)<br>dispersion.p<-1/lambda.error<br>lambda_1~dgamma(0.01,0.001)<br>alpha_1~dnorm(0.0,0.001)<br>beta~dnorm(0.0,0.0001)<br>}<br>list(y=c(1,0,5,2,1,2,0,4,0,0,0,3,1,1,0,0,2,2,0,2,6,3,2,1,2),s=10,n=25)<br>list(mu=c(.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1,.1),lambda.error=1,lambda_1=1,alpha_1=1,beta=1)
2016-07-28T09:57:12+01:00Abongohttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;fdca4129.1607Short course: Applied Bayesian modelling for ecologists and epidemiologists
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;754139a7.1607
Dear all<br><br>Please see below for details of a course that may be of interest, to be held on the shores of Loch Lomond, near Glasgow, at the end of October. The course will cover the theory and application of a number of Bayesian methods, with a focus on practical problems including participants’ own data where possible. [...]
2016-07-21T15:31:24+00:00Matt Denwoodhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;754139a7.1607Re: BUGS- compilation of data; network meta-analysis
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;e9813361.1607
If you're using the network meta-analysis code from the NICE Decision<br>Support Unit wedsite (<br>http://www.nicedsu.org.uk/Evidence-Synthesis-TSD-series(2391675).htm), then<br>it can handle NA's. You just have to be sure to load in all data inputs<br>(one as list, one as array).<br><br>Best wishes,<br><br>Nicky<br><br>On 14 July 2016 at 11:24, Michael Conroy <mjconroy52@gmail.com> wrote:<br><br>> If you have NA in the data array BUGS treats those element as missing and<br>> you must specify a prior .<br>><br>> I've found it easier many times to specify the likelihood only over those<br>> indices that are not missing to avoid the missing [...]
2016-07-14T11:46:25+01:00Nicky Weltonhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;e9813361.1607Re: BUGS- compilation of data; network meta-analysis
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;e9210851.1607
If you have NA in the data array BUGS treats those element as missing and you must specify a prior .<br><br>I've found it easier many times to specify the likelihood only over those indices that are not missing to avoid the missing problem.<br><br>Dr. Michael J. Conroy<br><br>MJ CONROY WILDLIFE BIOMETRICS<br>345 Three Oaks Dr.<br>Athens, GA 30607<br>(706) 308-6208 [...]
2016-07-14T06:24:44-04:00Michael Conroyhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;e9210851.1607BUGS- compilation of data; network meta-analysis
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;491bdf0.1607
Hi All,<br>I am running a winbugs code ( a fixed effect model for NMA). I can check the model, load the data. But the compilation failed. An error appears stating : made use if undefined node t. The t[,] is a matrix with unbalanced data containing the treatment number and NA. I don't know if the presence if NA is responsible for that? Does someone already encounter that problem? Please I need a help. [...]
2016-07-14T10:45:40+01:00SOLANGE Titnkheuhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;491bdf0.1607Re: Selecting Different Percentiles for the Posterior Distribution
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;8345a393.1607
> From: Chin, Paul W<br>> Sent: 13 July 2016 22:22<br>><br>> When using the Sample Monitor Tool, I can select several percentiles<br>> of my posterior distribution to view. The list is limited and jumps<br>> from the 75th percentile to the 90th percentile. Is there a way to<br>> request that WinBUGS displays the 80th percentile of my posterior<br>> distribution? [...]
2016-07-14T08:14:51+00:00MILLARD A.R.https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;8345a393.1607Selecting Different Percentiles for the Posterior Distribution
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;5cfac680.1607
Hi All,<br><br>When using the Sample Monitor Tool, I can select several percentiles of my posterior distribution to view. The list is limited and jumps from the 75th percentile to the 90th percentile. Is there a way to request that WinBUGS displays the 80th percentile of my posterior distribution?<br><br>Thanks for any help and have a great day, [...]
2016-07-13T21:22:29+00:00Chin, Paul Whttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;5cfac680.1607'STATS COURSE - Applied Bayesian modelling for ecologists for epidemiologists'
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;d16cf9c4.1607
Applied Bayesian modelling for ecologists and epidemiologists (ABME)<br><br>Delivered by Dr. Matt Denwood and Prof. Jason Matthiopoulos<br><br>http://prstatistics.com/course/applied-bayesian-modelling-for-ecologists-and-epidemiologists-abme/<br><br>This 6 day course will run from 26th – 31st October 2015 at SCENE field station, Loch Lomond national park, Scotland.<br><br>This application-driven course will provide a founding in the basic theory & practice of Bayesian statistics, with a focus on MCMC modeling for ecological & epidemiological problems.<br>Starting from a refresher on probability & likelihood, the course will take students all the way to cutting-edge applications such as state-space population modeling & spatial point-process modeling. By the end of the week, [...]
2016-07-12T09:59:51+01:00Oliver Hookerhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;d16cf9c4.1607Re: Short Courses: Survival analysis courses by the SSC, University of Reading, UK (July 2016)
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;6ff29db8.1607
Dears<br><br>I am estimating some occupancy parameters (Occupancy model) in Presence.<br>The model used was "Single season- Two species" and I compared between<br>these parametrization and I selected the model with the AIC<br><br>1. ψA, ψB, φ, *p*A, *p*B, *r*A, *r*B, λ<br>2. ψA, ψBA, ψBa, *p*A, *p*B, *r*A, *r*BA, rBa<br>3. ψA, ψBa ,*N*, pA, *p*B, *r*A, *r*Ba, *r* [...]
2016-07-04T13:20:45-05:00Néstor Javier Roncancio Duquehttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;6ff29db8.1607question regarding spatial.exp function
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;81e1639c.1606
Hello,<br><br>My name is Munir Ahmad Nayak and I'm a Ph.D student at the University of Iowa.<br>Recently, I started using OpenBugs for some of my analyses. I have a small question on the function spatial.exp[mu[], x[], y[], tau, phi, kappa].<br>I have measurement stations over a large area (United States), and I want to make sure that I use great-circle distance between the stations (rather than Pythagoras distance). [...]
2016-06-20T18:53:53+00:00Nayak, Munir Ahttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;81e1639c.1606Short Courses: Survival analysis courses by the SSC, University of Reading, UK (July 2016)
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;2aff4ee4.1606
Dear list members,<br><br>Apologies for cross-posting. Note these courses are taking place as planned ...<br><br>We are re-announcing the following short courses, which are scheduled to take place at the Statistical Services Centre in July 2016. Summary information is given below. For more detailed information and registration forms please see http://www.reading.ac.uk/ssc/, or email statistics-courses@lists.reading.ac.uk<mailto:statistics-courses@lists.reading.ac.uk>. [...]
2016-06-20T10:05:33+00:00James Gallagherhttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;2aff4ee4.1606Hierarchical CAR Model
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;2c0d236c.1606
Hello,<br><br>I trying a model both individual level predictors and spatial correlations. I know which counties the individuals live, and I want to determine if after controlling for individual level predictors if there is some spatial correlations.<br><br>Is this possible using a car.normal model.<br><br>Below is my current code:<br><br>model;<br>{<br>for(i in 1:N)<br>{<br>DV[i] ~ dbern(p[i])<br>logit(p[i]) <- a + b1*V1[i] + b2*V2[i] + b3*V3[i] + b4*V4[i] + S[[i]]<br>}<br>for(j in 1:M)<br>{<br>S[j] ~ car.normal(adj[], weights[], num[], precv)<br>}<br>a ~ dnorm(-6.4, 5.7E-2)<br>b1 ~ dnorm(1.6, 5.8E-2)<br>b2 ~ dnorm(2.6, 5.7E-2)<br>b3 ~ dnorm(2.8, 5.7E-2)<br>b4 [...]
2016-06-08T18:23:13+01:00Peter Georgantopouloshttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;2c0d236c.1606"Stan for Pharmacometrics" course and workshop in Paris September 2016
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;8b0a0892.1606
Dear All,<br><br>We are very pleased and excited to announce a "Stan for Pharmacometrics" course and workshop in Paris this September.<br><br>A three day course will be led by three core members of the Stan development team on 19--21 September 2016. You can find more details and register at:<br>http://bit.ly/stan-pkpd<br><br>A one day workshop will be held on 22 September 2016. If you would like to give a talk, please send an email by the end of July with a title and a short one-page abstract to julie.bertrand@inserm.fr. You can find more details and register at:<br>https://www.eventbrite.com/e/stan-for-pharmacometrics-day-tickets-25399716235 [...]
2016-06-07T16:42:26+00:00Bertrand, Juliehttps://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;8b0a0892.1606Trap 33 Precondition violated - using ode
https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;66f441d8.1606
Dear all,<br><br>I am not quite sure whether this is the right way to get help, but I am trying to fit a set of differential equations to data. I am new in Bayesian statistics and thus Openbugs.<br>Currently I am trying to fit this extremely simple model and I get the Trap 33 Precondition violated message. [...]
2016-06-06T12:09:45+00:00Fischer, E.A.J. (Egil)https://www.jiscmail.ac.uk:443/cgi-bin/webadmin?A2=BUGS;66f441d8.1606