Version: 1.3.1 Category: CHANGES Text: ... columns passed to create.dml from predict.crm so remove.unused.columns=FALSE in cases where prediction is for Version: 1.3.1 Category: a single animal and all numeric values are 0. Be careful using this as it could cause problems where a column should be dropped Text: Version: 1.3.0 Category: CHANGES Text: Using 0 as a strata label causes issues with MVMS model so it is not allowed. Version: 1.2.9 Category: CHANGES Text: *Fixed js which was storing scaled design matrix in model_data rather than unscaled version. This caused differences in results between compute_real and predict.crm when a continuous covariate was specified. predict.crm uses compute_real but it creates the unscaled design matrices before calling compute_real whereas the latter used the incorrect stored versions. Now both will produce the same result. Documentation for scale parameter was changed to specify that setting scale=1 will produce an unscaled design matrix. *Modified predict.crm and compute_real to only return any specified parameter rather than all parameters from model object when ddl and dml are both NULL. *Fixed initiate_pi and mvms_design_data to use stringsAsFactors=TRUE in data.frame call that creates strata fields. This was handled for other functions when default was changed in R but was missed in these functions. Version: 1.2.9 Category: CHANGES Text: Default method is now nlminb for crm. It is a much more reliable optimization method than BFGS-B. Version: 1.2.8 Date: 2023-10-19 Category: CHANGES Text: *Minor changes to function names, examples etc to fix warnings/notes on CRAN check. Version: 1.2.7 Category: NEW Text: added argument unit_scale to crm, predict.crm, and compute_real. If set to FALSE the real value estimates for variables like Phi and S are based on the interval length rather than on unit scale (e.g., annual if unit is 1 year). See ?crm for an example. Version: 1.2.7 Category: CHANGES Text: ddl argument to crm can now be specified as "external" and the ddl should be "save"d to a file "ddl.rda". This can save memory when the ddl gets extremely large (eg 6GB) and it is used with a model that simplifies the ddl. Version: 1.2.7 Category: CHANGES Text: in a related change for msld_tmb, the full ddl is written to a tmp.rda and read in when needed. This saves memory for tmb to build optimization function (MakeAdfun) and run the optimization. Version: 1.2.7 Category: CHANGES Text: in crm.wrapper previously it was passing entire model for the initial values if a saved external model was found. Now it only passes the beta list and it removes the model once it is loaded and beta list extracted. This can save a considerable amount of memory if the saved model is large. Version: 1.2.7 Category: CHANGES Text: global_decode will use the matrices in results$mat even if the model is not HMM. MSLD model outputs dmat and gamma from TMB code but does not have HMM functions for HMMLikelihood. Many of the TMB functions may be moved in this direction. Version: 1.2.7 Category: CHANGES Text: crmlist_fromfiles was changed so it saves name of external file including .rda so it is consistent with crm.wrapper when external=TRUE Version: 1.2.7 Category: CHANGES Text: patched code in cjs.initial such that if glm fails that the initial values are set to 0 rather than the code failing altogether. Thanks to Jason Baker for bringing this to my attention. Version: 1.2.7 Category: CHANGES Text: fixed wrong or unclear text in the help files for process.data and crm thanks to comments from Jason Baker. Version: 1.2.7 Category: CHANGES Text: MSLD and MSCJS models using TMB were changed to use simplification indexing for random effects as done for fixed effects. This is a transparent change to user. Version: 1.2.6 Date: 2019-12-09 Category: BUG FIXES Text: more changes to FORTRAN files to avoid warnings. Version: 1.2.6 Date: 2019-12-09 Category: BUG FIXES Text: process.data will now change any data.table to a data.frame. Caused issues with second vignette in RWildbook. Version: 1.2.5 Category: BUG FIXES Text: more changes to FORTRAN files to avoid warnings on use of multiple DO's to a single continue and tabs. Version: 1.2.4 Category: BUG FIXES Text: changes to FORTRAN files to avoid warnings on use of multiple DO's to a single continue and tabs. Version: 1.2.4 Category: BUG FIXES Text: changes to use is() function instead of class which can always return more than one value which can cause problems. Version: 1.2.3 Category: BUG FIXES Text: Fixed lit cited in vignette to remove DOI and URLs which were causing issues on CRAN. Problem with optimx on Debian Linux remains until optimx is patched. Version: 1.2.2 Category: NEW Text: added TMB version of mvmscjs in mvmscjs_tmb R code and mvmscjs_tmb.cpp code. It is an order of magnitude faster than ADMB code for the sealion example. For this model, the default method is nlminb for optimization because it performs better in general. Setting method can over-ride the default. Version: 1.2.2 Category: NEW Text: added model MSLD for multistate live recapture and dead recovery in TMB. Version: 1.2.2 Category: NEW Text: added simulated annealing method (SANN) with optim for MSLD, MSCJS and MVMSCJS models. Note that this method does not implement a convergence criterion and simply runs until itnmax iterations are completed. You can get a report of iteration results using control=list(trace=1). You can also set the temp with control. See ?optim for complete help on SANN. Version: 1.2.2 Category: BUG FIXES Text: In mvmscjs the parameters were not in the correct order so the admb code did not work with the sealions example because the hessian was not in the expected order. Thanks to Jaume Forcada for reporting the problem. Version: 1.2.2 Category: BUG FIXES Text: fixed bug in make.design.data that was preventing use of newdata in predict.crm and using subtract.stratum for some models. Version: 1.2.2 Category: CHANGES Text: for models hmmmscjs and mscjs, all parameter values can be fixed (eg, ddl$p$fix=1). Previously this resulted in R errors. Version: 1.2.2 Category: CHANGES Text: in crm.wrapper with external=TRUE, if file already exists code will be stopped unless new argument replace=TRUE is specified. Version: 1.2.2 Category: CHANGES Text: Change was made in create.dmdf. For model mvmscjs, previously if there were no uncertain states, the design data for delta was set to NULL. This doesn't work with the admb code so the design data is created and to make all states certain you need to set fix=1 for all records in the design data. The example for mstrata has been expanded to indclude a model for straight multistate data without uncertainty that uses admb code for mvmscjs. Version: 1.2.2 Category: CHANGES Text: code in simplify_ddl was modified to use package data.table which substantially sped up the simplification of the ddl for any given model. Version: 1.2.2 Category: CHANGES Text: in global_decode, if ddl is not specified and list of dmat, gamma, delta matrices are in object$results$mat it will use those values. Version: 1.2.2 Category: CHANGES Text: added argument useHess to TMB versions of code; if TRUE uses hessian in optimization. This is typically slower but can be more precise in finding max. Version: 1.2.2 Category: CHANGES Text: default value for save.matrices has been changed to FALSE for HMM and MVMS models. Version: 1.2.2 Category: CHANGES Text: in make.design.data rownames are reset so they are in numeric order; should not affect code and more for readability for some parameters. Version: 1.2.1 Date: 2018-03-30 Category: CHANGES Text: minor adjustment to Description file and new init.c for routine registration for CRAN pass Version: 1.2.0 Category: CHANGES Text: changed use of rBind and cBind to rbind and cbind for change in MATRIX package; requires >=R3.2.0 so version incremented Version: 1.1.14 Category: BUG FIXES Text: bug fixed in js.hessian which prevented computation of vcv matrix because of earlier change restructuring model object. Thanks to Bjorn Kroger for reporting problem. Version: 1.1.14 Category: NEW FEATURES Text: addition of priors for cjs with TMB - still in testing phase Version: 1.1.14 Category: CHANGES Text: make.design.data will now return an error if field names not in data are specified in static list. Version: 1.1.14 Category: CHANGES Text: in mvmscjs model, p and Phi can now also be all fixed values. Also, all parameters are now checked for empty columns because of fixed parameter values. Previously only Psi was. Version: 1.1.14 Category: CHANGES Text: create.dm now checks and traps for NA in any variable used in the formula and stops if it is encountered and reports problem variable. Previously a very obtuse R error was given. Version: 1.1.13 Date: 2016-12-06 Category: BUG FIXES Text: patched mvms.f for non-admb version of MVMSCJS. Code had an incorrect index that only worked if delta did not vary by individual. Version: 1.1.13 Date: 2016-12-06 Category: BUG FIXES Text: patched mvmscjs.r so that it allows ~0 for delta (all fixed values) in the ADMB version. Also, now all Psi can be fixed as well. Version: 1.1.13 Date: 2016-12-06 Category: CHANGES Text: added returns to output when debug=T for MVMS so warning about last line incomplete should not appear. Version: 1.1.13 Date: 2016-12-06 Category: CHANGES Text: added checks HMMLikelihood to outline errors when dmat, gamma or delta are not formed correctly. Typically this will be due to fix values being set incorrectly. Also, HMMLikelihood is now called prior to mvmscjs to check for those errors to avoid putting checks in ADMB code as well. Version: 1.1.12 Category: NEW FEATURES Text: parameter pi has been added to MVMSCJS model. Including pi allows the initial state at release to be uncertain. See ?initiate_pi for more details. initiate_pi is called from make.design.data for MVMSCJS models so existing code can be run without modification. Version: 1.1.12 Category: NEW FEATURES Text: modified mscjs and multistate.tpl file so fixed parameters can be included and design Version: 1.1.12 Category: matrix can be simplified to speed up execution times Text: added function setup_admb to simplify calling code like cjs and mscjs and standardize handling Version: 1.1.12 Category: of compile, clean arguments and use of existing admb executables in the marked package directory Text: created an ADMB version of the mvms model with simplified design matrix to speed up execution times. The sealion example runs in 1/10 of the time if use.admb=TRUE is set. Version: 1.1.12 Category: of compile, clean arguments and use of existing admb executables in the marked package directory Text: created a TMB version of cjs models that provides real predictions with std errors and is much faster fitting random effects. It allows accumulation of capture histories for time random effects or if group effects included in data. Version: 1.1.12 Category: CHANGES Text: pull request merged from Brett McClintock for probit.cjs. Changes Bayesian updating. Version: 1.1.12 Category: CHANGES Text: all values for a parameter can be fixed (e.g., S=1 or delta=1). This used to cause problems but now if Version: 1.1.12 Category: all values are fixed, a message is given that the formula is being reset to ~0 which has no parameters. Likewise, if formula is ~0 and not all values are fixed then an error is generated. Text: The above change made it possible to allow an mvmscjs specification in which all states are certain. In which case, Version: 1.1.12 Category: no design data is generated for the delta parameter Text: global_decode now requires ddl argument to be specified and does not compute on the fly. Version: 1.1.12 Category: no design data is generated for the delta parameter Text: changed AIC calculation for models with random effects to use only number of fixed effect estimates plus number of random effect sigma values. Version: 1.1.12 Category: BUG FIXES Text: patched mvms.f which had a bad index. Only worked if delta did not vary by individual. Version: 1.1.11 Date: 2016-03-07 Category: NEW FEATURES Text: can now add sigmaPhi and sigmap to list of initial values for random effects CJS model. If none are specified, initial values are now set to -2 instead of 0.1 Version: 1.1.11 Date: 2016-03-07 Category: BUG FIXES Text: patched create.dm which was leaving out one record when chunking through design data. This caused failure when using splines. Version: 1.1.11 Date: 2016-03-07 Category: BUG FIXES Text: beta v-c matrix not being properly stored on CJS with use.admb=TRUE Version: 1.1.11 Date: 2016-03-07 Category: BUG FIXES Text: modified cjs so it doesn't check for tplcpp before running admb. This was stopping it from using admb on Linux. Version: 1.1.11 Date: 2016-03-07 Category: BUG FIXES Text: run=FALSE was not working for non-HMM models because initial.list was not assigned. Version: 1.1.11 Date: 2016-03-07 Category: BUG FIXES Text: fixed bug that was preventing real predictions from being included in results where appropriate. Version: 1.1.11 Date: 2016-03-07 Category: CHANGES Text: in crm.wrapper if a model with the same name is stored externally in the workspace it will load that file and use it as initial values unless the initial argument has been specified in the call. Version: 1.1.11 Date: 2016-03-07 Category: CHANGES Text: modified crmlist_fromfiles and model.table to prevent issues with lack of memory by removing models after loading. Version: 1.1.11 Date: 2016-03-07 Category: CHANGES Text: modified reSplit so it would cope with left-hand-side of random effect being a formula with more than one term Version: 1.1.11 Date: 2016-03-07 Category: CHANGES Text: stop message was changed to warning when right-hand-side of random effect is not a factor variable. This was needed to do something like (1|time:post) where post is a 0/1 variable that restricts the random effect to occur where post=1. Version: 1.1.10 Date: 2015-08-20 Category: BUG FIXES Text: qualified call to choose_files and removed from import so it would build on Linux Version: 1.1.9 Date: 2015-08-20 Category: BUG FIXES Text: fixed bug in compute.real which would fail computing real values for Psi in some models Version: 1.1.9 Date: 2015-08-20 Category: BUG FIXES Text: made change to setup.parameters to handle include and mlogit parameter setting when model is re-run with rerun_crm. Version: 1.1.9 Date: 2015-08-20 Category: CHANGES Text: modified omega to accept a list of transition matrices Version: 1.1.9 Date: 2015-08-20 Category: CHANGES Text: added sealions data for example Version: 1.1.8 Date: 2015-03-17 Category: CHANGES Text: modified crmlist_fromfiles to avoid issue on non-Windows machines which do not support choose.files Version: 1.1.7 Category: CHANGES Text: modified R_HMMlikelihood and loglikelihood to be more useful for computations like in hmmDemo and local_decode Version: 1.1.7 Category: CHANGES Text: matrices delta, gamma, and dmat no longer stored in model object unless argument save.matrices=TRUE. They can be computed with compute_matrices. Version: 1.1.7 Category: CHANGES Text: specification of strata.labels is now required for multistate models. Code will no longer use values it finds in capture history. You can specify more strata that are in the data (unobserved strata). Version: 1.1.7 Category: CHANGES Text: modified mscjs.r to return standard crm model object rather than admb object. Version: 1.1.7 Category: NEW FEATURES Text: added compute_matrices, local_decode, global_decode, backward_prob, omega functions for HMM manipulations Version: 1.1.7 Category: NEW FEATURES Text: added load.model into print.crm so if it is given a file string it will load the external model and then print results. Version: 1.1.7 Category: NEW FEATURES Text: added function crmlist_fromfiles which allows you to select a set of .rda files for external models to create a crmlist with a model table. Version: 1.1.7 Category: NEW FEATURES Text: added function rerun_crm which will run through a model list and re-run any models that did not converge (convergence !=0). A different method can be specified. If a specific set of models are to be re-run regardless of the convergence criterion then modelnums set to a vector of model numbers in the model list and they are re-run instead of those that did not converge. An initial model can be specified which is useful to rerun a model with a model that is a subset of that model. Version: 1.1.7 Category: NEW FEATURES Text: added model class mvmscjs Version: 1.1.7 Category: NEW FEATURES Text: added data set mstrata from MARK and examples showing fitted models from hmmmscjs, mscjs and mvmscjs. The mscjs is not run unless ADMB is available. Code from help can be copied and pasted. Version: 1.1.7 Category: BUG FIXES Text: fixed bug in multistate.tpl that caused array out of bounds in ADMB when in safe mode Version: 1.1.7 Category: BUG FIXES Text: fixed bug in msgamma.f; PSISUM not defined as REAL*8 so sum of transitions slightly less (1E-8) than 1. Version: 1.1.6 Date: 2014-06-17 Category: CHANGES Text: Modification to documentation to avoid overflow on pdf Version: 1.1.6 Date: 2014-06-17 Category: CHANGES Text: Patches to crm so design.data and fixed values were handled when the ddl is specified Version: 1.1.6 Date: 2014-06-17 Category: CHANGES Text: cjs and js were patched so only the initial parameter values are stored; it was errantly storing the model when a model was used to specify initial values Version: 1.1.6 Date: 2014-06-17 Category: CHANGES Text: Additional changes to crm to use optimx to work with HMM models. This required changes to HMMLikelihood argument names which conflicted with optimx. Version: 1.1.6 Date: 2014-06-17 Category: CHANGES Text: cjs and *.tpl files for cjs were modified to use debug; not all exes have been changed yet Version: 1.1.5 Category: NEW FEATURES Text: Real parameters are now created for HMM models with predict.crm; modified compute.real so reduction to unique real parameters occurs before computation of v-c matrix; also certain fields are added (eg occ) so that a full set of real parameters is displayed. That includes showing all parameters in an mlogit set. Version: 1.1.5 Category: NEW FEATURES Text: Added tag loss models HMMCJS1TL and HMMCJS2TL Version: 1.1.5 Category: NEW FEATURES Text: Added switch (remove.unused.columns) to create.dml and create.dm functions that controls whether unused columns in the design matrix are removed or not. The default is TRUE. Version: 1.1.5 Category: CHANGES Text: marked now depends on lme4 rather than imports; only way that Rscript would work. Also added methods to depends. Version: 1.1.5 Category: CHANGES Text: ehmat (encounter history matrix) is now stored with ddl in make.design.data Version: 1.1.5 Category: CHANGES Text: Model BayesMSCJS which also contains uncertain state was added to models.txt in anticipation of its inclusion and probitMsCJS was never used and was removed. Version: 1.1.4 Category: NEW FEATURES Text: Random effects added to probitCJS Version: 1.1.4 Category: BUG FIXES Text: patch made to cjs so accumulate set to FALSE with fixed parameter values and bug in create.fixed.matrix for cjs model was patched Version: 1.1.4 Category: CHANGES Text: proc.form was modified to use code from lme4 to provide better handling of formula; removing intercept was not fully supported in previous version. Version: 1.1.4 Category: KNOWN PROBLEM Text: used of fixed= in parameter specification currently doesn't work; use ddl$fix approach to setting fixed values. Merging described below is not working properly. Version: 1.1.3 Date: 2013-11-23 Category: NEW FEATURES Text: use of fix column in ddl is now supported for all models. By assigning a non-NA value, the real parameter is fixed to the assigned value. If both ddl$fix is assigned and fixed argument is specified these values are merged and the fixed values in the parameter specification will override those in the ddl. Version: 1.1.3 Date: 2013-11-23 Category: NEW FEATURES Text: HMM models are now fitted with crm rather than fitHMM which has been removed. HMM models now include HMMCJS(cjs model),HMMMSCJS (multistate CJS), HMMUMSCJS (multi-state with state uncertainty), HMMU2MSCJS (2 strata level MS with state uncertainty), HMMU2IMSCJS (2 strata level MS with state uncertainty; independent state and second level strata transtions) Version: 1.1.3 Date: 2013-11-23 Category: NEW FEATURES Text: time interval specification can now be assigned via the design data for a parameter if it is not constant across individuals. Version: 1.1.3 Date: 2013-11-23 Category: CHANGES Text: model is no longer case-sensitive Version: 1.1.3 Date: 2013-11-23 Category: CHANGES Text: code for creating design data was improved and is now much faster Version: 1.1.3 Date: 2013-11-23 Category: CHANGES Text: error message given if data contains a field called id; id is forced to be a factor variable throughout Version: 1.1.3 Date: 2013-11-23 Category: CHANGES Text: in mscjs, dml now refers to dml$fe to use fixed effects dm because it currently does not support random effects Version: 1.1.3 Date: 2013-11-23 Category: CHANGES Text: removed use of ::: and exported more functions in marked to avoid that usage Version: 1.1.1 Text: fixes for CRAN submission; same as 1.1.0 except for some transparent changes for CRAN submission Version: 1.1.0 Category: NEW FEATURES Text: improved and added to ADMB options for mixed effect modeling which resulted in the modification of a number of calling arguments to crm. Version: 1.1.0 Category: NEW FEATURES Text: included code that looks for ADMB executable files in the package directory or workspace directory and uses them if they are found and !compile. This allows the user to avoid downloading ADMB and gcc but exe's are only available for windows at present. Version: 1.1.0 Category: NEW FEATURES Text: included code for HMM likelihood model fitting of c-r models (fitHMM) and simulation (simHMM) and supporting functions Version: 1.1.0 Category: NEW FEATURES Text: added clean argument to remove TPL and executable files remaining in the working directory Version: 1.1.0 Category: NEW FEATURES Text: capture history strings can now be comma separated strings to cope with multi-character states Version: 1.1.0 Category: CHANGES Text: modified print.marked.version so it would not conflict with loading devtools Version: 1.1.0 Category: CHANGES Text: modified code in cjs.f and ADMB TPL files to more closely match MARK in handling small values of Prob(ch) to avoid log(0) Version: 1.1.0 Category: CHANGES Text: code was moved from crm to process.data which sets accumulate=FALSE if MCMC method is selected. Version: 1.1.0 Category: CHANGES Text: removed code that prevented splitCH from working on a single ch value Version: 1.1.0 Category: CHANGES Text: added code to crm to assure accumulate=FALSE with any random effect; individual random effect is supposed to work with weights but Version: 1.1.0 Category: tpl and cpp code do not appear to be working together; with accumulate=FALSE, results are correct Text: added code to make.design.data which handles subtract.stratum for Psi type parameters and cjs type parameters which sets value to 1 when cohort=time. Those changes are for fitHMM code which uses subtract.stratum in MS models and uses first occasion seen parameters to make the hmm code work with JS and CJS type models. Version: 1.1.0 Category: tpl and cpp code do not appear to be working together; with accumulate=FALSE, results are correct Text: made changes to cjs and js for changes in optimx package Version: 1.1.0 Category: BUG FIXES Text: create.dmdf was not assigning id values correctly for MSCJS model Version: 1.1.0 Category: BUG FIXES Text: fixed problem introduced in v1.0.7 in which scaled parameters were not being adjusted back to original scale with R MLE optimization. Did not affect results with MCMC or when use.admb=T. Version: 1.1.0 Category: BUG FIXES Text: on Mac or Linux the code to assess whether admb was available, did not work. This was patched. Version: 1.1.0 Category: BUG FIXES Text: fixed bug in cjs which was written with a local copy of R2admb and did not work with the posted R2admb. This also involved changes to coef.crm Version: 1.0.9 Date: 2013-01-25 Category: CHANGES Text: modified iteration counter so it was stored in an environment rather than in global workspace to meet CRAN policy Version: 1.0.8 Category: NEW FEATURES Text: added multi-state model with fixed effects using admb. This is done with new function mscjs and multistate.tpl. It required changes to create.dm,create.dmdf, crm, process.ch and process.data. The code is functional but currently only returns admb object that will not work with other generic functions, so crm will not allow it currently. Version: 1.0.7 Category: NEW FEATURES Text: added use of admb and random effects for CJS. Controlled through arguments use.admb=TRUE and re=TRUE respectively. Doesn't currently work with predict. Version: 1.0.7 Category: BUG FIXES Text: fixed problem in create.dmdf which was removing id field if fields (static) specified Version: 1.0.7 Category: BUG FIXES Text: made change to cjs.f which was incorrectly fixed for loss on capture Version: 1.0.7 Category: BUG FIXES Text: made changes to bayesCR.cpp to remove compilation warnings Version: 1.0.6 Category: CHANGES Text: removed autoscale feature which complicated code and was not very effective Version: 1.0.6 Category: CHANGES Text: Removed simulation code and put in a separate package - simcr. Version: 1.0.6 Category: CHANGES Text: restructured the model object that is returned from crm Version: 1.0.6 Category: NEW FEATURES Text: Function load.model will return externally stored model. Most useful with crm.wrapper and external=TRUE Version: 1.0.6 Category: NEW FEATURES Text: Generic function predict will provide estimates of real parameters and optionally std errors, conf intervals and vcv matrix for the existing or newdata. Does not handle fixed parameters at present. Version: 1.0.6 Category: NEW FEATURES Text: crm.wrapper now returns a list like mark.wrapper in RMark with a model.table. The model list elements are the filenames of the externally saved model results. A default print method is included to print out model selection table. Also, an external argument was added. Version: 1.0.6 Category: NEW FEATURES Text: Modified cjs.initial to use either logit or probit and added initial value calculation to probitCJS Version: 1.0.6 Category: NEW FEATURES Text: Allows initial=value and all parameters given that initial value Version: 1.0.6 Category: NEW FEATURES Text: Can now handle specification of more than one method for optimization within optimx; setting control=list(follow.on=TRUE) in the arguments will let it use par estimates as starting values as it proceeds through list of methods. Version: 1.0.6 Category: NEW FEATURES Text: Added SANN as a stand alone method with optim. Default of itnmax should be increased substantially for this method. Version: 1.0.6 Category: NEW FEATURES Text: Modified cjs.initial to use freq as weight in glm to properly weight estimation when freq>1. Version: 1.0.6 Category: BUG FIXES Text: Fixed bug in js.lnl which was not getting lnl correct when there were no groups; cosmetic and did not affect estimates because was constant part (ui) Version: 1.0.6 Category: BUG FIXES Text: Fixed bug in which accumulate argument was not passed to process.data from crm Version: 1.0.5 Category: NEW FEATURES Text: Simplified code in cjs.f but no obvious payoff in speed. Set p=1e-307 if p=0 and freq>0 to avoid log(0) issue. This is close to value used in MARK. Version: 1.0.5 Category: BUG FIXES Text: Fixed bug in hessian calculation which was using 2*lnl instead of lnl Version: 1.0.4 Category: NEW FEATURES Text: js is now structured the same as cjs and includes all features. Version: 1.0.4 Category: NEW FEATURES Text: integrated probitCJS into crm. Version: 1.0.4 Category: NEW FEATURES Text: Added automatic data accumulation and creation of freq field where all data fields are the same. See argument accumulate in crm. Version: 1.0.4 Category: NEW FEATURES Text: added autoscaling of parameters; see autoscale argument Version: 1.0.4 Category: NEW FEATURES Text: added simcjs and simpopan and related functions for simulating data for cjs and js models Version: 1.0.3 Category: NEW FEATURES Text: Code was modularized to make package easier to extend. js is now structured the same as cjs. Version: 1.0.3 Category: NEW FEATURES Text: If time.interval is a field in the Phi design data it will be used as Phi^time.interval. This allows variation in time intervals across animals as with cohorts of pups branded at different times each year. Version: 1.0.3 Category: NEW FEATURES Text: Added probitCJS model and accompanying code. Version: 1.0.3 Category: BUG FIXES Text: Change made in cjs.accumulate to compare accumulated frequencies to sum of data frequencies. Previously was using number of rows and would fail if any freq>1. Version: 1.0.2 Category: NEW FEATURES Text: Added check on accumulator to report error if sum of frequencies does not match original number of records. Version: 1.0.2 Category: NEW FEATURES Text: Added model convergence check and reporting of model convergence message, if non-null. Version: 1.0.2 Category: NEW FEATURES Text: Added function fix.parameters to create matrix needed for fixing real parameters Version: 1.0.2 Category: NEW FEATURES Text: Added option to include remove.intercept=TRUE in model.parameters list for each parameter to force removal of intercept. Version: 1.0.2 Category: NEW FEATURES Text: Added refit argument to crm, cjs and js to control number of refittings if model doesn't converge. Version: 1.0.2 Category: NEW FEATURES Text: Added function create.links which works out which real parameters can use a sin link. Code was added to cjs and cjs.lnl to use the sin link where appropriate. This is commented out at present! Not sure it is working correctly. Version: 1.0.2 Category: NEW FEATURES Text: Uses optimx function for optimization which allows more methods and multiple methods to be selected. Version: 1.0.2 Category: BUG FIXES Text: Fixed issue with capture history accumulator when chunk_size was larger than needed. Added 1 to pieces. Version: 1.0.2 Category: BUG FIXES Text: Beta estimate names were lost once sparse matrices were implemented and this was fixed. Version: 1.0.2 Category: BUG FIXES Text: Made change to accumulation code to correct error introduced in last version with fixed parameters. Version: 1.0.2 Category: BUG FIXES Text: Made fixes to js to accommodate accumulation. Version: 1.0.1 Category: NEW FEATURES Text: Added use of sparse matrices for design matrices which sped up code and reduced memory consumption. The argument chunk_size was added to crm, create.dm,cjs, and js to control amount of memory usage. Version: 1.0.1 Category: NEW FEATURES Text: Added run timing and various print statements to track progress of model. If debug=FALSE, includes function evaluation counter (every 100) and neg lnl which remains on same line unless used in Rterm. Version: 1.0.0 Category: NEW FEATURES Text: Extracted crm and accompanying code from RMark and created initial base package.