Old Changes

April, 1998

1.    Making the PIM Chart interactive, so that you can edit the PIM's by dragging blocks of parameters, or by right-clicking on a block and changing the PIM.   This feature makes for rapid editing, and also allows editing of very large PIMs without actually opening up the PIM Window and all the edit boxes.

2.    The design matrix now allows you to paste in a block of values contained in the clipboard.  Either use the Ctrl+V command, or else select the Paste Clipboard command.

3.    The robust design model now has a saturated model value computed, although I'm not quite sure its correct.  However, without a saturated model value, the deviance was computed as zero, so that likelihood ratio tests were not possible.   Turns out that some data sets still have a zero deviance, meaning that the saturated model deviance is not correct.  The AIC values are okay, just the deviance is wrong.

4.    Barker's model has been thoroughly checked, and seems to work correctly.  Note that the model has been re-parameterized compared to earlier versions.  See the Help file for details.

May, 1998

5.    Because the identity design matrix is the default, I made the program "smarter" so that an actual matrix does not have to be passed between the interface and the numerical analysis program, and not stored in the output.   Instead, the key work "identity" is added to the design matrix statement.   No design matrix is printed in the output.  More importantly, no design matrix pops up on the screen when you retrieve a model that had the "identity" key word on the design matrix statement.

6.    I have modified the calculation of the threshold value for declaring a parameter not estimable.  The old threshold value was 0.3E-6.  The new value varies depending on the dimension of the variance-covariance matrix.  Thus, some models may give a different number of parameters estimated than previously.  If you find an error, let me know so that I can continue to refine this process.

June, 1998

7.    Seven new models were added to the list of data types:

Huggins (1991) method where the closed captures likelihood is conditioned on only the animals captured has been added.  This model allows individual covariates to be used for estimating the initial (p) and recapture (c) probabilities of the closed capture model.  Estimates of population size (N) are generated as derived parameters, because N does not actually occur in the likelihood with Huggins' model as it does in the closed captures models of Program CAPTURE.   Estimates of N are asymptotically equivalent to estimates of N from the closed captures models of Program CAPTURE, but are not quite as efficient because some information is lost by conditioning the likelihood.

Huggins (1991) method was also added to the robust design model, so that individual covariates can also be used with all the paramters.

The Jolly-Seber model has been added.  The first parameterization is from Ken Burnham, and uses the first population size N1 and the rate of population change, lambda1 through lambda(k-1) for k encounter occasions.   This model does not converge readily, so I suggest you use one of the Jolly-Seber models described below.

Pradel's (1996) model that estimates seniority from the encounter histories has been added.

Pradel's (1996) model that estimates both seniority and apparent survival from the encounter histories has been added.  This model is another parameterization of the Jolly-Seber model.

Pradel's (1996) model that estimates apparent survival and rate of population change (lambda) has been added.  Again, this model is another parameterization of the Jolly-Seber model.

Pradel's (1996) model was reparameterized to include apparent survival and fecundity (number of new individuals in the population at time i+1 per individual at time i).  Again, this model is another parameterization of the Jolly-Seber model.

8.    The Graphics Server graphics package was added as the default graphics package.  The reason for this second graphics package was to provide better graphics when multiple sets of parameters were displayed.  Both packages are available via from the Results Browser Window with the Output | Specific Model Output | Interactive Graphics and the Output | Specific Model Output | Interactive Graphics GS menu choices.

May, 1999

9.    The variance components code was changed to use the full precision of estimates for the variance-covariance matrix.  You will have to recompute parameter estimates of existing models to use the new code.

10.    The older graphics package was eliminated.  This change cut the size of the program distribution files.

11.    An option to re-compute all the models in a Results Database was appended to the Run menu of the Results Browser.

June, 1999

12.    PIM matrices can now be copied to the clipboard, and vice versa the clipboard can be pasted into a PIM.  This feature also means that you can create PIMs in a spreadsheet, and then paste the values into a PIM Window (or also on the PIM chart).

13.    An "All Different Matrix" option has been added to the PIM menus to allow you to create PIMs with every value different.

14.    The Full menu choice has been changed when you request a Design Matrix. Now, the Full menu choice produces a design matrix with an intercept, group effects (if needed), time effects, and group*time effects (if needed).  The limitation of the Full option is that the PIMs must have the number of parameters equal to a full g*t model.  However, you can create age structure across PIMs, and still use the Full menu choice as long as the number of parameters is equal to a full g*t model.   The columns of the design matrix are initially labeled, although these labels disappear if you manipulate the matrix, e.g., add or delete columns.  I'm thinking about how to keep the labels around.

    The old version of the Full menu choice has now been renamed to be a the Identity menu choice.

15.    The list of Pre-Defined Models has now been expanded to include both PIM and Design Matrix coding of the standard models.  In addition, the Design Matrix coding list includes g+t models.  All the Design Matrix models are constructed based on a full g*t set of PIMs.  Hence, you still can't use the Pre-Defined Models option to construct age-structured models.

16. A menu choice has been added to the Output menu to list the m(ij) arrays of live recaptures and dead recoveries of the encounter history matrix in a NotePad window.  This option is useful for identifying anomalies in the data. In the past, you could only run RELEASE for the CJS model to look for anomalies.

17. Variance components outputs are further developed, with additional options specified by check boxes for output other than the main numerical output and the graph of the estimates.

18. The problem with a semicolon in the encounter histories file between comment delimiters, e.g.,

        /* stuff; stuff */,

has been fixed.

19. Problem behavior with the individual covariate standardization method has been uncovered.  When you specify a model with a common a common intercept but 2 or more slopes for the individual covariate, e.g.,

            1 weight 0

            1 0 weight

and specify to standardize the individual covariate, you will get a different value of the deviance than from the model run with unstandardized individual covariates.  This behavior is because the centering effect of the standardization method affects the intercept differently depending on the value of the slope parameter. The effect is caused by the nonlinearity of the logit link function.  You get the same effect if you standardize variables in a logistic regression, and run them with a common intercept.   The result is that the estimates are not scale independent, but depend on how much centering is performed by subtracting the mean value.

20.  I think I have gotten rid of the problem that caused the message "Unique ID > 8000.  Tell Gary".  If you still get this message with the new version, let me know.

July, 1999

21.  The ability to open one of the last 4 files previously used has been added to the File menu.

22.  Real parameter estimates are now labeled as "Estimates" instead of "S(I)" in the output.

23.  An option was added to make the default to tell you when a model is retrieved.  With the File | Preferences menu choice, you can change this option back to not making you respond to this message.

24.  The option to list the Akaike Weights in the Results Browser has now been made the default.  You can turn this option off with the File | Preferences menu choice.

December, 1999

25.  A problem with losses on capture in estimating p and c of the robust design and Huggins robust design models has been fixed.  Previously, S, gamma'', and gamma' were estimated correctly with losses on capture, but not p and c, and hence N.

26.  Options for export of parameter estimates (both beta and real), and SE's and confidence intervals to Excel spreadsheets or the clipboard were added.

27.  Options for export of variance-covariance matrices to dBase files or the clipboard were added.  These changes forced a re-organization of the menu choices under Output | Specific Model Output to keep the menu from becoming too clumsy.    Also, the model averaging variance-covariance matrix can now be exported to a dBase file.

28.  Extensive testing of the Jolly-Seber model demonstrates that it works correctly, but is just difficult to achieve convergence with for sparse data.  With simulations and large sample sizes, it works fine.

29.  More work was done on the simulation procedure.  However, I still have not removed the double-click trap to keep the faint of heart out of the simulation procedures.  I now believe that all models are simulated correctly except the robust design,  Huggins robust design, and Pradel model with recruitment but no survival.   Currently, you are not allowed to simulate these models.  All the rest should work correctly, including the Jolly-Seber model.  The problem with simulating the robust design models is how to specify the poulation size so that it is not less than what is consistent with the S, gamma'', and gamma' parameters for the previous interval and preceding N.  This problem involves fixing the input interface, not the actual numerical simulation procedure.

March, 2000 (updated to Version 1.7)

30.  A "robust design" extension of the Barker model is now included in MARK.

31.  A VPA - Virtual population model - is now included.

32.  The ability to submit simulations to Program RELEASE is included in the Simulation menu.

33.  The ability to submit simulations to Program CAPTURE is included in the Simulation menu.

34.  The design matrix now displays different colors for cells with zero, one, and other values. This feature is handy for spotting the pattern in complex matrices. You can change the colors displayed with the File | Preferences menu choice.

35.    The ability to run more than one data type within the same Results Browser window has been implemented.  For example, you can now run both the dead recoveries {S, r} models and the Brownie et al. {S, f} models, and have the results appear in the same Results Browser window.  To change the data type for models, select the PIM | Change Data Type menu choice from the Results Browser window.  This convenience also introduces some problems, such as model averaging (not all the models have the same PIMs), so that the model averaging procedure only averages across the models with the same data type.  Another example of where problems occur is when the likelihoods are very different between the data types, e.g., the closed captures models and Huggins' models.  The AIC values are not on the same scale for these different data types.  However, this feature is very useful to compare the 2 sets of estimates from the same Results Brower window.  With this change, I updated the version number of MARK to 1.7.

36.    An option to use multi-threading was added to the File | Preferences menu choice.  This option allows you to be building a new model while numerical estimation is running for a previous model in background. 

37.  An option was added to the Run Window that specifies to use the mean values of individual covariates to compute the real parameter estimates reported in the output.  Note that when individual covariates are standardized, the mean value for a standardized covariate is zero.  

June, 2000 (updated to Version 1.8)

38.  The "robust design" extension of the Barker model was modified to include the permanent emigration (F) parameter, and the notation for the temporary emigration parameters was changed to be gamma'' (old F) and gamma' (old F') to be consistent with Lindberg et al. (In Press).

39.  Calls to Programs ESTIMATE and BROWNIE were added to the Test menu choice in the Results Browser window. The programs are useful in assessing goodness-of-fit for dead recovery data.

40.  The capability to "undelete" a model that was deleted from the Results Browser window was added. This menu choice is available under the Delete menu of the Results Browser window. You are presented with a list of all the models you have deleted during this execution of MARK, and allowed to select the models you want to undelete. However, once you end MARK, the deleted models permanently disappear.

41.  A "Partial Zero" menu choice was added to the Design Matrix window to allow the user to "clear" or zero out a portion of a column. This option is useful when you copy a column, and only want to retain a portion of the values in the column.

42.  The variance components output was modified to include a naive estimator of the process variance. Additionally, the estimate of the process variance and associated confidence intervals are now reported with negative values, if necessary. However, the estimate of the process standard deviation is reported as zero for negative process variances (and similarly for the lower confidence bound).

43.    A new column has been added to the Results Browser -- Model Likelihood.  The values in this column are the Akaike weight of the model in the row divided by the Akaike weight of the minimum AICc model.  This value is the model likelihood, the strength of evidence for the model.  The best model has a value of 1, with all other models <1.

44.    The Variance Components module now has a check box to run the Random Effects model that corresponds to the variance components analysis you have requested.  Specifically, a Random Effects model can be viewed as intermediate in its number of parameters between the time-specific model (t) and the dot (.) model where all values are the same.  The approach developed by Ken Burnham is to fix the parameter values that were used to estimate the process variance to their S-tilde values from the variance components  analysis.  The number of parameters that this set now represents is estimated from the trace of the G matrix.  Adding trace(G) to the number of parameters estimated from the numerical analysis run with fixed values from the S-tilde gives the number of parameters estimated for the random effects model.  Because trace(G) is not an integer, the number of parameters for the random effects model appears as a non-integer.  To accommodate these non-integers, existing Results Files will be updated automatically the first time they are accessed.  You should see a message stating so.  To be safe, you may want to back up your MARK files before opening them with this new version.

45.    Although the background colors used in the design matrix to represent the pattern of the matrix are very helpful, especially at first, the cost is a lot of time watching design matrices being created on the screen.  Therefore, I put an option in the File | Preferences dialog to turn off the color in design matrices if you desire, which definitely increases speed for large matrices.

46.    I fixed a bug with the probability of surviving the study (Kaplan-Meier estimate) for the Known Fate data type when time intervals were unequal.  Previous versions assumed all time intervals were equal to 1.  Now, the length of the time interval is used to compute the product of survivals and its standard error and confidence intervals.

47.    Labels for the real parameter estimates are now provided in the output.  Labels for real parameter estimates are taken from the "Parm" column that is added to the design matrix -- giving you some idea about what the parameter type is for each estimate.  Note however, that some parameter types can be set to others, e.g., initial capture probability (p) and recapture probability (c) in the closed capture models.  Hence, be careful to not be mislead.  You do not need a design matrix to obtain these labels -- the process is automatic.  Labels are generated from the parameter type for the first occurence of the parameter index in the PIMs.  If you want to obtain these labels for existing model output, you need to re-run the model, i.e., from the Results Browser, Run | ReRun Models menu choices.  The real parameter estimate labels are limited to 20 characters.

48.    Labels for the beta parameter estimates and columns of the design matrix are now available.  When in the design matrix, if you right-click the matrix, the menu choice "Label" allows you to enter 20 characters of information to include in the column heading of the column with the highlighted cell.  If you start with the Design | Full menu choice, the headings in the design matrix will be automatically labeled.  Likewise, the pre-defined design matrix models also have the columns automatically labeled.  When the model is run, this information is passed to the numerical estimation process, and these labels will appear beside the beta parameter estimates.  When models are retrieved, the design matrix column headings also are retrieved from the output.

49.    Columns in the design matrix can be moved by dragging the column heading to a new position.  The new ordering can be preserved by right-clicking the design matrix and selecting "Reorder Columns" from the pop-up menu.  Basically reordering columns is equivalent to renumbering the beta parameters.

August, 2000

50.    A bug introduced in June concerning parsing of real parameter estimates from the output file when derived parameter estimates were present was fixed.  This bug was detected because of incorrect output in the model averaging procedure when derived parameter estimates were present. 

September, 2000

51.    The computer time required to process individual covariates has been dramatically reduced, at least 10 times.

52.    A bug with the naive variance component estimator has been fixed -- previously, 2 consecutive runs of the variance components module would give incorrect estimates for the naive estimator for the second and additional runs.

January, 2001

53.    The variance components estimator has been modified to show estimates of sigma^2 < 0, then to set the estimate of sigma to zero, and re-compute the beta-hats and their standard errors with sigma = 0.

54.    Likelihood ratio tests when c > 0 have been modified to incorporate the quasi-likelihood parameter.  Likewise, the deviance is now displayed in the results browser is corrected for c.

55.    Design matrix window now has an option to set the font size, so that large matrices can be miniaturized and the colors used to determine if the pattern is correct.

56.    Derived parameters are now handled separately from the real parameter estimates.  Eventually, derived parameters will be a fully-supported third parameter type, and you will be able to graph them and perform variance components.

May, 2001 (updated to Version 2.0 -- Greater Prairie Chicken photo)

57.  Two additional data types for closed captures have been added. Two heterogeneity models for closed captures have been created based on mixture distributions. The simple version of this model only has a constant capture probability for all occasions and for recaptures. The full heterogeneity model has a set of p and c parameters for each mixture. Encounter histories for these 2 models are identical to the closed captures data type. These models required a fairly major change to the PIM structure to allow for large rectangular PIMs.

58.  A model for nest survival estimation was added. This model is different from the known fate data type because the exact time of failure of the nest is not known. Input for nest survival data can be in the form of a nest survival data type, where the time of first finding the nest, next to last time the nest was observed, and the last time the nest was observed are entered, followed by the fate of the nest (0=successful, 1=unsuccessful). Individual covariates may also be used. More details on the structure of this model and the encounter histories file can be found in the help file.

59.  A bug that caused incorrect standard errors for the derived parameter estimates of N was fixed in the Huggins robust design model.

60.  Some really large problems (e.g., very large PIMs or a large number of PIMs) will not work correctly with the existing model averaging parameter selection interface. Therefore, I added another interface that will work. To use this new interface, click the "Use non-interactive model averaging parameter specification window" choice from the File | Preferences menu choice. With this interface, you have to pick the PIM and then specify the parameter within the PIM to model average. Not as visual, but more effective for large problems.

June, 2001

61.  A bug in computing the variance of the derived parameter estimate N-hat for the Huggins data type was fixed.  Standard errors of N-hat from the Huggins models now correspond closely with standard errors from the closed captures data types.

62.  The scaling of the gamma'' and gamma' for the length of the interval was removed.  These parameters should be viewed as the probability of being available for capture conditional their previous status, and hence are not a function of the length of the time interval.

63.  When setting up a new analysis, you can now paste in time interval lengths from the clipboard.  Thus, if you have a complicated set of time intervals, enter them into a spreadsheet, and then paste the values into MARK.

64.  A "Save As" capability was added to the File menu choice to save a copy of the current Results File.

65.  The dialog to select a set of "Pre-defined Models" has been re-written to allow the user to pick models for each parameter, rather than select from a list of all possible models.  This change allows pre-defined models for data types that previously would generate too long of potential models, allows the user to more easily select just models desired, and makes it harder to "Select All" models.  In other words, rather than shamelessly data dredging with a few mouse clicks, you will now have to work at it!

66.  The ability to summarize derived parameters from simulations was added.

67.  Several bugs that caused incorrect estimates for the nest survival data type were fixed.

July, 2001

68.  The nest success data type was further modified to produce warnings when invalid data are entered for a nest record.  The help file under the 'Nest Survival' topic was updated to present examples of these invalid cases.

September, 2001

69.  The multi-strata model with live and dead encounters is now working correctly after another bug was fixed on 3 September, 2001.  The help file has also been updated to describe this model.

November, 2001

70.  The occupancy estimation model of MacKenzie et al. (2002) has been implemented.  The help file has also been updated to describe this model.

January, 2002

71.  Minor fixes to several bugs that have surfaced, including aborts when no pre-selected models are selected, incorrect label on beta labeling window, and specification of default preferences when the program is installed.

March, 2002

72.    The Run Window has been modified to allow a user to specify values of individual covariates to be used in computing the real parameter estimates.  With these changes, 3 options are available for specifying values of the individual covariates: values from the first encounter history in the encounter histories file, mean values, and values the user specifies.

73.    Paste buttons were added to the simulation input screens for beta parameter estimates and numbers of releases, and initial input screens for group labels, individual covariate names, and strata names and labels.  Now, the user can generate a set of input values for any of these screens in a spreadsheet, and then paste the values into MARK.

74.    The Run Window has been modified to allow a user to specify different link functions for parameters in a model.  However, this procedure is tricky, and can lead to numerical problems if not done intelligently.  Default initial parameter values are more difficult to set, so numerical convergence becomes more of an issue.  The user will have to take more responsibility in specifying initial parameter values if multiple link functions are used in one model.

September, 2002 -- Version 3.0

75.    The robust-design occupancy estimation model of MacKenzie et al. (submitted) has been implemented.

76.    An updated Graphics Server graphics package has replaced the older version.  The interface to the graphics package now includes legends for each set of parameters plotted.

77.    Some minor bugs with labeling columns in the design matrix, PIM Charts off the top of the screen, and some other little annoyances that have been pointed out have been fixed.

78.    The capability to construct interaction terms between individual covariates has been implemented with the product function in the design matrix.  In addition, an add function has also been implemented.  See the revised help file for details of how these functions operate within the design matrix.

79.    The Pearson chi-square statistic is now computed routinely in the full output, and can be saved in the simulation and bootstrap files.  The Pearson chi-square statistic is sometimes better behaved than the deviance statistic, and may provide an improvement in estimating the over-dispersion parameter, c.

November, 2002 -- Version 3.1

80.    An entry in the design matrix menu now allows you to see the list of individual covariates available for the current data set, and to insert 1 or more of these individual covariate names in the current highlighted cell in the design matrix.

81.    Added the capability to retrieve one or more columns from a previous model's design matrix and insert them into the current design matrix.  Note that the number of rows in the design matrix has to be the same for both design matrices.

December, 2002

82.    An option was added to the Run Window to allow estimation of profile likelihood confidence intervals for real parameters.  However, numerical problems can cause the estimates to be wrong, so use this option carefully.  The help file explains the procedure, and shows an example of incorrect profile likelihood confidence intervals.

March, 2003

83.    The multinomial logit (MLogit) and cumulative logit (CLogit) link functions were added as options under the Parm.-Specific link function option of the Run Window.  The MLogit link is particularly useful for the multi-strata model to constrain the transitions from a strata to sum to <=1.  Another application is the probability of entry parameter of the POPAN model (see below).  See the revised link functions help file write-up for details.

84.    The POPAN model was revised, and is now working correctly for the real parameter values.  This is the Jolly-Seber model from Schwarz and Arnason (1996), where recruitment to the population is modeled as the probability of entry (pent in MARK, beta in the original paper) from a super population.

85.    Problems with convergence of the robust design occupancy model were corrected, plus 2 additional parameterizations of this model were added.  See the help file for details.

86.    An additional parameterization of the Jolly-Seber model developed by Link and Barker (submitted to Biometrika) was added.  To use this parameterization, start with one of the other Jolly-Seber data types (e.g., Burnham's Jolly-Seber data type, one of the 3 Pradel data types, or the POPAN data type), and use the PIM | Change Data Type menu choice to obtain the Link-Barker parameterization.  Details are in the help file.

June, 2003

87.    Fixed a bug in the Huggins closed-capture heterogeneity models so that N is correctly estimated.

August, 2003 -- Version 3.2

88.    There are 6 different data types for estimation under the closed capture models: Mt (p and c), Mh (pi and p) and full Mh (pi, p, and c); with each of these data types available with population size (N) in the likelihood, or a Huggins version where the likelihood does not contain N.  Now, when the Closed Captures data type is selected, you are given a choice of which of these models of capture probabilities to use for your initial model.  However, you can select different data types with the Change Data Type menu choice that is available under the PIM menu.  Note that the likelihood for the Huggins models is not comparable to the likelihood for the data types that include N in the likelihood, so that AIC model selection should only be done within the 2 categories, not between them.

89.    These same 6 closed models are now available for the robust design model, as well as the robust design version of Barker's model.  When you select either the robust design, or the Barker robust design, you are asked to select one of the 6 closed models.  However, you can change the closed model with the Change Data Type menu choice.

90.    The robust design multi-strata model has been added to MARK.  The open model of Kendall and Bjorkland (2001 Biometrics 57:1113-1122) has been added, along with a closed version that can operate with any of the 6 closed models described above.  When you select the closed robust design multi-strata data type, you are asked to select one of the 6 models, but you can change between them with the Change Data Type menu choice.

91.    A new version of the installation program was used that has caused some difficulty, but seems to be working correctly.

April, 2004

92.    The derived parameters from the Huggins models (population size) can now be used in model averaging and variance components.  Huggins models can be used with closed capture data, robust-design data, and robust-design multi-strata data.  For robust-design multi-strata data, a population estimate is generated for each strata at each primary occasion.  To allow derived parameters to be used in the model averaging and variance components analyses, the variance-covariance matrix of the derived parameters must be computed, and so this matrix is also available in either a notepad window or else an Excel spreadsheet.

93.    For all of the multi-strata data types, including the robust-design multi-strata models, the transition probability (psi) obtained by subtraction can be selected with the Change PIM definitions menu choice.  Previously, the default value of psi was the probability of remaining in the strata (e.g., psi A to A).  Now, the user can select the transition probability to obtain by subtraction.  This change allows fixing the probability of remaining in the strata to zero, a model that was not possible in the previous versions of MARK.  One problem with this change is that care must be taken to only model average parameter values for psi PIMs that share the same definition.  More details on this capability can be found in the MARK help file.

94.    A new method of estimating the overdispersion parameter, c, has been implemented with the Median chat menu choice.  The approach is to simulate data for a range of c values, and perform a logistic regression to estimate the value of c for which the probability of obtaining a deviance c-hat greater than the observed deviance c-hat is 0.5.   More details on this procedure can be found in the MARK help file.

May, 2004 -- Version 4.0 -- Female Lark Bunting

95.    Numerous small irritations were fixed with this version (even though I'm sure you have many more that you would like to see addressed).

96.    The capability to run multiple multiple models simultaneously has been re-instituted into the program, no matter what operating system is being used.  Thus, you can start the numerical optimization of a model, and then begin preparing and then start of the optimization of additional models while the first one continues to execute.

97.    Markov Chain Monte Carlo (MCMC) estimation for all data types is now implemented, mainly to provide increased capability for estimation of variance components.  However, the code is for this estimation procedure is aimed primarily at modeling the beta parameters for models constructed with the logit link function, although inputs can be provided to accommodate other link functions for specific data types, such as the log link for population estimates (N).  The one or more hyperdistributions being modeled are thus generally on the logit scale.  However, considerably flexibility has been provided to model the hyperdistributions, including a design matrix on the means and a full variance-covariance matrix to estimate process covariances (actually correlations).  MCMC estimation is obtained by checking the MCMC check box in the Run Window.  Typically the parameter estimates from the ML procedure would be used as the initial parameter values for the MCMC procedure to ensure that the Markov Chain is started at reasonable values.  Output is displayed in a NotePad Window, consisting of the mean, standard deviation, mode, and median of the posterior distribution, plus the 2.5, 5, 10, 20, 80, 90, 95, and 97.5 percentiles.  Note that no output is stored in the Results Browser.  In addition, a binary file of the sample from the posterior distribution is provided, along with example SAS code in the help file for reading and summarizing this sample.  Full details on the MCMC estimation procedure are provided in the MARK help file.

June, 2004

98.    Bug with the parameter labels in the reparameterizations of the multi-strata data types was fixed.  This fix will not correct models that have already been run and are retrieved.  Rather, the model will have to be reconstructed to fix the incorrect labels.

99.    MCMC algorithm was modified to make the step size for the jumping distribution be adaptive.  After 100 steps, the percent of jumps is evaluated and a new value computed to try to obtain 40% of the steps making a new jump.

July, 2004

100.    Bug with the multi-strata model with unequal time intervals was fixed.  This bug was introduced during the update of April, 2004, when the modification to allow the user to specify the psi obtain by subtraction was implemented.

101.    All 6 of the closed captures models now work with the median c-hat estimation procedure, and can be simulated.  Note that for the closed capture models with N in the likelihood, the value of N is specified as the actual value, and the link function specified for the true model is ignored for the N parameters.

July, 2004 -- Version 4.1 -- Female Lark Bunting

102.    A new Visual Objects compiler was used to create the interface on July 14, 2004.  The new compiler uses a different set of DLL files.  Therefore, uninstall MARK and then install this new version.  The old executable files will not work with the new DLL files, and vice versa.  This new version seems to be more stable.  As usual, please let me know of any bugs you encounter.

103.    The simulation capability was added for the robust design models.  Unfortunately, specification of the inputs for these models is tricky, in that for the second and following primary sessions, you do not actually specify the population size, but rather the number of new recruits.  For the closed population models where N is in the likelihood, you must specify the size of the initial population and the numbers of new recruits as actual values, regardless of the link function for the rest of the parameters.  For the Huggins closed captures models, a separate tab window requests the initial population size, and the numbers of new recruits.

104.    AICc values were incorrectly computed for models with deviance equal zero when a new value of c-hat was applied.  This bug is now fixed, but will not correct the AICc values in results files that have already been affected by this problem.  To check to see if the AICc value is correct, recompute the model's estimates and verify that the old and new values of AICc are the same.  If not, then you will have to recompute the estimates for all the models to get back the correct AICc values.

August, 2004

105.    The ability to simulate multiple Markov chains was implemented in the MCMC estimation algorithm.

106.    The "gibbsit" algorithm to estimate samples sizes for Markov chains in the MCMC estimation algorithm was implemented.

December, 2004

107.    A bug with viewing the selection of the parameter to constrain in the multistrata models was fixed.

108.    The default value for the maximum number displayed in the PIMs was set to 9999.  However, memory limits on your machine likely will still limit the size of MARK models.

February, 2005

107.    A bug with fixed parameters in simulating median c-hat values was fixed.

April, 2005 -- Version 4.2 -- Female Lark Bunting

108.    The models for estimation of closed populations when there is mis-identification of animals developed by Paul Lukacs, Colorado State University, have been added to MARK.  These models were developed for use with DNA marking, when there is a "small" probability of the genotype of the animal being incorrectly determined because of low quantity or poor quality DNA.  The existing 6 closed captures models in MARK were all extended to include the alpha parameter (probability that the animal was correctly identified, with 1 - alpha the probability that the animal was incorrectly identified).  Thus there are 12 closed captures models now in MARK.  Because the population size is no longer included in the likelihood with the mis-identification models, all of the closed captures models now also report the population estimate as a derived parameter, so that model averaging and variance components analyses can be conducted across these different data types.  That is, model averaging is appropriate for the closed captures models with and without the mis-identification parameter is appropriate.  However, as before, the Huggins models and the models that include the population size (N) or more technically, the number of animals never encountered (f₀), do not have comparable likelihood values, so you should not try to perform model selection when both of these models are in the same results browser window.

109.    The 6 robust design data types were extended to incorporate the mis-identification parameter, so that there are now 12 different robust design data types.

110.    The 6 multi-strata robust design data types were extended to incorporate the mis-identification parameter, so that there are now 12 different multi-strata robust design data types.  Note that the Barker robust design data types have NOT been extended to incorporate the mis-identification parameter because the most likely place in these models where the mis-identification would take place is the resightings during an interval (coded as a "2"), and the theory has not been worked out to handle mis-identification with this type of encounter.

111.    A problem with retrieving the muti-strata models when the psi parameters had been redefined was fixed.  Previously, the redefinitions of the psi parameters was lost when a model was retrieved, so that the default of obtaining the probability of staying in the current stratum was obtained by subtraction was always the case.  Now, the redefinitions are maintained when the model is retrieved from the results browser.

112.    A problem with retrieving models with design matrices that used the POWER function has now been fixed.

August, 2005

113.    An additional derived parameter (gross births, B*) was added to 2 existing derived parameters (net births, B; population size, N) for the POPAN data type.  The capability to do model averaging and variance components for these 3 derived parameters was added.

114.    The estimation of the derived parameters for the known fate model (probability of surviving the entire interval) was cleaned up so that model averaging and variance components can be performed with this derived parameter.

November, 2005

115.    Modifications to the MCMC estimation code have been made: 1) the order of the beta parameters is randomized on each Markov chain iteration; 2) -2log likelihood is now stored and reported in the output, including the MCMC.BIN file, 3) the deviance information criterion (DIC) is reported in the output; 4) a "tuning" phase has been included prior to the burn-in phase to adjust the standard deviation of the proposal distribution; 5) the standard deviation of the proposal distribution is no longer changed during the iterations where the parameter values are sampled; and 5) values of the SAS macro variables to process the MCMC.BIN file are printed at the bottom of the MCMC output, and this output block can be copied and pasted into the SAS code to process the MCMC.BIN file.

116.    A problem with reading encounter history file records longer than 256 characters was fixed.  Encounter history records can now be up to 2048 characters.

December, 2005

117.    The open robust-design multistrata data type has been modified to now show the phi parameter as an upper-triangular matrix.  This modification means that age models can be constructed, i.e., models that recognize the time since the animal first entered the primary session.  Bill Kendall has developed this model based on the original Kendall and Bjorkland (2001) paper: Kendall, W.L. and R. Bjorkland. 2001. Using open robust design models to estimate temporary emigration from capture-recapture data.  Biometrics 57(4): 1113-1122.  Unfortunately, with this change in the PIM structure, you can no longer recall the previous models without getting errors.  If you are using this data type currently, you may not want to update your version of MARK until you finish your current analysis.  Additional derived parameters will be added to this data type in the near future to provide estimates of residency time and population size.

118.    A labeling bug was fixed with the variance components module when a user-specified design matrix was used.

119.    A bug when simulating models with fixed parameters was fixed.