1.5.1

BUGFIXING

• Bug in formulaNLR() when type = "b" fixed.
• Bug in estimNLR() when method = "plf" fixed.
• Bug in NLR() calculating SEs when no item is converged fixed.
• Bug in re-calculation of starting values in the NLR() function was fixed.
• Bugs in coef.difNLR() when convergence issues are present were fixed. Thanks to @netique .

MAJOR UPDATES

• Testing with the testthat package has been started.

MINOR UPDATES

• References updated.
• Warnings and messages in the NLR() and difNLR() functions were updated.
• Documentation of the coef.difNLR() was updated.
• Some typos were fixed.
• README file was updated.
• Description of the package was updated.

1.5.0-1

BUGFIXING

• Bug in re-calculation of starting values in the NLR() function was fixed.
• Bugs in coef.difNLR() when convergence issues are present were fixed. Thanks to Jan Netik.
• Bug caused by mixing two parametrizations in predict.difNLR() was fixed.
• Bug in the calculation of starting values with startNLR() was fixed.

MAJOR UPDATES

• Options "em" and "plf" were added for the method argument in the estimNLR() function to estimate item parameters with either the EM algorithm or algorithm based on the parametric link function (PLF).
• Options for the parameterization argument of the formulaNLR() and startNLR() functions were updated (renamed).
• Argument constraints were added to the startNLR() function.

MINOR UPDATES

• Warnings and messages in the NLR() and difNLR() functions were updated.
• Documentation of the coef.difNLR() was updated.
• Some typos were fixed.
• README file was updated.
• Description of the package was updated.
• References were updated.

1.4.2

BUGFIXING

• plot.ddfMLR() now correctly plots ordinal data.
• Option test = "W" was fixed for the difNLR() and NLR() functions.
• Constraints handling was fixed for the difNLR() and NLR() functions.

MINOR UPDATES

• startNLR now handles missing values. Returns error when not enough complete observations are provided.
• Arguments in ggplot2 plotting methods were updated to follow changes in the ggplot2 package.
• Linetype and its legend appearance in ggplot2 plotting methods were updated.
• ggplot2 v.3.4.0 is now imported.
• Examples for the difORD() and ORD() functions were updated. Now using the Anxiety dataset from the ShinyItemAnalysis package.
• Some typos fixed.
• class handling was updated.

1.4.0

BUGFIXING

• Option parameterization = "logistic" was fixed in the formulaNLR() function.
• Bug in plotting empirical probabilities in the plot.difNLR() fixed.

MAJOR UPDATES

• Iteratively reweighted least squares algorithm was implemented in the difNLR(), NLR(), and estimNLR() functions for 2PL model.
• coef.difNLR(), coef.difORD(), and coef.ddfMLR() methods now include delta method for IRT and logistic parameterizations.
• coef.difNLR(), coef.difORD(), and coef.ddfMLR() methods now include calculation of confidence intervals.
• predict() method for the difORD object was implemented.

MINOR UPDATES

• Some typos fixed.
• Examples in functions were updated.
• References were updated.
• Output of the estimNLR() function is now unified via the print() method.
• Packages CTT, grDevices, methods and reshape2 are no longer imported.

1.3.7

MINOR UPDATES

• doi in help pages updated.

1.3.6

BUGFIXING

• Empty factor levels were removed from Data in ddfMLR() to fix bug
  when plotting.

MAJOR UPDATES

• Sandwich estimator for covariance matrix in case that method = "nls" was
  implemented into the vcov() method for the output of the estimNLR()
  function.
• Wald test for implemented for the difNLR() function.
• Sandwich estimator for covariance matrix in case that method = "nls" was
  implemented into the difNLR() function via an argument sandwich = TRUE.

MINOR UPDATES

• doi in DESCRIPTION file updated.
• CITATION file updated.
• References were updated.

1.3.4

BUG FIXING

• Error when covariance matrix cannot be computed for some items in difNLR() function
  was fixed.

MINOR UPDATES

• NEWS and README files were updated.
• CITATION file was updated. Paper discribing the package was accepted for publication
  in The R Journal.

1.3.3

BUG FIXING

• plotDIFLogistic() now correctly plots matching criterion when item purification
  is applied.

MAJOR UPDATES

• Use of DIF matching criterion other than Total score is now enabled in shiny app
  in DIF sections.
• Downloadable tables are newly provided in shiny app in DIF sections.
• Iterations of purification process are now displayed in downloadable tables
  in the DIF sections of shiny app.
• DIF method comparison table is provided in DIF section of the shiny app.

MINOR UPDATES

• Help page for the ShinyItemAnalysis package was added.
• NEWS file is now formatted using markdown.
• On attach message was updated.
• Wording, Figures and page structure were updated in shiny app in DIF sections,
  histogram of total scores now better compares the groups.

1.3.2

BUGFIXING

• Legend in plot.difORD() is now correctly displayed.
• Method plot.difNLR() now correctly uses matching criterion when item purification is applied.
• Mismatch in null and alternative models was fixed for all functions.
• MLR() function now returns correct value of log-likelihood for alternative model.
• Function MLR() was fixed for binary data and IRT parametrization.
• Typo fixed in print.difORD() method.
• Method plot.ddfMLR() was fixed for binary data.

MAJOR UPDATES

• Function ddfORD() was renamed to difORD().
• Default option of argument type in difNLR() and NLR() functions was set to "all" instead of "both".
• Seed was added for re-calculation of bootstrapped initial values in NLR() estimation.
• Input Data in difNLR() function can be also a vector now.

MINOR UPDATES

• NEWS file is now generated using markdown.
• README file was updated.
• Function genNLR() with an option itemtype = "nominal" returns nominal items as factors with levels presented by capital letters.
• Legend for plot.ddfMLR() was updated to show P(Y = option) instead of option alone.
• Argument item for S3 methods of difNLR class can be now name of the column in Data.
• Legends in plot.ddfMLR() and plot.ddfORD() were updated.
• Package styler was used to improve formatting of the code.
• Package ShinyItemAnalysis was added into Suggests.
• Figures for README file were updated.

DOCUMENTATION

• Documentation of the package was updated.
• Some typos were fixed.
• Authors' details were updated.
• Documentation for all data was updated. Format of data was fixed.

1.3.0

BUG FIXING

• logLik.ddfMLR() now works properly.
• Anchor items are now used for calculation of matching criterion in plor.difNLR(), plot.ddfORD() and plot.ddfMLR()
• Check for input 'item' in S3 methods for difNLR(), ddfMLR() and ddfORD() was fixed.
• plot.ddfMLR() now handles also binary data.
• ddfMLR() returns consistently "No DDF item detected" when no DDF item was detected.

MAJOR UPDATES

• New functions ddfOrd() and ORD() for DDF detection for ordinal data with adjacent and cumulative logistic regression models were added. S3 methods plot(), print(), coef(), AIC(), BIC(), logLik(), residuals() were implemented.
• genNLR() now generates ordinal data using adjacent category logit model with argument itemtype = "ordinal".
• Names of reference and focal group in plot.difNLR(), plot.ddfMLR() and plot.ddfORD() can be changed with 'group.name' argument.
• IRT parametrization now available in ddfMLR() and ddfORD() functions with argument 'parametrization'. SE calculated with delta method.
• S3 methods plot() outputs for difNLR, ddfMLR and ddfOrd were unified.
• Matching criterion for plot.ddfMLR() was improved for displaying more smooth curves.

MINOR CHANGES

• Some typos fixed.
• Authors' details were updated.
• CITATION file was updated.
• Matching criterion for plot.ddfORD() uses anchor items.
• Help pages for difNLR(), ddfMLR(), ddfORD(), MLR() and ORD() functions were updated.
• All static DOI links were updated.