NEWS

DAAG 1.25.5

Add alc2018 dataset supplementing grog dataset with data for the years 2005 to 2018.
The qgam package, with functions qgam (single quantile) and mgam (multiple quantiles) that allow for automatic selection of smoothing parameters when fitting quantiles, replaces quantreg in vignette simulate-varselect and in function bsnVaryNvar().

Add repPsych dataset holding summary results from Reproducibility Study: Psychology
The function DAAGtheme() has been modified to left adjust main titles and to use font 1.
Correct stated number of rows in the frostedflakes dataset to 100.

A number of URLs have been updated.
The bomregions2021 dataset updates the bomregions data to 2021.
The dataset mdbAVtJtoD, with monthly Murray-Darling basin data from January 1950 through to February 2022, has been added.

Vignettes sources are now R Markdown, with the format bookdown::html_document2 format combined with base_format: rmarkdown::html_vignette (This combination allows automatic numbering of figures, and cross-references.)
The 'bomregions2018' dataset updates 'bomregions' to include data through to 2018. The updates 'bomregions2012' and 'bomregions2015' have been omitted. A further column 'avDMI' has been added, holding values (from 1960 onwards) of the Annual average (Indian) Dipole Mode Index.
Web links on what is now the 'bomregions2018' help page have been updated. The url in the DESCRIPTION file now links to the development version of DAAG on the gitlab site.
The 'head.injury' dataset (which was a copy of the 'headInjury' dataset) has been removed.
Help pages for related data sets have, in several cases, been combined.

The dataset coralPval, holding 3072 p-values from a gene expression array comparison between post- and pre-settlement coral larvae, has been added.
The dataset bomregions2015 (updates bomregions to include data up to 2015) has been added.
The dataset lognihills, which replaces values in all 4 columns of nihills by log(values), has been added. Column names are logdist, logclimb, logtime and logtimef.
The generic function sumry(), with a method and print method for glm (but not other) objects, has been added.
The function DAAGtheme(), used to generate themes for use in an intended 4th edition of Data Analysis and Graphics Using R, has been added.
The functions CVlm() (alias cv.lm()), simulateLinear(), qreference() and sampdist() all now return lattice objects. If the result is returned to the command line, the graph is plotted just as before.

A line that subtracted 1 from the bottom and top margins, causing an error if the margin was already set to < 1, was removed from onetPermutation(), twot() and twotPermutation().
Code has been amended so that term labels ('term.labels') are given by attr(obj$terms, "term.labels"). The code now replaced expected to find 'term.labels' among the attributes of the summary object; this would have failed starting several R releases back.

The dataset hurricNamed (deaths from Atlantic hurricanes, damage assessments, and other measures, from 1950 onwards) has been added.
The dataset cps3 (a subset of cps10 has been added
The help file CVlm.Rd has been amended to give the correct return value for CVlm() and cv.lm().
An enhanced version of overlapDensity() (an alias is overlap.density()), as assumed by the code for Figure 13.3 on page 421 of the 3rd edition of "Data Analysis and Graphics using R", has replaced the former version.

The function align2D() has been added. This finds the linear transformation which, applied to one set of points in the (x, y) plane, gives the best match in a least squares sense to a second set of points.

The functionality of lmdiags(), called from plotSimDiags(), has been slightly improved.

There are improvements to the way that dependencies are handled.
The file simulate-reg.pdf has been replaced by two knitr vignettes -- 'Functions for Assessing the Sampling Properties of plot.lm() Regression Diagnostics', and '{Functions that Demonstrate the Sampling Properties of Variable Selection'.

The function plotSimScat was added. This plots either residuals or y-variable simulated values, against the x-variable from a straight line regression.
The datasets tomato and bomregions2013 were added.
A pdf file simulate-reg.pdf has been placed in the subdirectory inst/doc. This demonstrates the use of the functions plotSimDiags, plotSimScat and bsnVaryNvar.

The NAMESPACE file has been edited to ensure that the function layer, used in plotSimDiags and plotSimScat, is taken from the latticeExtra package.

The function plotSimDiags was added. This returns diagnostics for multiple sets of simulated data.

The function CVlm (alias cv.lm) has been modified to allow multiple explanatory variables. Predicted values can be subtracted off from the points that are shown. If there is more than one explanatory variable, the lines that are shown for individual folds are approximations to the small points that show the fitted values for the respective fold. Arguments main (main title) and legend.pos have been added.

the function bsnOpt has been added - this finds the regression model that is optimal according to the stated criteria
the functions bestsetNoise, bestset.noise, bsnCV, bsnOpt, and bsnVaryNvar all allow the optional logical argument intercept. Except for bsnVaryNvar, the user now has the option to supply arguments X and y.
the function lmdiags and the generic function plotSimDiags with a method for lm models have been added. The function takes as argument a model object, then returning a lattice graphics object or objects that show regression diagnostics for simulated data

In the dataset possumsites, the first two column names have been corrected to "Longitude" and "Latitude". The code under 'Examples' has been changed accordingly, so that the map that appears is the same as before.