NEWS

        CHANGES IN spatstat.core VERSION 2.4-4.010

OVERVIEW

    o Internal improvements.


        CHANGES IN spatstat.core VERSION 2.4-4

OVERVIEW

    o Bug fixes and minor improvements.

BUG FIXES

    o rhohat.ppp
    The argument 'subset' was not handled correctly in the internal data.
    The estimated function 'rho' was correct, but if 'predict.rhohat' was applied,
    predictions were computed only in the 'subset', and were possibly incorrect values.
    Fixed.

    o Gfox, Jfox
    Warnings were issued about text formatting errors (mentioning 'sprintf' or 'fmt').
    Fixed.
    
        CHANGES IN spatstat.core VERSION 2.4-3

OVERVIEW

    o We thank Art Stock for contributions.
    
    o Bug fixes and minor improvements.

BUG FIXES

    o Smooth.ppp
    Crashed when 'kernel' was a function, 'at="points"' and 'scalekernel=FALSE'.
    Fixed.

    o Finhom, Ginhom, Jinhom
    Crashed when ratio=TRUE.
    [Spotted by Art Stock.]
    Fixed.
    
    o envelope
    Crashed for some of the summary functions when ratio=TRUE.
    [Spotted by Art Stock.]
    Fixed.

    o "[.rat"
    Crashed in some cases.
    Fixed.

    o Kcross
    The result of Kcross() was 'invisible',
    i.e. it was not automatically printed.
    Fixed.
    
        CHANGES IN spatstat.core VERSION 2.4-2

OVERVIEW

    o Internal bug fixes.


        CHANGES IN spatstat.core VERSION 2.4-1

OVERVIEW

    o We thank Frederic Lavancier, Sebastian Meyer, Suman Rakshit and Sven Wagner
    for contributions.
    
    o Improved approximation of intensity of Gibbs models.
    
    o Experimental code to represent (theoretical) point process models
    
    o Extract more information about a point process model.

    o Internal improvements and bug fixes.

NEW CLASSES

    o zgibbsmodel
    Experimental. An object of class 'zgibbsmodel' represents a
    Gibbs point process model with specified parameter values
    (whereas 'ppm' represents a model fitted to data).

NEW FUNCTIONS

    o hardcoredist
    Extract the hard core distance of a point process model.

    o interactionorder
    Extract the order of interpoint interaction of a point process model.

    o zgibbsmodel
    Experimental. Create an object of class 'zgibbsmodel'.

    o print.zgibbsmodel
    Experimental. Print an object of class 'zgibbsmodel'.
    
    o is.poisson.zgibbsmodel, is.stationary.zgibbsmodel
    Experimental. Methods for class 'zgibbsmodel'

SIGNIFICANT USER-VISIBLE CHANGES

    o intensity.ppm
    Can now calculate the Coeurjolly-Lavancier DPP approximation of intensity.
    [Code kindly contributed by Frederic Lavancier]
    New argument 'approx' specifies the choice of approximation.

    o relrisk.ppp
    Ratios which are close to 0/0 are handled more effectively,
    reducing the likelihood of strange-looking plots when 'sigma' is very small.

BUG FIXES

    o density.ppp
    Crashed if the observation window had zero area.
    Fixed.

    o dirichletVoronoi.ppp
    Crashed randomly, with obscure error messages from 'im' or 'eval.im',
    when argument 'f' had a small value.
    [Spotted by Suman Rakshit.]
    Fixed.

    o dirichletVoronoi.ppp
    Rarely, produced an image containing NA values.
    [Spotted by Suman Rakshit.]
    Fixed.

    o vcov.ppm
    Crashed in some cases, with message 'object lamdel not found'.
    [Spotted by Sven Wagner.]
    Fixed.
    
        CHANGES IN spatstat.core VERSION 2.4-0

OVERVIEW

    o We thank Sriram Ramamurthy for contributions.

    o spatstat.core now depends on the new package 'spatstat.random'.

    o Functions for generating random patterns have been removed.
    
    o Important bug fixes in anova.mppm and vcov.mppm.

    o Minor improvements and bug fixes

SIGNIFICANT USER-VISIBLE CHANGES

    o package structure
    The code for generating random spatial patterns (including 'rpoispp',
    'rMatClust', 'rThomas', 'rNeymanScott', 'rStrauss', 'rmh')
    has been removed from 'spatstat.core' and placed in a new package
    'spatstat.random'. This new package is required by 'spatstat.core'.

    o anova.mppm
    Improved labelling of models in output.

    o qqplot.ppm, plot.qqppm
    Improved the text label indicating the type of residuals.

    o reload.or.compute
    New argument 'context' 

BUG FIXES

    o vcov.mppm
    For Gibbs (non-Poisson) models, the variance matrix was
    calculated incorrectly in some cases. 
    Fixed.

    o anova.mppm
    Results were sometimes incorrect if the two models
    had different interactions (e.g. Strauss vs Poisson).
    Fixed.

    o anova.mppm
    Crashed for some models with a message about 'coefficient missing
    from new.coef'.
    Fixed.
    
    o anova.mppm
    Gave a warning for some models about "Internal error: 
    unable to map submodels to full model".
    Fixed.

    o addvar
    If the covariate contained NA, NaN or Infinite values,
    the calculations were sometimes incorrect.
    Fixed.

    o pcfmodel.ppm
    Refused to handle an inhomogeneous Poisson process.
    Fixed.

    o fitted.ppm
    Crashed if leaveoneout=TRUE and the model had no fitted coefficients.
    Fixed.

    o reload.or.compute
    Scoping error (crashed sometimes if called from a non-global environment).
    Fixed.
    
        CHANGES IN spatstat.core VERSION 2.3-2

OVERVIEW

    o More diagnostics for spatial logistic regression models.

    o Important bug fix in kppm.

    o Increased numerical stability in kppm.
    
    o Minor improvements and bug fixes.

    o We thank Jonas Brehmer for contributions.

NEW FUNCTIONS

    o lurking.slrm
    Lurking variable plot for spatial logistic regression models.

    o eem.slrm
    Exponential energy marks for spatial logistic regression models.

    o eem.ppm
    Exponential energy marks for Gibbs and Poisson point process models
    (this function was previously called 'eem').

SIGNIFICANT USER-VISIBLE CHANGES

    o eem
    The function 'eem' is now generic, with methods for 'ppm' and 'slrm'.
    The function previously named 'eem' is now called 'eem.ppm'.

    o objsurf.dppm, objsurf.kppm, objsurf.mincontrast
    New arguments 'xlim', 'ylim' determine the range of parameter values
    to be considered.

    o Hybrid
    Printed output from hybrid models has been improved slightly.

    o kppm
    New default settings ensure greater numerical stability of the
    optimization algorithm against the effects of the scale of the
    spatial coordinates.  New argument 'stabilize' specifies whether
    the optimization algorithm should be numerically stabilized.

    o pcf
    Improved error message
    
BUG FIXES

    o kppm
    Results were sometimes incorrect for method='clik2' and method='palm'
    because the log composite likelihood was erroneously truncated to
    positive values. Any fitted model for which logLik(model) = 2.2e-16
    should be suspected of being incorrect.
    Fixed.

    o edge.Ripley
    Results were incorrect for data points lying exactly at the corners
    of a rectangle.
    Fixed.

    o MultiHard
    A hybrid of 'MultiHard' with another multitype interaction caused an error.
    Fixed.

    o simulate.dppm, simulate.detpointprocfamily
    Crashed, rarely, with error message 'The dimension differs from the number of columns in index'.
    Fixed.
    
    o as.fv.kppm, as.fv.dppm
    The default plot labels in 'as.fv(x)' implied that the model 'x' was inhomogeneous.
    Fixed.
    
        CHANGES IN spatstat.core VERSION 2.3-1

OVERVIEW

    o Covariates in ppm and mppm may be functions that depend on the marks
    as well as the spatial coordinates.

    o Automatic selection of threshold for defining a binary predictor.
    
    o Random perturbation of line segments.
    
    o Minor extensions, performance improvements, and bug fixes.

NEW FUNCTIONS

    o thresholdSelect, thresholdCI
    Select the optimal threshold for converting a numerical predictor
    to a binary predictor.
    
    o coef<-.fii
    Changes the coefficients of a fitted interaction object
    (a method for the generic "coef<-")

    o rjitter.psp
    Random perturbation of a spatial pattern of line segments.
    A method for 'rjitter' for class 'psp'.
    
SIGNIFICANT USER-VISIBLE CHANGES

    o ppm
    Covariates which are functions may now depend on the marks
    as well as the spatial coordinates: function(x,y,marks).
    
    o mppm
    Covariates which are functions may now depend on the marks
    as well as the spatial coordinates: function(x,y,marks).
    
    o reach.kppm
    Now returns a result for LGCP models as well as cluster models.

    o distcdf
    Improved regularisation algorithm.
    Argument 'nr=NULL' is now accepted.
    New argument 'delta' allows the result to be interpolated
    onto a finer grid.

    o collapse.fv
    Columns identified by the arguments 'same' and 'different'
    may now be absent from some of the 'fv' objects that will be collapsed.

    o Kest
    When the argument 'domain' is given, the calculation of estimates of K(r)
    has changed slightly, to adhere more closely to the description in
    the help file.
    
    o reload.or.compute
    Now prints a message indicating whether the data were recomputed
    or reloaded from file.
    New argument 'verbose'.

    o update.kppm
    New argument 'envir'.
    
    o clusterfit
    Minor changes to the argument list.

    o mincontrast
    Minor changes to the argument list.

    o mincontrast
    Improved algorithm for handling NA, NaN or infinite values.

    o varcount
    Argument B has a sensible default.

    o pool.envelope
    Now uses the value of 'nrank' which was used in the original envelopes.
    
    o Kmulti
    New argument 'rmax'.

    o Kinhom
    No longer issues a warning about changed behaviour in the case where
    'lambda' is a fitted model.
    
    o pcfinhom
    No longer issues a warning about changed behaviour in the case where
    'lambda' is a fitted model.
    
BUG FIXES

    o subfits
    The fitted interaction coefficients were garbled.
    If 'mfit' is the mppm object and 'a <- subfits(mfit)[[i]]' is one of the
    sub-models, then coef(a) was correct, but coef(fitin(a)) was incorrect.
    The fitted interaction was shown correctly by printing 'mfit'
    but incorrectly by printing 'a'.
    Fixed.
    
    o varcount
    The result was very inaccurate if the cluster radius was small 
    compared to the size of the window 'B', due to discretisation error.
    Fixed.

    o segregation.test
    The test statistic was calculated as the mean, rather than the sum,
    of discrepancies between probabilities. (The p-value was not affected.)
    Fixed.

    o Kest
    If 'domain' was specified, 'rmax' was ignored.
    Fixed.

    o edge.Ripley
    Value was incorrect for a point lying exactly on a corner.
    Fixed.

    o edge.Ripley
    Crashed when method="interpreted", if a point lay exactly on a corner.
    Fixed.

    o plot.fv, plot.envelope
    Crashed when trying to display a significance band of width zero
    around a constant function.
    Fixed.
    
    o collapse.fv
    Crashed if 'length(same) > 1'.
    Fixed.

        CHANGES IN spatstat.core VERSION 2.3-0

OVERVIEW

    o We thank Chiara Fend for contributions.
    
    o Extensive support for spatial logistic regression models.

    o New fitting method in kppm and dppm.

    o Transect of an image along a curve.

    o Image cross-correlation and cross-covariance.
    
    o Summary method for 'objsurf'.

    o Minor bug fixes.

NEW FUNCTIONS

    o response
    Generic function which extracts the values of the response
    in a fitted model. There are methods for lm, glm (which extract the
    numeric vector of responses), and ppm, kppm, slrm, lppm and mppm
    (which extract the original data point pattern).

    o cov.im, cor.im
    Correlation or covariance between several pixel images.
    
    o summary.objsurf, print.summary.objsurf
    Summary method for 'objsurf'

    o residuals.slrm
    Residuals for spatial logistic regression models.
    
    o leverage.slrm, influence.slrm, dfbetas.slrm, dffit.slrm
    Leverage and influence diagnostics for spatial logistic regression models.
    
    o rhohat.slrm
    Method for 'rhohat' for spatial logistic regression models.

    o envelope.slrm
    Method for 'envelope' for spatial logistic regression models.

    o intensity.slrm
    Method for 'intensity' for spatial logistic regression models.

    o deviance.slrm
    Method for 'deviance' for spatial logistic regression models.

    o pseudoR2.slrm
    Method for 'pseudoR2' for spatial logistic regression models.

    o quadrat.test.slrm
    Method for 'quadrat.test' for spatial logistic regression models.

    o parameters.slrm
    Method for 'parameters' for spatial logistic regression models.

    o valid.slrm
    Method for 'valid' for spatial logistic regression models.

    o emend.slrm
    Method for 'emend' for spatial logistic regression models.

    o roc.slrm
    Method for 'roc' for spatial logistic regression models.

    o auc.slrm
    Method for 'auc' for spatial logistic regression models.

    o Window.slrm, as.owin.slrm
    Methods for 'Window' and 'as.owin' for spatial logistic regression models.

SIGNIFICANT USER-VISIBLE CHANGES

    o kppm, dppm
    New option 'method="adapcl"' performs adaptive composite likelihood fitting.
    [Contributed by Chiara Fend.]
    
    o transect.im
    New argument 'curve' allows the user to specify a curved transect.

BUG FIXES

    o clusterfield
    Values of the cluster field were slightly incorrect (slightly higher
    than the correct values) near the edge of the window,
    because an 'edge correction' was mistakenly applied.
    Fixed.

    o rhohat
    The rug plot (produced by plot.rhohat) was incorrect
    when rhohat was called with method="piecewise".
    Fixed.

    o markcrosscorr
    Did not recognise the option 'correction="none"'.
    Fixed.

   o roc.ppp
   The default plot of the result of roc.ppp
   did not include the diagonal line 'y=x'.
   Fixed.
   
        CHANGES IN spatstat.core VERSION 2.2-0

OVERVIEW

    o We thank Abdollah Jalilian, Yongtao Guan and Rasmus Waagepetersen
    for contributions.
    
    o summary method for spatial logistic regression models

    o estimation of the spatial covariance function of a pixel image

    o simulation of the product shot noise Cox process.

    o extensions to rhohat
    
NEW FUNCTIONS

    o rPSNCP
    Generate simulated realisations of the product shot noise Cox process.
    Contributed by Abdollah Jalilian, Yongtao Guan and Rasmus Waagepetersen.
    
    o spatcov
    Estimate the spatial covariance function of a pixel image.

    o summary.slrm, print.summary.slrm
    Summary method for spatial logistic regression models

    o coef.summary.slrm
    Print the fitted coefficients, confidence interval and p-values
    for a spatial logistic regression model.

    o pairMean
    Compute the mean of a specified function of interpoint distance
    between random points in a window.

SIGNIFICANT USER-VISIBLE CHANGES

    o rhohat
    New option (smoother='piecewise') computes a piecewise-constant
    estimate of rho(z).
    
    o rhohat
    The result now includes the 'average' intensity rho.

    o distcdf
    Arguments which are NULL will be treated as missing.

    o distcdf
    New argument 'savedenom'.
    

        CHANGES IN spatstat.core VERSION 2.1-2

OVERVIEW

    o Reduced CRAN check time.

        CHANGES IN spatstat.core VERSION 2.1-1

OVERVIEW

    o Minor bug fix

BUG FIXES

    o simulate.kppm
    Conditional simulation crashed on rare occasions,
    with an error about negative probabilities.
    Fixed.

        CHANGES IN spatstat.core VERSION 2.1-0

OVERVIEW

    o We thank Achmad Choiruddin, Jean-Francois Coeurjolly
    and Rasmus Waagepetersen for contributions.

    o Conditional simulation in kppm
    
    o Information criteria for model selection in kppm

    o Modified handling of covariates in slrm

    o densityfun.ppp handles query points outside original window

    o Improved output in summary.mppm

    o Minor improvements and bug fixes.

NEW FUNCTIONS

    o ic
    Information criteria for model selection in ppm and kppm.
    Kindly contributed by Achmad Choiruddin, Jean-Francois Coeurjolly
    and Rasmus Waagepetersen.
    
SIGNIFICANT USER-VISIBLE CHANGES

    o simulate.kppm
    Conditional simulation of the model, given a fixed number of points,
    is now supported using the new arguments 'n.cond' and 'w.cond'.
    
    o densityfun.ppp
    The resulting function can now handle query points which lie
    outside the window of the original data,
    and has argument 'drop=TRUE' which specifies how to handle them.

    o rpoint
    New argument 'forcewin' forces the code to use the window 'win'
    when 'f' is a pixel image.

    o slrm
    In the default case (where dataAtPoints is not given)
    all spatial covariates, including the spatial coordinates x and y,
    are now evaluated at the centre of each pixel. This improves
    consistency with other implementations of spatial logistic regression.
    
    o slrm
    Silently ignores any arguments '...' that are not recognised by 'as.mask'

    o summary.mppm
    Improved summary of the dependence of the
    interpoint interaction on the covariates.

    o pairs.im
    New argument 'drop'.
    
BUG FIXES

    o model.matrix.mppm
    If the model was fitted using 'gam', the resulting matrix
    did not have an 'assign' attribute.
    Fixed.
    
    o model.depends
    Crashed for models fitted using 'gam'.
    Fixed.

    o predict.slrm, fitted.slrm
    Crashed if the model was fitted using split pixels (argument 'splitby').
    Fixed.

    o predict.slrm, fitted.slrm
    Crashed in some cases when 'window' was given.
    Fixed.

    o update.slrm
    Failed to find covariates that were provided in 'env'.
    Fixed.

    o cdf.test
    Crashed if the covariate was constant.
    Fixed.

        CHANGES IN spatstat.core VERSION 2.0-0

OVERVIEW

    o We thank Tilman Davies,  Greg McSwiggan and Suman Rakshit
    for contributions.

    o We thank Corey Anderson, Michael Chirico, Andy Craig,
    Marcelino de la Cruz, Tilman Davies, Pavel Fibich,
    Kurt Hornik, Gopalan Nair, Yonatan Rosen and Rasmus Waagepetersen
    for contributions.
    
    o Diffusion kernel smoothing.
    
    o More support for spatial logistic regression models.

    o predict.mppm now works for multitype point process models.

    o Improved handling of 'newdata' in predict.mppm.

    o More support for multi-dimensional patterns.
    
NEW FUNCTIONS

    o densityHeat
    New generic function for diffusion kernel estimation of intensity

    o densityHeat.ppp
    Diffusion kernel estimation of intensity for point pattern in 2 dimensions.
    This is an alternative to density.ppp.

    o intersect.boxx
    Compute intersection of boxes in multi-dimensional space

    o scale.boxx, scale.ppx
    Methods for 'scale' for boxes and patterns in multi-dimensional space

    o shift.boxx, shift.ppx
    Methods for 'shift' for boxes and patterns in multi-dimensional space

    o is.boxx
    Determine whether an object is a multidimensional box

    o mincontrast
    New argument 'action.bad.values' specifies what action is taken
    when the summary function produces NA or NaN or infinite values.
    

    
SIGNIFICANT USER-VISIBLE CHANGES

    o slrm
    'step' can now be applied to models fitted using 'slrm'.

    o predict.mppm
    Now supports multitype point process models.

    o predict.mppm
    Improved handling of argument 'newdata'
    
    o rotmean
    The result now has the same 'unitname' as the input object X.
    New argument 'adjust' controls the smoothing bandwidth.
    
    o rlabel
    New argument 'group' specifies that the points are divided into
    several groups, and that relabelling is applied within each group.
    
    o simulate.ppm
    Now recognises the argument 'window' as an alternative to 'w'.

    o kppm
    Improved numerical robustness.

    o Kcross, Gcross, Jcross
    Function labels (shown on the plot legend) have been
    improved when i = j.

    o anova.mppm
    Issues a warning when applied to random-effects models
    (models fitted using the argument 'random').

BUG FIXES

    o Gest
    If correction="rs" or correction="km", then both the reduced-sample
    (border correction) and Kaplan-Meier corrected estimates were calculated.
    [Spotted by Gopalan Nair.]
    Fixed.

    o rMatClust, rThomas, rCauchy, rVarGamma
    If the fitted model was effectively a Poisson process,
    the result did not have attributes 'Lambda' and 'parents'
    even when the user requested them.
    Fixed.
    
    o model.matrix.mppm
    Crashed with random-effects models.
    Fixed.

    o anova.mppm
    Crashed with random-effects models.
    Fixed.

    o simulate.rhohat
    Crashed when applied to rhohat objects computed from data
    on a linear network.
    Fixed.

    o objsurf.kppm
    Crashed if the model was fitted by Palm likelihood (method="palm")
    or second order composite likelihood (method="clik2").
    Fixed.
    
    
        CHANGES IN spatstat.core VERSION 1.65-11

OVERVIEW

    o Internal tweaks.
    
        CHANGES IN spatstat.core VERSION 1.65-10

OVERVIEW

    o Minor corrections to documentation.

        CHANGES IN spatstat.core VERSION 1.65-9

OVERVIEW

    o We thank Ian Buller for a suggestion.

    o weights permitted in density calculation for line segments.

SIGNIFICANT USER-VISIBLE CHANGES

    o density.psp
    New argument 'weights'.

        CHANGES IN spatstat.core VERSION 1.65-8

OVERVIEW

    o Minor changes to appease the compiler.

        CHANGES IN spatstat.core VERSION 1.65-7

OVERVIEW

    o We thank Michael Chirico for a contribution.

    o Minor changes to appease the compiler.
    
        CHANGES IN spatstat.core VERSION 1.65-6

OVERVIEW

    o We thank Tilman Davies and Pavel Fibich for contributions.

    o Important bug fix in simulation of log-Gaussian Cox processes.

    o Increased speed for large datasets.

    o variance calculations handle larger datasets.
    
SIGNIFICANT USER-VISIBLE CHANGES

    o vcov.ppm, summary.ppm
    Variance calculations now handle larger datasets
    (using sparse arrays).
    
    o rSSI
    Accelerated.
    
    o overall speed
    Changes have been made to the internal code of spatstat 
    which should accelerate computations involving large datasets.

    o localpcf, localpcfinhom
    New argument 'rvalue'.

BUG FIXES

    o rLGCP, simulate.kppm
    Simulation results for log-Gaussian Cox processes were incorrect
    unless the pixel dimensions and pixel spacings were identical
    on the horizontal and vertical axes. (If pixel dimensions were not
    specified, then the results were incorrect whenever the Frame of the
    simulation window was not a square.) 
    [Spotted by Tilman Davies.]
    Fixed.
    
    o Vmark
    Crashed if normalise=TRUE when there was only one column of marks.
    (Spotted by Pavel Fibich.)
    Fixed.
    
    o nnclean
    Crashed if k >= npoints(X).
    Fixed.

    o print.ppm
    Crashed sometimes when applied to the result of subfits().
    Fixed.
    
    CHANGES IN spatstat.core VERSION 1.65-5

OVERVIEW

    o Minor changes required by CRAN.
    

        CHANGES IN spatstat.core VERSION 1.65-1

OVERVIEW

    o Added NEWS file.


        CHANGES IN spatstat.core VERSION 1.65-0

OVERVIEW

    o Package initialised at version 1.65-0

SIGNIFICANT USER-VISIBLE CHANGES

    o spatstat.core
    The package 'spatstat.core' has been created from a subset
    of the code in the original 'spatstat' package version 1.65-0.
    It contains the core functionality for statistical analysis of
    spatial data. For an overview, see help("spatstat.core-package")

    o Execution
    The 'spatstat.core' package is slightly faster than the corresponding
    code in the 'spatstat' package, because the procedure for calling
    internal C functions has been streamlined.