2.11.1 (September 29, 2023)
    * weibreg: Arguments xlim, ylim, xlab, ylab, main are now "working".
    * Submitted to CRAN.
    
2.11.0 (September 6, 2023)
    * Missing '#include <R_ext/BLAS>' added where necessary (internal).
    * Submitted to CRAN.

2.10.3.9310 (June 24, 2024)
    * ltxCoef3: No printing of number of cases (internal).

2.10.3.9300 (June 22, 2023)
	* HiscoHisclass: Bug in 'out[23 & (hisclass == 8)] <- 10':
	Should be 'out[s23 & (hisclass == 8)] <- 10'. Fixed. (All
	farmers were thrown out!!)

2.10.3.9200 (May 21, 2023)
    * HiscoHisclass: New function, still under  development.

2.10.3.9100 (April 24, 2023)
    * summary and print functions: begin update. Nothing visible yet.

2.10.3.9000 (February 17, 2023)
    * aftgomp.R added but not exported (still under development).

2.10.3 (February 10, 2023)
    * Change old-style format to new-style in CITATION file.

2.10.2 (February 9, 2023)
    * Submitted to CRAN (and rejected, see 2.10.3).

2.10.1.9020 (February 9, 2023)
    * More plot fixes (mostly cosmetic).

2.10.1.9010 (February 9, 2023)
    * Bugs in plot.phreg, plot.aftreg fixed (wrong No. of covariates).

2.10.1.9000 (January 4, 2023)
    * The drop1 method drop1.coxreg removed due to bug; 
    stats::drop1.default should work for coxreg.
    
2.10.1 (November 16, 2022)
    * Removed GB_zeroin.c (not needed) and resaved files in utf-8.
    * Submitted to CRAN.
    
2.10.0.9300 (October 5, 2022)
    * ltx2: Add number of observations to output.

2.10.0.9250 (September 26, 2022)
    * More bugs in ltx2 fixed.

2.10.0.9200 (September 24, 2022)
    * Fix bug in output in ltx2.
    
2.10.0.9150 (September 21, 2022)
    * Remove class "phreg" from aftreg objects.

2.10.0.9100 (September 21, 2022)
    * Bug in ltx2.phreg eliminated.
    * Comparing class(x) to string "y" changed to inherits(x, "y") in two places.
    
2.10.0.9050 (April 20, 2022)
    * Fix new bug (when interactions present) in ltx2.

2.10.0.9000 (April 20, 2022)
    * Fix bug (when interactions present) in ltx2.
    
2.10.0 (April 17, 2022)
    * Submission to CRAN.
    
2.9.0.9500 (April 17, 2022)
    * fix error in drop1.coxreg

2.9.0.9450 (April 16, 2022)
    * fix examples taking too long.
    * add ltx2 with methods as alternatives to ltx.
    
2.9.0.9400 (April 15, 2022)
    * drop1 method for coxreg added.
    
2.9.0.9300 (April 3, 2022)
    * Add logLik methods to coxreg and phreg.

2.9.0.9210 (Dec 5, 2021)
    * plot.coxreg: Add (0, 0) to cumulative hazards plot.

2.9.0.9200 (Dec 4, 2021)
    * hazards.coxreg: Add (0, 0) to cumhaz.
    * make.communal: Deal with zero length interval output and allow 
    input for 'com.dat' with more than one variable.

2.9.0.9100 (July 8, 2021)
    * coxreg2: New function, copied from coxreg. Not modified yet.

2.9.0.9050 (June 30, 2021)
    * coxreg: Checking calls to survival package.

2.9.0.9000 (June 21, 2021)
    * Adding math mode to the methods of ltx.

2.9.0 (Apr 30, 2021)
    * Submission to CRAN.
    
2.8.5.9100 (Apr 28, 2021)
    * coxreg: Remove calculation of boot.sd (too freqent failures).

2.8.5.9070 (Apr 18, 2021)
    * plot.tpchreg: Error with arguments 'xlab' and 'ylab' fixed.

2.8.5.9060 (Feb 24, 2021)
    * Data for the year 2020 included in swepop and swdeaths.

2.8.5.9050 (Jan 17, 2021)
    * Vignettes updated and completed.

2.8.5. 9010 (Jan 14, 2021)
    * plotHaz: New arguments 'leglab' and 'log = ""'.

2.8.5.9000 (Jan 13, 2021)
    * hazards: Fix bug appearing when no covariates are present.
    
2.8.5 (Jan 11, 2021)
    * Submission to CRAN.

2.8.4.9450 (Jan 11, 2021)
    * hazards: Fix bug with parametric model and fixed shape.

2.8.4.9400 (Jan 9, 2021)
    * print.summary.aftreg: Fix error in heading.

2.8.4.9300 (Jan 4, 2021)
    * hazards: Added method for aftreg objects.

2.8.4.9250 (Dec 30, 2020)
    * tpchreg: try for singular hessian, continue without error.

2.8.4.9240 (Dec 29, 2020)
    * pchreg.fit: Put names on coefficients.

2.8.4.9230 (Dec 28, 2020)
    * ltx: Move caption to top of table.

2.8.4.9220 (Dec 27, 2020)
    * ltx.tpchreg: Bug fix.

2.8.4.9210 (Dec 25, 2020)
    * plotHaz: cut plot at 'interval', if given.

2.8.4.9200 (Dec 23, 2020)
    * compHaz: New function, comparing cumulative hazards functions for 
    fitted models. See also 'check.dist'.

2.8.4.9120 (Dec 22, 2020)
    * age.window: No warning is given for an empty result (check for NULL).

2.8.4.9100 (Dec 21, 2020)
   * pchreg: New function for piecewise constant PH regression.
   * phreg: 'dist = "pch"' deprecated. Use 'pchreg' instead.

2.8.4.9020 (Dec 20, 2020)
   * Fix wrong description of C function 'breslow_rs2' in 'init.c'. Triggered by LTO.
   * CRAN release.

2.8.4.9010 (Dec 19, 2020)
   * ltx.phreg: Remove double printout in the end.

2.8.4.9000 (Dec 16, 2020)
   * plotHaz: New function for plotting two baseline cumulative 
   hazards functions.
   * attrassign and untangle.specials (survival package) removed from 
   importFrom.

2.8.4 (Dec 15, 2020)
    * Fixed error in doi field in inst/CITATION, resubmitted.

2.8.3 (Dec 14, 2020)
    * Vignette problems fixed.
    * check.dist: Bug fixed.

2.8.2 (Dec 13, 2020)
    * CRAN Release.

2.8.1.99993 (Dec 8, 2020)
    * hazards: made generic with methods for regression outputs. 
    
2.8.1.99991 (Dec 2, 2020)
    * ltx: Introduce "\footnotesize" in latex table printing.
    
2.8.1.9999 (Nov 29, 2020)
    * tpchreg: fix error in example.
    * hazards: fix bug.

2.8.1.9998 (Nov 28, 2020)
    * Fix bug in ltx.coxreg.
    * Add example to tpchreg help page.

2.8.1.9997 (Nov 27, 2020)
    * coxfun0: New function, loglihood for null coxreg model.
    * ltx, print: Add and modify methods for printing regression fits.

2.8.1.9996 (Nov 22, 2020)
    * coxreg: Removed warning concerning not using option 'coxph'.
    
2.8.1.9995 (Nov 20, 2020)
    * hazards: New function, that calculates the baseline hazards from a 
    coxreg fit.

2.8.1.9994 (Nov 19, 2020)
    * coxreg: Add log likelihood and df to null models.
    
2.8.1.9993 (Nov 18, 2020)
    * child: The data set is expanded with info about mother.

2.8.1.9992 (Nov 16, 2020)
    * coxreg, phreg, aftreg: Fixed bug in calculating weighted means, which
    affects dates as covariates.
    
2.8.1.9991 (Nov 15, 2020)
    * print.summary.coxreg: New argument 'short' added.
    * plot.hazdata: Fix y limits.
    
2.8.1.9990 (Nov 12, 2020)
    * summary.[aft,cox,ph,tpch]reg: Change p-value label to 'Wald p'.

2.8.1.9989 (Nov 10, 2020)
    * logrank: Fix error in evaluation of arguments.

2.18.1.9988 (Nov 09, 2020)
    * coxreg.fit: Bug fix (when one covariate).
    * plot.logrank: New plot method for logrank objects.

2.18.1.9987 (Nov 09, 2020)
    * coxreg: fix stratum names when no covariates.

2.18.1.9986 (Nov 08, 2020)
    * logrank: 'hazards' is a new output component.
    * plot.logrank: New function (plot method).

2.8.1.9985 (Nov 07, 2020)
    * logrank: New function with a print method (the log-rank test).
    
2.8.1.9975 (Oct 30, 2020)
    * 'survival' moved from Depends to Imports.
    * No import of functions 'frailty' and 'cluster' from survival. 
    (Not used by eha, but by some innocent users of eha. Want to stop that.) 

2.8.1.9971 (Oct 19, 2020)
    * toTime: Change result from 'timediff' to 'numeric'.

2.8.1.9970 (0ct 11, 2020)
	* tpchreg.fit: Bug fixes.

2.8.1.9965 (Oct 07, 2020)
    * tpchreg: Estimates of sd of hazards introduced. Not completed.

2.8.1.9961 (Sep 27, 2020)
    * tpchreg: Failed for cases with zero exposure, fixed.
    
2.8.1.9960 (Sep 21, 2020)
    * summary: Failed for null models with tpchreg, pchreg, now fixed.
    
2.8.1.9950 (Sep 08, 2020)
    * coxreg: Default value for argument 'coxph' changed to TRUE.
    * plot.phreg: Argument 'new.data' removed, 'score' added.

2.8.1.9945 (Sep 07, 2020)
    * coxreg: Fixed (bad) shift in 'linear.predictors'.
    
2.8.1.9942 (Aug 12, 2020)
    * phreg: Added 'pfixed = FALSE' to return values of distributions
    where pfixed is irrelevant. (Yet a reason to split 'phreg' into functions by
    narrow groups of distributions.)

2.8.1.9941 (Aug 11, 2020)
    * phreg.summary: Bug fixed. It was introduced by differences between the
    distribution 'pch' and the other distributions. The piecewise constant 
    hazards regression will eventually break out of phreg and form a separate
    function, named 'pchreg'. Have the same thoughts about the Gompertz 
    distribution, to be put in 'gompreg'. We'll see... Main reason is that 
    these guys need special treatment in the programming.

2.8.1.9935 (Aug 8, 2020)
    * phreg: Number of events in output was denoted 'events', now changed
    to 'n.events'.
    * aftreg: Number of events in output was denoted 'events', now changed
    to 'n.events'.
    * weibreg: Number of events in output was denoted 'events', now changed
    to 'n.events'.
    * tpchreg: Number of events in output was denoted 'events', now changed
    to 'n.events'.
    * coxreg: Number of events in output was denoted 'events', now changed
    to 'n.events' (to match output from coxreg).

2.8.1.9930 (Jul 26, 2020)
    * phreg: Fix issue with 'shape = ' something positive.
    * summary.phreg: Rewritten to conform with other summaries.

2.8.1.9925 (Jul 24, 2020)
    * plot.phreg: Only one function per plot is the rule now. Except of course
    if the analysis was stratified.
    * Gompertz: Fix errors in documentation.
    * plot.coxreg: Some minor anomalies fixed.
    * coxreg: Default for coxph changed to FALSE.
    * print.summary.coxreg: Bug wrt LR p fixed.
    * child: New data set added.

2.8.1.9920 (Jul 23, 2020)
    * phreg: fixed bug in Gompertz regresion with 'rate' parametrization. 
    BTW, the 'rate' paramerization will be the default soon!

2.8.1.9910 (Jul 22, 2020)
    * Gompertz: added argument 'rate' to all functions.
    
2.8.1.9900 (Jul 17, 2020)
    plot.coxreg: Test for presence of 'hazards' object, avoid bug when 
    'coxph = FALSE'.

2.8.1.9850 (Jul 13, 2020)
    * risksets: New argument 'collate_sets', a logical. Defaults to FALSE,
    which conserves old behaviour. If TRUE, output is a list of lists, where 
    each level 2 list describes one risk set.
    
2.8.1.9800 (Jul 13, 2020)
	* coxfunk: New function calculating a partial likelihood.

2.8.1.9750 (Jul 10, 2020)
    * Vignette 'Proportional hazards regression with tabular data' develops.

2.8.1.9700 (Jul 8, 2020)
    * coxreg: New argument 'coxph' (logical) is introduced. If TRUE, 
    survival::agreg.fit (or survival::coxph.fit) is called for in the 
    optimization. If FALSE, 'own' optimization is called for, can be slow 
    for (even moderately) large data sets (many distinct event time points).
    Default value is TRUE.

2.8.1.9650 (Jul 7, 2020)
    * plot.coxreg: Fixed newly introduced bug in the cases 'method = m(pp)l'.
    * eha vignette: First serious attempt.
    * summary.tpchreg: New function, with a print method.

2.8.1.9600 (June 30, 2020)
    * tpchreg: add argument 'last', and allow 'time' to be numeric. 
    Documentation changed accordingly.

2.8.1.9550 (June 28, 2020)
    * swepop: New data set, Swedish population by age, sex, and 
    year 1968-2019.
    * swedeaths: New data set, Swedish deaths by age, sex, and 
    year 1968-2019.
    * swe07: Old data set is now obsolete and deprecated. Will be removed.

2.8.1.9500 (June 17, 2020)
	* tcphreg: Added debug info.

2.8.1.9450 (June 15, 2020)
    * tpchreg: Add rownames and colnames to hazards (if matrix).
    * summary.tpchreg: New function.
    
2.8.1.9400 (June 14, 2020)
    * tpchreg: Finish first step of development, now useful.

2.8.1.9300 (June 11, 2020)
	* plot.coxreg: Bug fix.

2.8.1.9200 (June 9, 2020)
    * coxreg revised for "coxhp case".
    * plot.coxreg rewritten, using survival::(survfit, plot.survfit)
    * toTpch: Bug fixed.
    * nobs.coxreg: Changed to count events instead of rows (bug fix).
    * plot.hazards: Removed.
    * plot.hazdata: Removed.
    
2.8.1.9150 (june 7, 2020)
    * coxreg: Arguments 'hazards' and 'center' deprecated.
    * phreg: Argument 'center' deprecated.
    * Bugs (some ...) fixed.
    
2.8.1.9100 (June 6, 2020)
    * coxreg: Default values of arguments 'hazards' and 'center' changed to FALSE.
              A null model output has a 'terms' added.
              
2.8.1.9050 (Jun 4, 2020)
    * tpchreg: Argument 'pieces' removed (is now calculated from 'time').

2.8.1.9000 (Jun 3, 2020)
	* tpchreg: and friends. Work continues.

2.8.1 (Mar 31, 2020)
    * Published on CRAN.

2.8.0.9050 (Mar 3, 2020)
	* gompstartRate: Simpler starting values introduced (internal).
	* risksets: Bug fix: Now works when 'members = FALSE'.

2.8.0.9000 (Feb 21, 2020)
	* getsums.f: Replace 'dcopy' in three instances to please gfortran 10. 

2.8.0 (Dec 19, 2019)

	* Rewriting tpchreg.
	* Add tpchreg methods: print, plot, extractAIC.
	* Add new function: oe (occurrence/exposure data).

2.7.6.9200 (Dec 09, 2019)

	* C header files conformed to PKG_CFLAGS=-fno-common.

2.7.6.9100 (Oct 22, 2019)

    * toTpch: Take care of possibly empty interval(s).

2.7.6.9000: (Sep 26, 2019)
    * tcphreg: Expanding and writing summary, print, and plot methods. Not finished yet.

2.7.6: (Sep 15, 2019)
    * cdgemv.c: Removed. Finally correct handling of C and Fortran calls to BLAS and LAPACK routines (thanks to Tomas Kalibera).

2.7.5: (Sep 11, 2019)
	* cdgemv.c: C wrapper for dgemv added as replacement for gdgemv.f (removed).

2.7.1: (Sep 4, 2019)
	* gdgemv.f: New function, essentially a copy of the LAPACK subroutine dgemv. Purpose: Avoid bug in gfortran 7+.
	* risksets: Warning about sample_fraction fixed.
	* Errors in plot.aftreg fixed (thanks to Mark Clements).
	* In eha_fun.c, changed 'finite' to 'isfinite' (C99).
        * New functions: tpchreg, toTpch.

2.6.0.9200: (Jul 22, 2018)
        * Bug in phreg, Gompertz distribution, 'param ="rate"', fixed. Again!

2.6.0.9100: (Jul 19, 2018)
        * Bug in phreg, Gompertz distribution, 'param ="rate"', with strata, fixed.

2.6.0.9000: (Jul 15, 2018)
        * plot.phreg, error in Gompertz with 'param = "rate"' fixed.

2.6.0: (Jul 13, 2018)
        * Depends on survival (>= 2.42-5)
        * Released to CRAN.

2.5.2.9100: (Jun 5, 2018)
        * plot.Surv removed: is now a method from the survival package,
        version >= 2.42-5

2.5.2.9000: (Jun 5, 2018)
        * plot.Surv: bug fix, illegal value to 'fn' gave nonsense-plot.

2.5.2:  (Feb 26, 2018)
        * mpch is a new function: Gives the mean of a piecewise constant 
        hazards distribution.

2.5.1:  (Dec 04, 2017)
        * ppch, dpch, hpch, Hpch, qpch were not exported. Fixed.

2.5.0:  (Oct 11, 2017)
        * New versioning system (*Hadley Wickham*'s suggestion).
        * Documentation 'roxygenized'.
        * Development moved to [gitHub](https://github.com/goranbrostrom/eha).
        * The package glmmML was merged into eha in version 2.0. This has
	turned out to be a very bad idea ("double entry"), and all
	relevant
	functions are now Defunct and removed. Use glmmML instead.

        * glmmML, glmmboot, ghq, glmmbootFit, glmmML.fit are all Defunct
	and removed. No reverse dependency used any of them. Those
	functions are all found in the package glmmML, on CRAN.

2.4-6:  (Sep 22, 2017)
	* phreg: Start values by 'coxreg' (bad idea) removed.
	* aftreg: in 'addMeans' and 'aftreg1', an error
	(spotted by Jingchunzi Shi) with fixed scale corrected.

2.4-5:	(May 2, 2017)
	* init.c: Added after R-3.4.0.
	* coxreg: "Geometric bootstrap" removed (it never worked and was
	never used).
	* cal.window, age.window: Works now with 'tibbles'.

2.4-4:	(Sep 21, 2016)
	* rpch: Bug fixed (spotted by Brigitte Mueller).
	* plot.phreg: parameters 'changeLog' and '...'
	enhanced. Documentation updated.
	* plot.coxreg: Bug concerning 'printLegend' fixed.
	* weibreg: Note added in documentation: "Use phreg
	instead". weibreg will soon be deprecated.

2.4-3: (May 5, 2015)
	* coxreg: Changed the default for 'hazards' to TRUE.
	* getHaz: Modified to handle missing strats and score.
	* print.risksets: Added '...' to argument list.
	* coxreg: Fitting a NULL model didn't produce enough information
	like log likelihood ('loglik'), for comparison of models. Fixed.
	* glmmboot: 'frail' set to Inf or -Inf for appropriate clusters.
	* plot.hazdata: Updated to acknowledge the argument 'hazards' that was
	included in 'coxreg' in version 2.4-2.
	* plot.coxreg: Updated to acknowledge the argument 'hazards' that was
	included in 'coxreg' in version 2.4-2. Errors in the manual page
	corrected.
	* risksets: Risk set sampling did not always work (memory leak in
	C code); fixed.
	* risksets: New value in the Value list: sample_fraction.
	* plot.aftreg: Now registered as an S3 method.
	* print.aftreg: Scale and shape parameters are only printed on the
	log scale.

2.4-2: (Dec 15, 2014)
	* src: 'abs' corrected to 'fabs' in two places in C code.
	* coxreg: Argument 'hazards' added, with default = FALSE.
	* plot.Surv: Argument 'printLegend' added.
	* coxreg: Changed labels in case of ordered factors.
	* Foreign calls: 'DUP = FALSE' removed everywhere.
	* print.pch: Bug preventing printing(!) fixed.
	* plot.pch: Bug preventing plotting(!) fixed.

2.4-1: (Feb 14, 2014)
	* phreg: Fails often for the lognormal and loglogistic models with
	badly fitting data. A warning is included in the documentation,
	and better error handling in case of a failure. Example under
	'check.dist' changed accordingly.

2.4-0:	(Feb 10, 2014)
	* coxreg: Call to survival::[agreg,coxph].fit in "standard" cases
	reintroduced; Thanks to Terry Therneau, who exports these
	functions from survival_2.37-6. So this version of 'eha' requires
	survival (>= 2.37-6). Argument 'center' reintroduced, with slightly
	different meaning.
	* plot.coxreg: argument 'newdata' not used.
	* phreg: Stratified analysis in the piecewise constant hazards
	model introduced. Argument 'center' reintroduced, with slightly
	different meaning.
	* plot.phreg: Argument 'newdata' not used.

2.3-5:  (Jan 17, 2014)
	* plot.coxreg: Added the argument lty, for line type, and the
	logical parameter 'printLegend' (default: TRUE). Setting it to FALSE
	skips the legend in the figure.

2.3-4:	(Dec 5, 2013)
	* plot.coxreg: added the argument col, for color line(s).
	* plot.hazdata: Ditto.


2.3-3: (Oct 10, 2013)
	* coxreg: Restoring equity.
	* glmmML: Removed spurious printing in 'glmmml.c'.

2.3-2: (Oct 10, 2013)
	* aftreg: Updated start values for the Gompertz distribution,
	matching the (now mandatory) 'canonical' parametrisation.

2.3-1: (Oct 03, 2013)
	* coxreg: Downgraded due to limitations in using
	non-exported function (agreg) from 'survival'. This will be taken
	care of ASAP. Meanwhile, live with slightly slower 'coxreg' and
	choking on huge data sets.

	* Coxreg: Removed for the same reason as the downgrade of coxreg.

2.3-0: (Oct 03, 2013)
	* phreg: The lognormal and loglogistic distributions have
	an intercept added to the linear predictor; in effect extending
	these distributions to three-parameter families.
	A bug (found by Claire Williams) affecting the covariance matrix
	of the parameter estimates has been eliminated.
	* aftreg: The same bug as for phreg eliminated. The available
	parametrizations are changed to 'lifeAcc' and 'lifeExp'. For the
	Gompertz distribution, the 'canonical' representation is used
	regardless.
	* hlnorm, Hlnorm, hllogis, Hllogis: New parameter, 'prop',
	added. It simply multiplies the previous definitions by its value.
	* age.window, cal.window: speed up and handling of empty intervals
	* perstat: speed up.

2.2-6:	(June 23, 2013)
	* plot.phreg, plot.coxreg: legend added for stratified fits.
	* Coxreg: New function designed to handle huge data sets. coxreg
	chokes on 'integer overflow', and memory overflow,  when the
	collection of risk sets is created and saved. Coxreg is
	essentially a wrapper for 'coxph' in the 'survival' package, but
	with methods for nice(r) printing.
	* coxreg: Error in the calculation of 'total time at risk' fixed
	(gave negative values when no left truncation!).

2.2-5:	(May 13, 2013)
	* Gompertz, ltx: Too long lines in the documentation fixed.

2.2-4:	(May 13, 2013)
	* phreg: The gompertz 'scale' parameter is now a real scale
	parameter through a reparametrization of the baseline gomperz
	distribution. It follows now the 'canonical' parametrization.
	There is also a new argument, 'param', which can take the values
	"canonical" (default) and "rate". Better start values
	for the Gompertz regressions.
	* check.dist : Has failed due to different scaling in coxreg and
	phreg since 2.2-2. Now fixed.

2.2-3:	(April 17, 2013)
	* risksets: Added an argument, 'members', a logical with default
	value 'TRUE'. If TRUE, the members of all risk sets are returned.
	With huge data sets, this may be too much, and not wanted. In such
	cases, set 'members = FALSE'.
	* coxreg: Memory leaks fixed in getHaz.f, sizes.c and risk.c.

2.2-2:	(March 23, 2013)
	* phreg: argument 'center' deprecated; centering of covariates is
	done internally, but uncentered results reported. If you want
	centered results, center covariates yourself!

2.2-1:	(November 6, 2013)
	* weibreg, phreg, aftreg: Error messages refreshed.

2.2-0: 	(October 30, 2012)
	* ltx is made generic, with methods for coxreg, phreg, and aftreg

2.1-3: 	(September 6, 2012)
	* Gompertz distribution: new parameter 'param'; values 'default'
	(default, nothing changes), and 'canonical', which gives meaning
	'PH' and 'AFT' to the shape and scale parameters, respectively.
	* aftreg: New input argument: 'param', with three values, 'default'
	(which is the default;), 'survreg', and 'canonical'. The value
	'default' uses 	the standard parametrization, while 'survreg' uses
	the same parametrization as the survreg function in the survival
	package. 'canonical' is especially useful in the Gompertz
	distribution for separating PH and AFT parameters.

2.1-2	(August 30, 2012)
	* glmmboot, glmmbootFit: Reduced the numbers of iterations in the
	examples.

2.1-1	(August 29, 2012)
	* aftreg: There was a bug when using 'dist="gompertz"' and
	the 'id' argument with more than one record per id. This is now
	hopefully solved.

2.1-0	(June 25, 2012)
	* glmmML: Added the gamma distribution to the possible priors in
	the case of the Poisson model.

2.0-7	(January 3, 2012)
	* C code cleaned (avoiding some Warnings at CRAN).
	* check.dist: New argument, col, added. Default is now
	black-and-white plotting.

2.0-6	(December 9, 2011)
	* plot.Surv: Can now handle missing values in strata variable (by
	deleting corresponding cases!)

2.0-5	(October 13, 2011)
	* ltx: New function for printing coxreg fits in latex format.
	* plot.Surv: New arguments, col and lty, added. The default is now
	black-and-white plotting.
	* data: Two new data sets, "logrye" and "scania".

2.0-4   (August 27, 2011)
	* oldmort: Data set updated.
	* aftreg.fit: Errors in documentation corrected.

2.0-3	(August 12, 2011)
	* SurvSplit: Bug fixed. It works now.
	* glmmboot: Bug in calculation of 'frail' fixed.
	* plot.phreg: Bug in argument 'ylim' fixed. Bug with stratified
	fits fixed.
	* aftreg, phreg: summary method added (just printing).

2.0-2	(August 8, 2011)
	* coxreg: 'x = TRUE' now really returns the design matrix.

2.0-1   (August 7, 2011)
	* sw_fun: C function (internal) now declared and defined.

2.0	(August 6, 2011)
	* glmmML: Merged into eha. This adds binary and count data
	regression for clustered data to eha. glmmML is still a standalone
	package on CRAN, for how long remains to be seen.
	* Pch: New class of distributions; Piecewise constant hazards (Pch).
	* phreg: piecewise constant baseline hazard function introduced.
	* phreg: 'center' reinstalled as a logical, default = TRUE, so
	nothing changes if 'center' is left untouched.
	* age.window: If the selected window is empty, an error was
	previously thrown. Now NULL is returned, with a warning.
	* cal.window: Now returns NULL, with a warning, if the selected
	window is empty.

1.4	(June 26, 2011)
	* aftreg: Gompertz distribution added (back again). Better start values.
	* phreg: Better start values for Gompertz distribution.

1.3-7 	(June 20, 2011)
	* eha: Depends (R >= 2.13.0)
	* phreg, aftreg, weibreg: Errors in documentation corrected.
	* ChangeLog: Errors in 1.3-6 corrected.

1.3-6	(June 18, 2011)
	* coxph, phreg, aftreg: Bug fix: Logical covariates now OK.
	  (converted to factor).
	* phreg: Removed "coxph" from "class".
	* aftreg: Ditto.
	* weibreg: Ditto.
	* oldmort: New data set.
	* coxreg: added return values 'df' and 'n'.
	* coxreg, phreg, aftreg: Added extractAIC and nobs methods.

1.3-5	(May 20, 2011)
	* infants: New data set containing matched data on infant and
	maternal mortality.

1.3-4	(May 10, 2011)
	* toBinary: Now includes also risksets without survivors (if any).

	* fert: Row names changed to row numbers.

	* swe07: New data set.

	* Spelling errors (some!) corrected in documentation.

1.3-3	(May 4, 2011)
	* male.mortality: Old name (male.mortality) of data frame mort
	reinstalled.
	Now known under two names. Overlapping spells no longer exist.

	* fert: Documentation corrected.

	* toDate: Now works as intended.

1.3-2	(March 1, 2011)
	* dllogis: Bug fixed.

	* hllogis: Bux fixed.

1.3-1	(January 15, 2011)
	* phreg: Fixed error and (some) numerical instability in the
	estimation of the Gompertz proportionla hazards model (seems to be
	very sensitive to good start values!).

	* plot.phreg: Fixed 'ylim' for cumulative hazards plot.

	* male.mortality: Data set renamed to

	* mort: The new name. (Just for convenience!)

1.3-0	(January 4, 2011)
	* coxreg: Major revision, calls 'coxph' in 'survival' (thanks to
	Terry Therneau!) for ordinary requests. Reason: speed. The argument
	'center' is deprecated, centering of covariates is now routinely
	made. Affects the estimation of the baseline hazard, which is done
	at the means of the columns of the original design matrix.

	* phreg: Argument 'center' is deprecated. Centering is routinely
	performed, so the baseline hazard is estimated at the means of the
	columns of the original design matrix.

	* aftreg: Corrected error in documentation and made the argument
	'center' deprecated.

	* aftreg: Estimation of intercept ("scale") is now comparable with
	corresponding estimate of Intercept in 'survreg'. 'shape' in
	aftreg is still '1 / scale' in survreg. Regression coefficients still
	have different signs but the same absolute values.

	* join.spells: a few bugs fixed, now allows for factor covariates.

	* male.mortality: data set "joined".

	* fert: is a new data set. Contains data on 19th century marital
	fertility from a parish in northern Sweden.

1.2-18	(February 28, 2010)
	* aftreg.fit: Check of data consistency in case of more than one
	record per individual added.
	* aftreg.fit: Fixed memory allocation bug in C code. This bug was
	not fatal (too much memory was allocated), but caused a severe
	performance problem, especially on Windows.

1.2-17	(February 23, 2010)
	* aftreg: 'dist = "gompertz"' should now work (realizing that it
	is a special case of 'dist = "EV").
	* phreg: Fixed bugs for the cases with fixed shape (includes 'dist
	= "gompertz"'.
	* aftreg: 'offset' works now.
	* phreg: Ditto.
	* check.dist: Bug fixed for 'dist = "gompertz"' (again!)
	* plot.phreg: Ditto.

1.2-16	(February 19, 2010)
	* aftreg: 'dist = "gompertz"' is unreliable, so changed the
	example and added a note with a warning. Is also an item on the
	ToDo list now.

1.2-15	(February 18, 2010)
	* aftreg: argument 'id' now works as intended, i.e., is looked for
	in 'data' (or 'parent.frame()')
	* print: functions return invisibly 'x'.
	* coxreg: argument 'weights' now works as intended, i.e., is looked for
	in 'data' (or 'parent.frame()')

1.2-14	(October 29, 2009)
	* check.dist: Bad 'ylim' fixed.
	* aftreg: Gompertz distribution had a bug; now fixed.

1.2-13	(October 26, 2009)
	* plot.coxreg: Bug fixed in call to plot.hazdata (wrong order of args).

1.2-12	(October 15, 2009)
	* aftreg: Works now for time-varying covariates. Not polished.

1.2-7	(August 13, 2009)

	* coxreg: Time-dependent offset introduced. 


1.2-6	(August 11, 2009)

	* coxreg.fit: Time-dependent weights work now. 

1.2-5	(August 2, 2009)

	* coxreg: Added (time-varying) weighted Cox regression. Returns
	error code 1 if failure, with a warning.    
	* check.dist: Can now handle stratified models.

1.2-4	(Jan 12, 2009)

	* Bugs in qEV, rEV, all Gompertz functions, plus documentation of
	EV and Gompertz fixed. Note that these bugs did not affect aftreg
	or phreg! However,
	* check.dist had to be revised due to the above bugs.

1.2-2	(Aug 25, 2008)

	* Bug in coxreg, method = "ml" fixed.
	* Bug in coxreg, calculation of hazard, fixed.
	* Several errors in the documentation fixed.

1.2-0	(Jul 18, 2008)

	* Added functionality: Accelerated failure time models with 
	several distributions, parametric proportional hazards models 
	with the same distributions. All models allow for left truncated 
	and right censored data, as well as for stratification. 
	Plot methods for the new models.  

1.0-1	(Jan 5, 2008) 

	* Added 'risktime' to toBinary output.

0.99	(June 9, 2007) This is the last release before version 1.0!
	
	* Added full profiling to the "ml" and "mppl" methods in
	coxreg.
	* mlreg is now deprecated. Use coxreg with method = ml or = mppl
	
0.98

0.97

0.96-7	(Dec 29, 2006)

	* Corrected a bug in the documentation of 'weibreg'.
	
0.96-6	(Dec 28, 2006)

	* Corrected a bug in weibreg concerning 'center = FALSE'.

0.96-5  (Oct 24, 2006)

        * added a 'terms' object to the output of coxreg and mlreg.

0.96-1  (May 15, 2006) 

	* plot.Surv improved. Suggestions by Dimitri Szerman. (Thanks!)

0.96-0  (April 14, 2006)

	* Weibull regression updated; It is now possible to fit a Weibull
	model to left-truncated and right-censored data without
	covariates. A small error in the documentation of 'weibreg'
	corrected. 

0.94-3	(Aug 22, 2005)

	* 'check.surv': If the 'id' was a factor with unused labels,
	the result was unpredictable. Corrected.

0.94-2  (Jun 27, 2005)

	* Errors in the documentation of 'weibreg' corrected.
	
0.93-0	(Sep 20, 2004) 

	* NAMESPACE introduced for R-2.0.0.

0.92-4 (May 5, 2004)

	* coxreg: 'Error' changed to 'warning' for a singular hessian, in
	which case the new return value is NULL.

0.92-3 (March 19, 2004)

	* table.events: New parameter: 'strict = TRUE'.

	* make.communal: In case of 'communal=FALSE', spells are no longer 
	truncated. Instead a value of NA is given if birthdate is out of
	range. Previous behavior was unpredictable when birthdate was out 
	of range.
	
0.92 (November 12, 2003)
	
	* mlreg: Geometric distribution (i.e., constant baseline discrete
	hazard) added. Not for frailty models, yet (on the TODO list).

	* mlreg: New argument, 'n.points', added to 'control'. Controls
	the number of points used in the Gauss-Hermite quadrature.

	* mlreg: Stricter control of numerical problems, especially in the
	frailty fit.
	
	* clean: Replaced by check.spells and join.spells.

	* Return values changed to conform with R-1.8.0 (and later).

