Commit f148276eccf3873b3d4b1906ea081838d3d32b1b - hmisc

Update upstream source from tag 'upstream/4.4-2' Update to upstream version '4.4-2' with Debian dir 91ae7b43dac6d8901b9ada4f3a3a4c646096a9b0 Dirk Eddelbuettel 3 years ago

20 changed file(s) with 839 addition(s) and 162 deletion(s). Raw diff Collapse all Expand all

-4

DESCRIPTION less more

0	0	Package: Hmisc
1		Version: 4.4-1
2		Date: 2020-08-10
	1	Version: 4.4-2
	2	Date: 2020-11-25
3	3	Title: Harrell Miscellaneous
4	4	Author: Frank E Harrell Jr <fh@fharrell.com>, with
5	5	contributions from Charles Dupont and many others.

22	22	Encoding: UTF-8
23	23	RoxygenNote: 7.1.1
24	24	NeedsCompilation: yes
25		Packaged: 2020-08-10 12:34:23 UTC; harrelfe
	25	Packaged: 2020-11-27 22:05:50 UTC; harrelfe
26	26	Repository: CRAN
27		Date/Publication: 2020-08-10 22:02:11 UTC
	27	Date/Publication: 2020-11-29 06:00:16 UTC

+19

-14

MD5 less more

0	0	16f8ea43efba90112351c5ecb33021c4 *COPYING
1		cef04548862dc17bed4924432f389f75 *DESCRIPTION
2		ac280b683a20ca6bf4395c2435fc30de *NAMESPACE
3		20fbb1071259a378a8c11b6c4cf62fd1 *NEWS
	1	be0222e063fcd3b8dd31e0fca3c984a3 *DESCRIPTION
	2	03aeb100e13bd948cd29ce779bb9da64 *NAMESPACE
	3	396507751e208423836df055a792b621 *NEWS
4	4	76f90acbe9bf8ad2eaaa6b8813eb7c82 *R/AFirst.lib.s
5	5	62ef16fc333551d6c65f15dedfb5a697 *R/Cs.s
6	6	85683865508739ff7a19b280e562ab67 *R/GiniMd.s

32	32	46e1eda21de21ba281f098f1f9e59bcb *R/describe.s
33	33	6cb5b3a77860b605f102bb528e84a071 *R/discrete.s
34	34	dcc7f27cea0aa7b110b0eb2be3144d8a *R/dotchart3.s
35		ab7c8cc449c83d6612b125a5da844a5b *R/dotchartpl.s
	35	266f05a603d2ddd37ad68ac269a8974d *R/dotchartpl.s
36	36	919e94c2c87f720601387c5171b57ffc *R/drawPlot.s
37	37	b44d1ac43c71a3a4d3f4820354f78598 *R/ecdf.s
38	38	c1e489a07ca933fb02e004489fd3ee8e *R/epi.s

44	44	c238614fb72943362407d74442bf236a *R/format.pval.s
45	45	aace929824aad6ebdfba910950b6cc2b *R/ftu.s
46	46	473bbef2ffd42d37d059d9a4c66fe8a4 *R/gbayes.s
	47	7d361bc56d0c8ca82e48b90493e1f0e9 *R/gbayesSeqSim.r
47	48	f76f66eae7faef0e445143d9e367619d *R/gettext.s
48	49	769b009b031f6f656a0e9b748057bfae *R/ggMisc.s
49		dc8fa7871825b028089e45097975a973 *R/ggfreqScatter.r
	50	2d9c334a5b7dddb5f8815e7bc2257c6d *R/ggfreqScatter.r
50	51	b70800bb3730a5b84500d43e97d951f4 *R/groupn.s
51	52	926950a356c2e54ecd8b2317ec08a5d8 *R/hidingTOC.r
52	53	29cf6d5ee0c8444cb58ed0afb85ac57b *R/hist.data.frame.s

84	85	1ada10fb83652630cf67024c3f1b421c *R/nobsY.s
85	86	27019e9e90730ac0b25014b51d68ec66 *R/nstr.s
86	87	cf2866606281d01f39e99d3205f50ae3 *R/num.intercepts.s
	88	a50c8048a7546e224ca402d362c084cd *R/pairUpDiff.r
87	89	215f9a06ffc6df02fb67576ad0da83b9 *R/panel.abwplot.s
88	90	41cf42963c4ceb1cb1233cc1e01b4b99 *R/panel.bpplot.s
89	91	1630c1be85251dac0e8cd0243bedd201 *R/pc1.s
90	92	25fc3a3579da1d81be1209bea9e63d2b *R/plot.describe.s
91	93	c746aa382838bbb7063eec0487f2a64d *R/plotlyM.r
92	94	62e6fe8152bc03bb23f83ff687ae4cc2 *R/plsmo.s
93		6c7831f266dc8e510adef7de3be688ef *R/popower.s
	95	8393f1ae893c0a37c8d367c1e8cd1c98 *R/popower.s
94	96	9a1119c7949407c39968645e921a87a6 *R/print.char.list.s
95	97	efe9a9a6c90594d1e7425cc46f2fc061 *R/pstamp.s
96	98	620401a9617131be8b2d162fd0466ff7 *R/rcorr.cens.s
97		06948100269e0b2705e6f0b0b1a8d4b9 *R/rcorr.s
	99	b9759f7459856f6eeb85e4ee3832f7ad *R/rcorr.s
98	100	0c9a682b270e97639d2f7d06c18aaa7f *R/rcorrp.cens.s
99	101	62264d939e616b9d1dc7070096b679e5 *R/rcspline.eval.s
100	102	bd5869619261cea0beb0bbc69f9d539f *R/rcspline.plot.s

107	109	3978f2ee3511549cb906af4618024a53 *R/rm.boot.s
108	110	a1763e6429f086dd1783694da9b85342 *R/samplesize.bin.s
109	111	130de98e37b80bad7896e6b083be8964 *R/sas.get.s
110		e0871a4ba840ef9892e26282ad7e9bd5 *R/scat1d.s
	112	bec8bca50b20965e503b4045b1cd1027 *R/scat1d.s
111	113	bcd4fb26ff4be0623165a02d1f874733 *R/score.binary.s
112	114	c1639df1f0faf833c9cb440e8367aa53 *R/sedit.s
113	115	f361dab6e800a20722ecc51bc08e056d *R/show.pch.s

127	129	b54b4f47cd9f797bb3bd49378f0367af *R/summaryM.s
128	130	f4e9b0484893f71186d0e527cfcab4b8 *R/summaryP.s
129	131	dc27150ba262b8c3844ce46d36f1e492 *R/summaryRc.s
130		f9c4fedbc00b494b1a4322d74e175ee5 *R/summaryS.s
	132	0ce190af103865fe51ba98c7bc4dab9d *R/summaryS.s
131	133	0af65ac91b7c04b13aac7791bdf758af *R/symbol.freq.s
132	134	474107a89a3a998eb78bbee1ac2dfdbe *R/sys.s
133	135	0917e77a727d026b5b12511b7710b283 *R/t.test.cluster.s

258	260	4c2942aab548faa49c5a389117b48b70 *man/discrete.Rd
259	261	94748a0c9bb7d0290a8b2dc7f4a0bb7d *man/dotchart2.Rd
260	262	67f28b3925d62e99e612e36532d3149f *man/dotchart3.Rd
261		a716e5e2bf83ff413350835c7058923c *man/dotchartpl.Rd
	263	2240c79711c6ab7bc4d204483d3d8258 *man/dotchartpl.Rd
262	264	553f6215baa4cad0ac3da850a3cbc2d2 *man/epi.Rd
263	265	36c0c98e4e6fcc78979da741c8dfb386 *man/equalBins.Rd
264	266	04771822c60d38658c3720cd91bc2113 *man/errbar.Rd
265	267	3bc08c67bb196770cb9ff90bdfa15672 *man/escapeRegex.Rd
	268	7529c6702e5cd078996c56b78b826c78 *man/estSeqSim.Rd
266	269	b8c947ebb9ea63a9ecf543e9248a512b *man/event.chart.Rd
267	270	6f9b356b12e05d93c091e033a7d3f4a8 *man/event.convert.Rd
268	271	1beee8ab5e6bde656f0daaf7dfe1cbc5 *man/event.history.Rd

270	273	6a5adf7af2c0cc119bbcded8040bd34f *man/first.word.Rd
271	274	262cfe8e5568594c1ebf3915f81f693a *man/format.df.Rd
272	275	041f76e1a74ad1e2361aab071b496dac *man/format.pval.Rd
273		4f6bea3b84c269f61ef89eaf205d8f00 *man/gbayes.Rd
	276	ebdd21800ffc4beea994eaac7d5b4668 *man/gbayes.Rd
	277	5522826f50906b5ec42f9679d29a5af1 *man/gbayesSeqSim.Rd
274	278	65798e0cb002195035db2991427ce63c *man/getHdata.Rd
275	279	993b4834e5bbff1af338a24d08cbb6cf *man/getRs.Rd
276	280	fa2378b377a46a23ee77876829ebf8c4 *man/getZip.Rd
277	281	ac60ee276185df5b27c0819a3e781bae *man/ggMisc.Rd
278		80e90e865489fc64f8414fa058030112 *man/ggfreqScatter.Rd
	282	10239b9be9e48b4526e454000124280c *man/ggfreqScatter.Rd
279	283	bd30d6597a6591b2d3b6202931502245 *man/hdquantile.Rd
280	284	9bde85cef5a8815e6a4adbc2163aa870 *man/hidingTOC.Rd
281	285	8256dd2a02e789ad86ec974fa90420cd *man/hist.data.frame.Rd

309	313	be66aaec2a64401335915c4671d50918 *man/nobsY.Rd
310	314	2f7e82883fd0be5a86a3bec4d4dc250a *man/nstr.Rd
311	315	1d560c844c38c8f7289bb27e21efe171 *man/num.intercepts.Rd
	316	76b1975307ab89cc2eb19f12040342ce *man/pairUpDiff.Rd
312	317	ab8edbb03663977e40ac2ffa3025d39f *man/panel.bpplot.Rd
313	318	b8247448b0c2c6d876ce693f23c9fe85 *man/partition.Rd
314	319	b34470017ff70767108ad8ee9e169f02 *man/pc1.Rd
315	320	17652e750bd0249ced3eac8e145d070d *man/plotCorrPrecision.Rd
316	321	88d93c5117d432cbe4ba996c5c74d0d8 *man/plotlyM.Rd
317	322	81937c2b724d0ec42f817ebb88748e70 *man/plsmo.Rd
318		53e306cc4774582993e485b474988555 *man/popower.Rd
	323	0a2e6d1c460ef2be08a7162c85709e43 *man/popower.Rd
319	324	52be83076b1f0442cb5ab5c15bb16a3f *man/print.char.list.Rd
320	325	0a25e1c6f349789159b7239ae0f0f620 *man/print.char.matrix.Rd
321	326	057da98d466a023bda3991abbf880855 *man/prnz.Rd

334	339	0363a389eb98c4c065a4cecae63c7e56 *man/rm.boot.Rd
335	340	326928a75e512ba40912a48fc355934e *man/samplesize.bin.Rd
336	341	89c4eefbaac065e35e5becf8f6e5319d *man/sasxport.get.Rd
337		c7247e422a999499fc292af52da803a8 *man/scat1d.Rd
	342	ca36423dc97de82f1aef3c6d260c7dd8 *man/scat1d.Rd
338	343	db05fb7a0b4ab6f2e3296104ab70dca1 *man/score.binary.Rd
339	344	08744258b9fd7ca14e1b32ba62ca245f *man/sedit.Rd
340	345	8dbcd26d8e34394832eaa6c2f32931a3 *man/show.pch.Rd

-1

NAMESPACE less more

0		export("%nin%",abs.error.pred,addMarginal,all.digits,all.is.numeric,approxExtrap,areg,areg.boot,aregImpute,aregTran,arrGrob,as.discrete,asNumericMatrix,ballocation,bezier,binconf,biVar,bootkm,bpower,bpower.sim,bpplot,bpplotM,bpplt,bppltp,bpx,bsamsize,bystats,bystats2,capitalize,catTestchisq,Cbind,ceil,character.table,chiSquare,ciapower,cleanup.import,clowess,cnvrt.coords,code.levels,colorFacet,combine.levels,combineLabels,combplotp,confbar,consolidate,contents,conTestkw,cpower,Cs,csv.get,cumcategory,curveRep,curveSmooth,cut2,datadensity,dataDensityString,dataframeReduce,dataRep,ddmmmyy,deff,describe,discrete,dhistboxp,dotchart2,dotchart3,dotchartp,dotchartpl,Dotplot,drawPlot,dvi,dvigv,dvips,ecdfpM,Ecdf,equalBins,errbar,escapeBS,escapeRegex,event.chart,event.convert,event.history,expr.tree,fillin,find.matches,first.word,fit.mult.impute,format.df,format.pval,formatCats,formatCons,formatDateTime,formatdescribeSingle,formatSep,formatTestStats,ftupwr,ftuss,Function,gbayes,gbayes1PowerNP,gbayes2,gbayesMixPost,gbayesMixPowerNP,gbayesMixPredNoData,get2rowHeads,getHdata,getLatestSource,GetModelFrame,getRs,getZip,ggfreqScatter,GiniMd,Gompertz2,groupn,grType,hdquantile,hidingTOC,hist.data.frame,histbackback,histboxp,histboxpM,histSpike,histSpikeg,hoeffd,html,htmlGreek,htmlSN,htmlSpecial,htmlSpecialType,htmlTranslate,htmlVerbatim,importConvertDateTime,improveProb,impute,impute.transcan,inmChoice,inverseFunction,invertTabulated,is.discrete,is.imputed,is.mChoice,is.present,is.special.miss,james.stein,jitter2,keepHattrib,Key,Key2,km.quick,knitrSet,labcurve,label,Label,labelLatex,labelPlotmath,Lag,largest.empty,latex,latexBuild,latexCheckOptions,latexDotchart,latexNeedle,latexSN,latexTabular,latexTherm,latexTranslate,latexVerbatim,list.tree,llist,lm.fit.qr.bare,Load,Lognorm2,logrank,lookupSASContents,lrcum,makeNames,makeNstr,makeSteps,mApply,markupSpecs,mask,match.mChoice,matchCases,matrix2dataFrame,matxv,mbarclPanel,mbarclpl,mChoice,mdb.get,Mean,medvPanel,medvpl,Merge,mgp.axis,mgp.axis.labels,mhgr,minor.tick,monotone,monthDays,mtitle,multEventChart,multLines,na.delete,na.detail.response,na.include,na.keep,na.pattern,na.retain,naclus,nafitted.delete,Names2names,naplot,napredict.delete,naprint.delete,naprint.keep,naresid.delete,naresid.keep,nFm,nobsY,nomiss,nstr,num.denom.setup,num.intercepts,numeric.string,numericScale,oPar,optionsCmds,ordGridFun,ordTestpo,outerText,panel.bpplot,panel.Dotplot,panel.Ecdf,panel.plsmo,panel.xYplot,parGrid,partition.matrix,partition.vector,pasteFit,pBlock,pc1,plotCorrPrecision,plotp,plotlyM,plotlyParm,plotlySave,plotmathTranslate,plotMultSim,plotpsummaryM,plsmo,pngNeedle,pomodm,popower,posamsize,prepanel.Dotplot,prepanel.Ecdf,prepanel.xYplot,print.char.matrix,prList,prn,propsPO,propsTrans,prselect,prType,pstamp,putHcap,putHfig,putKey,putKeyEmpty,Quantile,Quantile2,rcorr,rcorr.cens,rcorrcens,rcorrp.cens,rcspline.eval,rcspline.plot,rcspline.restate,rcsplineFunction,read.xportDataload,readSAScsv,redun,reformM,reLabelled,replace.substring.wild,reShape,responseSummary,restoreHattrib,rlegend,rlegendg,rm.boot,rMultinom,roundN,roundPOSIXt,samplesize.bin,sas.codes,sas.get,sas.get.macro,sasdsLabels,sasxport.get,Save,scat1d,score.binary,sedit,sepUnitsTrans,setParNro,setTrellis,show.col,show.pch,showPsfrag,simplifyDims,simRegOrd,sKey,smean.cl.boot,smean.cl.normal,smean.sd,smean.sdl,smearingEst,smedian.hilow,solvet,somers2,spearman,spearman.test,spearman2,spower,spss.get,src,stat_plsmo,stata.get,StatPlsmo,stepfun.eval,stratify,strgraphwrap,string.bounding.box,string.break.line,stringDims,stripChart,subplot,substi,substi.source,substring.location,substring2,summarize,summaryD,summaryDp,summaryM,summaryP,summaryRc,summaryS,symbol.freq,sys,t.test.cluster,table_formatpct,table_freq,table_latexdefs,table_N,table_pc,table_trio,tabulr,termsDrop,testDateTime,tex,tobase64image,transace,transcan,translate,trap.rule,trellis.strip.blank,truncPOSIXt,uncbind,units,unPaste,upData,upFirst,valueLabel,valueName,valueTags,valueUnit,var.inner,varclus,Weibull2,whichClosek,whichClosePW,whichClosest,wtd.Ecdf,wtd.loess.noiter,wtd.mean,wtd.quantile,wtd.rank,wtd.table,wtd.var,xInch,xless,xy.group,xYplot,xySortNoDupNoNA,yearDays,yInch,ynbind,zoom,"[<-.discrete","consolidate<-","is.na<-.discrete","label<-","length<-.discrete","substring2<-","valueLabel<-","valueName<-","valueTags<-","valueUnit<-")
	0	export("%nin%",abs.error.pred,addMarginal,all.digits,all.is.numeric,approxExtrap,areg,areg.boot,aregImpute,aregTran,arrGrob,as.discrete,asNumericMatrix,ballocation,bezier,binconf,biVar,bootkm,bpower,bpower.sim,bpplot,bpplotM,bpplt,bppltp,bpx,bsamsize,bystats,bystats2,capitalize,catTestchisq,Cbind,ceil,character.table,chiSquare,ciapower,cleanup.import,clowess,cnvrt.coords,code.levels,colorFacet,combine.levels,combineLabels,combplotp,confbar,consolidate,contents,conTestkw,cpower,Cs,csv.get,cumcategory,curveRep,curveSmooth,cut2,datadensity,dataDensityString,dataframeReduce,dataRep,ddmmmyy,deff,describe,discrete,dhistboxp,dotchart2,dotchart3,dotchartp,dotchartpl,Dotplot,drawPlot,dvi,dvigv,dvips,ecdfpM,Ecdf,equalBins,errbar,escapeBS,escapeRegex,estSeqSim,event.chart,event.convert,event.history,expr.tree,fillin,find.matches,first.word,fit.mult.impute,format.df,format.pval,formatCats,formatCons,formatDateTime,formatdescribeSingle,formatSep,formatTestStats,ftupwr,ftuss,Function,gbayes,gbayes1PowerNP,gbayes2,gbayesMixPost,gbayesMixPowerNP,gbayesMixPredNoData,gbayesSeqSim,get2rowHeads,getHdata,getLatestSource,GetModelFrame,getRs,getZip,ggfreqScatter,GiniMd,Gompertz2,groupn,grType,hdquantile,hidingTOC,hist.data.frame,histbackback,histboxp,histboxpM,histSpike,histSpikeg,hoeffd,html,htmlGreek,htmlSN,htmlSpecial,htmlSpecialType,htmlTranslate,htmlVerbatim,importConvertDateTime,improveProb,impute,impute.transcan,inmChoice,inverseFunction,invertTabulated,is.discrete,is.imputed,is.mChoice,is.present,is.special.miss,james.stein,jitter2,keepHattrib,Key,Key2,km.quick,knitrSet,labcurve,label,Label,labelLatex,labelPlotmath,Lag,largest.empty,latex,latexBuild,latexCheckOptions,latexDotchart,latexNeedle,latexSN,latexTabular,latexTherm,latexTranslate,latexVerbatim,list.tree,llist,lm.fit.qr.bare,Load,Lognorm2,logrank,lookupSASContents,lrcum,makeNames,makeNstr,makeSteps,mApply,markupSpecs,mask,match.mChoice,matchCases,matrix2dataFrame,matxv,mbarclPanel,mbarclpl,mChoice,mdb.get,Mean,medvPanel,medvpl,Merge,mgp.axis,mgp.axis.labels,mhgr,minor.tick,monotone,monthDays,mtitle,multEventChart,multLines,na.delete,na.detail.response,na.include,na.keep,na.pattern,na.retain,naclus,nafitted.delete,Names2names,naplot,napredict.delete,naprint.delete,naprint.keep,naresid.delete,naresid.keep,nFm,nobsY,nomiss,nstr,num.denom.setup,num.intercepts,numeric.string,numericScale,oPar,optionsCmds,ordGridFun,ordTestpo,outerText,pairUpDiff,panel.bpplot,panel.Dotplot,panel.Ecdf,panel.plsmo,panel.xYplot,parGrid,partition.matrix,partition.vector,pasteFit,pBlock,pc1,plotCorrPrecision,plotp,plotlyM,plotlyParm,plotlySave,plotmathTranslate,plotMultSim,plotpsummaryM,plsmo,pngNeedle,pomodm,popower,posamsize,prepanel.Dotplot,prepanel.Ecdf,prepanel.xYplot,print.char.matrix,prList,prn,propsPO,propsTrans,prselect,prType,pstamp,putHcap,putHfig,putKey,putKeyEmpty,Quantile,Quantile2,rcorr,rcorr.cens,rcorrcens,rcorrp.cens,rcspline.eval,rcspline.plot,rcspline.restate,rcsplineFunction,read.xportDataload,readSAScsv,redun,reformM,reLabelled,replace.substring.wild,reShape,responseSummary,restoreHattrib,rlegend,rlegendg,rm.boot,rMultinom,roundN,roundPOSIXt,samplesize.bin,sas.codes,sas.get,sas.get.macro,sasdsLabels,sasxport.get,Save,scat1d,score.binary,sedit,sepUnitsTrans,setParNro,setTrellis,show.col,show.pch,showPsfrag,simplifyDims,simRegOrd,sKey,smean.cl.boot,smean.cl.normal,smean.sd,smean.sdl,smearingEst,smedian.hilow,solvet,somers2,spearman,spearman.test,spearman2,spower,spss.get,src,stat_plsmo,stata.get,StatPlsmo,stepfun.eval,stratify,strgraphwrap,string.bounding.box,string.break.line,stringDims,stripChart,subplot,substi,substi.source,substring.location,substring2,summarize,summaryD,summaryDp,summaryM,summaryP,summaryRc,summaryS,symbol.freq,sys,t.test.cluster,table_formatpct,table_freq,table_latexdefs,table_N,table_pc,table_trio,tabulr,termsDrop,testDateTime,tex,tobase64image,transace,transcan,translate,trap.rule,trellis.strip.blank,truncPOSIXt,uncbind,units,unPaste,upData,upFirst,valueLabel,valueName,valueTags,valueUnit,var.inner,varclus,Weibull2,whichClosek,whichClosePW,whichClosest,wtd.Ecdf,wtd.loess.noiter,wtd.mean,wtd.quantile,wtd.rank,wtd.table,wtd.var,xInch,xless,xy.group,xYplot,xySortNoDupNoNA,yearDays,yInch,ynbind,zoom,"[<-.discrete","consolidate<-","is.na<-.discrete","label<-","length<-.discrete","substring2<-","valueLabel<-","valueName<-","valueTags<-","valueUnit<-")
1	1
2	2	useDynLib(Hmisc, .registration=TRUE, .fixes="F_")
3	3

+11

-0

NEWS less more

	0	Changes in version 4.4-2 (2020-11-25)
	1	* rcorr: captured loss of precision leading to square root of a negative number. Thanks: Ann Voss <avoss@novanthealth.org>
	2	* summaryS: sapply was doubling names
	3	* pairUpDiff: created for dotchartpl - function to pair up grouped observations for sorting by descending differences and computing approximate confidence intervals for the difference given individual confidence limits
	4	* dotchartpl: added new arguments and functionality to handle quantities other than proportions (e.g., hazards) and streamlined code using pairUpDiff
	5	* propsPO: added tooltip text for gpplot that will be transmitted to ggplotly; reversed order of categories so lowest category put at bottom of bar
	6	* dotchartpl: suppressed confidence intervals when the number of groups is not 2; fixed bug where hover text confidence intervals were repeats of the last computed interval instead of properly using all the intervals; added dec argument
	7	* added estSeqSim and gbayesSeqSim functions
	8	* ggfreqScatter: stopped varying alpha and just varied color, as alpha collided with the color scheme
	9	* histSpike: added minimal=TRUE and bins arguments
	10
0	11	Changes in version 4.4-1 (2020-08-07)
1	12	* popower: added approximate S.E. of log(OR) in results
2	13	* propsPO: new function for exploring proportional odds

+96

-99

R/dotchartpl.s less more

0		dotchartpl <- function(x, major=NULL, minor=NULL, group=NULL, mult=NULL,
	0	dotchartpl <- function(x, major=NULL, minor=NULL,
	1	group=NULL, mult=NULL,
1	2	big=NULL, htext=NULL,
2	3	num=NULL, denom=NULL,
	4	numlabel='', denomlabel='',
	5	fun=function(x) x, ifun=function(x) x,
	6	op='-',
3	7	lower=NULL, upper=NULL,
4	8	refgroup=NULL, sortdiff=TRUE, conf.int=0.95,
5	9	minkeep=NULL,
6	10	xlim=NULL, xlab='Proportion',
7	11	tracename=NULL, limitstracename='Limits',
8	12	nonbigtracename='Stratified Estimates',
9		width=800, height=NULL,
	13	dec=3, width=800, height=NULL,
10	14	col=colorspace::rainbow_hcl
11	15	) {
	16
12	17	mu <- markupSpecs$html
13	18	bold <- mu$bold
14	19

16	21	max(c(x, upper), na.rm=TRUE))
17	22
18	23	majorpres <- length(major) > 0
19		major <- if(majorpres) as.character(major) else rep('', length(x))
	24	major <- if(majorpres) as.character(major) else rep(' ', length(x))
20	25	minorpres <- length(minor) > 0
	26	if(! (majorpres \|\| minorpres)) stop('must specify major or minor or both')
	27
21	28	grouppres <- length(group) > 0 ## superpositioning variable
22	29	multpres <- length(mult) > 0
23	30	limspres <- length(lower) * length(upper) > 0
24	31	rgpres <- length(refgroup) > 0
25	32
26		if(minorpres) minor <- as.character(minor)
27		if(grouppres) group <- as.character(group)
28		if(multpres) mult <- as.character(mult)
29		ugroup <- if(grouppres) unique(group)
	33	if(minorpres) minor <- as.character(minor)
	34	if(grouppres) group <- as.character(group)
	35	if(multpres) mult <- as.character(mult)
	36	ugroup <- if(grouppres) unique(group) else ''
	37
	38	fmt <- function(x) format(round(x, dec))
30	39
31	40	if(rgpres) {
32	41	if(! grouppres \|\| multpres \|\| length(big))

48	57	col(1) }
49	58	if(! length(col) && ! grouppres) cols <- 'black'
50	59
51		levelsRemoved <- character(0)
52
53		if(rgpres && (sortdiff \|\| length(minkeep)) && minorpres) {
54		## For each major x minor subgroup, hold on to x for both groups
55
56		g <- function(i, wcat) {
57		gr <- group[i]
58		if(any(gr == wcat)) x[i[gr == wcat]] else NA
59		}
60
61		xa <- tapply(1 : length(x), llist(major, minor), g, wcat=refgroup)
62		xb <- tapply(1 : length(x), llist(major, minor), g, wcat=altgroup)
63
64		## Bug in x[cbind()] if major is constant
65		if(! majorpres) {
66		xa <- xa[, minor]
67		xb <- xb[, minor]
68		}
69		else {
70		xa <- xa[cbind(major, minor)]
71		xb <- xb[cbind(major, minor)]
72		}
73
74		if(length(minkeep)) {
75		w <- if(majorpres) paste0(major, ':', minor) else minor
76		levelsRemoved <- unique(w[! is.na(xa + xb) &
77		xa < minkeep & xb < minkeep])
78		j <- which(! is.na(xa + xb) & (xa >= minkeep \| xb >= minkeep))
79		}
80		else
81		j <- 1 : length(x)
82
83		## Sort minor categories by descending order of between-group differences
84
85		i <- order(major, ifelse(is.na(xa + xb), -Inf, - (xb - xa)))
86		i <- intersect(i, j)
	60	dropped <- character(0)
	61	D <- NULL
	62
	63	if(rgpres && minorpres) {
	64
	65	z <- pairUpDiff(x, major, minor, group,
	66	refgroup=refgroup, lower=lower, upper=upper,
	67	minkeep=minkeep, sortdiff=sortdiff, conf.int=conf.int)
	68
	69	Z <- z$X
	70	D <- z$D
	71	dropped <- z$dropped
	72	i <- Z[,'subscripts']
	73
87	74	x <- x[i]
88	75	major <- major[i]
89	76	minor <- minor[i]

95	82	if(length(lower)) lower <- lower[i]
96	83	if(length(upper)) upper <- upper[i]
97	84	if(length(htext)) htext <- htext[i]
98		}
	85	} # end if(rgpres && ...
99	86
100	87
101	88	ht <- htext
	89	if(numlabel != '') numlabel <- paste0(' ', numlabel)
	90	if(denomlabel != '') denomlabel <- paste0(' ', denomlabel)
102	91	if(length(num))
103	92	ht <- paste0(ht, if(length(htext)) mu$lspace,
104		round(x, 3), mu$lspace,
105		mu$frac(num, denom, size=95))
	93	fmt(fun(x)), mu$lspace,
	94	mu$frac(paste0(num, numlabel),
	95	paste0(denom, denomlabel), size=95))
	96
	97	## if confidence interval for differences are not to be displayed,
	98	## put point estimate confidence intervals in point hypertext
	99	if(length(ugroup) != 2 && limspres)
	100	ht <- paste0(ht, ' [',
	101	fmt(fun(lower)), ', ',
	102	fmt(fun(upper)), ']')
	103
106	104	ht <- paste0(ht, if(length(ht)) '<br>',
107	105	if(majorpres) paste0(major, ': '))
108	106	if(minorpres) ht <- paste0(ht, minor)

148	146	y <- y - 0.4
149	147	lines <- lines + (mi != '')
150	148	j <- which(major == ma & minor == mi)
151		m <- length(j)
152	149	X <- c(X, x[j])
153	150	Y <- c(Y, ifelse(big[j], y, y - .14))
154		if(rgpres) {
155		ja <- which(major == ma & minor == mi & group == refgroup)
156		jb <- which(major == ma & minor == mi & group == altgroup)
157		diff <- x[jb] - x[ja]
	151	if(length(D)) {
	152	k <- which(D$major == ma & D$minor == mi)
	153	if(length(k) != 1)
	154	stop('must have one observation for a major/minor/group combination')
	155	diff <- D$diff[k]
158	156	coldiff <- c(coldiff,
159	157	ifelse(diff > 0,
160	158	paste0(altgroup, ' ',

165	163
166	164	htd <- if(majorpres) paste0(ma, ': ') else ''
167	165	if(minorpres) htd <- paste0(htd, mi)
168
169		htd <- paste0(htd, '<br>', altgroup, ' - ', refgroup, ': ',
170		round(diff, 3))
171
172		if(! is.logical(conf.int) && length(num) && length(denom)) {
173		zcrit <- qnorm((1 + conf.int) / 2)
174		## If either proportion is 0 or 1, backsolve variance from
175		## Wilson interval
176		wi <- function(p, n) {
177		if(p == 0 \|\| p == 1) {
178		ci <- binconf(round(n * p), n)[1, ]
179		w <- ci['Upper'] - ci['Lower']
180		(0.5 * w / zcrit) ^ 2
181		} else p * (1. - p) / n
182		}
183		va <- wi(x[ja], denom[ja])
184		vb <- wi(x[jb], denom[jb])
185		se <- sqrt(va + vb)
186		dlower <- format(round(x[jb] - x[ja] - zcrit * se, 3), nsmall=3)
187		dupper <- format(round(x[jb] - x[ja] + zcrit * se, 3), nsmall=3)
	166
	167	htd <- paste0(htd, '<br>', altgroup, ' ', op, ' ',
	168	refgroup, ': ', fmt(fun(diff)))
	169	if(! is.logical(conf.int) && limspres && length(D)) {
	170	dlower <- fmt(fun(D$lower[k]))
	171	dupper <- fmt(fun(D$upper[k]))
188	172	htd <- paste0(htd, '<br>', conf.int, ' C.L.: [', dlower,
189	173	', ', dupper, ']')
190		Diff <- c(Diff, x[jb] - x[ja])
191		difflower <- c(difflower, (x[jb] + x[ja]) / 2 - 0.5 * zcrit * se)
192		diffupper <- c(diffupper, (x[jb] + x[ja]) / 2 + 0.5 * zcrit * se)
	174	Diff <- c(Diff, diff)
	175	difflower <- c(difflower, D$lowermid[k])
	176	diffupper <- c(diffupper, D$uppermid[k])
193	177	}
194	178	htdiff <- c(htdiff, htd)
195		}
196		if(limspres) {
	179	} # end if(length(D))
	180	if(limspres && ! length(D)) {
197	181	Lower <- c(Lower, lower[j])
198	182	Upper <- c(Upper, upper[j])
199		Htextl <- paste0('[', format(lower[j], digits=4), ', ',
200		format(upper[j], digits=4), ']')
	183	Htextl <- c(Htextl,
	184	paste0('[',
	185	fmt(fun(lower[j])), ', ',
	186	fmt(fun(upper[j])), ']' ) )
201	187	}
202	188	Group <- c(Group, group[j])
203		Big <- c(Big, big[j])
	189	Big <- c(Big, big[j])
204	190	Htext <- c(Htext, ht[j])
205	191
206	192	if(mi != '') {

213	199
214	200	d <- data.frame(X, Y, Group, Htext, Big)
215	201
216		if(limspres) {
	202	if(limspres && ! length(D)) {
217	203	d$Lower <- Lower
218	204	d$Upper <- Upper
219	205	d$Htextl <- Htextl
220	206	}
221
222		if(! grouppres) d$Group <- NULL ####
223		if(any(d$Big)) {
	207
	208	if(! grouppres) d$Group <- NULL
	209
	210	if(any(d$Big)) {
224	211	db <- subset(d, Big) # non-stratified estimates
225	212	## For some reason, colors= in add_marker did not always take
226	213	p <- plotly::plot_ly(data=db, colors=cols,
227	214	height=if(length(height)) height else
228	215	plotlyParm$heightDotchart(lines), width=width)
229
230		if(length(difflower)) {
	216
	217	if(limspres && length(D)) {
231	218	ddiff <- data.frame(Diff, difflower, diffupper, Ydiff, coldiff, htdiff)
	219
232	220	## Could not get color=coldiff to work; perhaps conflicted with
233	221	## earlier use of color =
	222
234	223	if(any(Diff > 0))
235	224	p <- plotly::add_segments(p, data=subset(ddiff, Diff > 0),
236	225	x= ~ difflower, xend= ~ diffupper,

240	229	name = paste0(htmlSpecial('half'),
241	230	' CL of difference<br>',
242	231	coldiff[Diff > 0][1]))
	232
243	233	if(any(Diff <= 0))
244	234	p <- plotly::add_segments(p, data=subset(ddiff, Diff <= 0),
245	235	x= ~ difflower, xend= ~ diffupper,

250	240	' CL of difference<br>',
251	241	coldiff[Diff <= 0][1]))
252	242	}
	243
253	244	## tracename and limitstracename are used if groups not used
254		if(limspres)
	245
	246	if(limspres && ! length(D) && length(ugroup) == 2)
255	247	p <- if(grouppres)
256		plotly::add_segments(p, x=~ Lower, xend=~ Upper,
	248	plotly::add_segments(p, data=db,
	249	x=~ Lower, xend=~ Upper,
257	250	y=~ Y, yend=~ Y,
258	251	color=~ Group, text= ~ Htextl,
259	252	colors=cols, hoverinfo='text')

263	256	text= ~ Htextl,
264	257	color=I('lightgray'), hoverinfo='text',
265	258	name=limitstracename)
266
	259
267	260	p <- if(grouppres)
268	261	plotly::add_markers(p, x=~ X, y=~ Y,
269	262	color=~ Group, text=~ Htext,

275	268	name=if(length(tracename)) tracename
276	269	else
277	270	if(any(! d$Big)) 'All' else '')
278
	271
279	272	} else p <- plotly::plot_ly(colors=cols) # Big is not used
280	273
281	274	if(any(! d$Big)) {
282	275	dnb <- subset(d, ! Big) # stratified estimates
283	276	if(limspres)
284		p <- if(grouppres)
	277	p <- if(grouppres && length(ugroup) == 2)
285	278	plotly::add_segments(p, data=dnb,
286	279	x=~ Lower, xend=~ Upper,
287	280	y=~ Y, yend=~ Y,

314	307	marker=list(opacity=0.45, size=4),
315	308	color=I('black'), hoverinfo='text',
316	309	name=nonbigtracename)
317
318		} # end if(any(! Big))
	310
	311	} # end if(any(! Big))
	312
319	313	leftmargin <- plotlyParm$lrmargin(ytnb)
	314	tickvals <- pretty(fun(xlim), n=10)
	315	xaxis <- list(title=xlab, range=xlim, zeroline=FALSE,
	316	tickvals=ifun(tickvals), ticktext=tickvals)
	317	yaxis <- list(title='',
	318	range=c(min(Y) - 0.2, 0.2),
	319	zeroline=FALSE, tickvals=yl, ticktext=yt)
	320
320	321	p <- plotly::layout(p,
321		xaxis=list(title=xlab,
322		range=xlim,
323		zeroline=FALSE),
324		yaxis=list(title='',
325		range=c(min(Y) - 0.2, 0.2),
326		zeroline=FALSE, tickvals=yl, ticktext=yt),
	322	xaxis=xaxis,
	323	yaxis=yaxis,
327	324	margin=list(l=leftmargin),
328	325	legend=list(traceorder=if(length(difflower))
329	326	'reversed' else 'normal'))
330	327
331		attr(p, 'levelsRemoved') <- levelsRemoved
	328	attr(p, 'levelsRemoved') <- dropped
332	329	p
333	330	}

+215

-0

R/gbayesSeqSim.r less more

	0	##' Simulate Bayesian Sequential Treatment Comparisons Using a Gaussian Model
	1	##'
	2	##' Simulate a sequential trial under a Gaussian model for parameter estimates, and Gaussian priors using simulated estimates and variances returned by `estSeqSim`. For each row of the data frame `est` and for each prior/assertion combination, computes the posterior probability of the assertion.
	3	##' @title gbayesSeqSim
	4	##' @param est data frame created by `estSeqSim()`
	5	##' @param asserts list of lists. The first element of each list is the user-specified name for each assertion/prior combination, e.g., `"efficacy"`. The other elements are, in order, a character string equal to "<", ">", or "in", a parameter value `cutoff` (for "<" and ">") or a 2-vector specifying an interval for "in", and either a prior distribution mean and standard deviation named `mu` and `sigma` respectively, or a parameter value (`"cutprior"`) and tail area `"tailprob"`. If the latter is used, `mu` is assumed to be zero and `sigma` is solved for such that P(parameter > 'cutprior') = P(parameter < - 'cutprior') = `tailprob`.
	6	##' @return a data frame with number of rows equal to that of `est` with a number of new columns equal to the number of assertions added. The new columns are named `p1`, `p2`, `p3`, ... (posterior probabilities), `mean1`, `mean2`, ... (posterior means), and `sd1`, `sd2`, ... (posterior standard deviations). The returned data frame also has an attribute `asserts` added which is the original `asserts` augmented with any derived `mu` and `sigma` and converted to a data frame, and another attribute `alabels` which is a named vector used to map `p1`, `p2`, ... to the user-provided labels in `asserts`.
	7	##' @author Frank Harrell
	8	##' @seealso `gbayes()`, `estSeqSim()`
	9	##' @examples
	10	##' \dontrun{
	11	##' # Simulate Bayesian operating characteristics for an unadjusted
	12	##' # proportional odds comparison (Wilcoxon test)
	13	##' # For 100 simulations, 5 looks, 2 true parameter values, and
	14	##' # 2 assertion/prior combinations, compute the posterior probability
	15	##' # Use a low-level logistic regression call to speed up simuluations
	16	##' # Use data.table to compute various summary measures
	17	##' # Total simulation time: 2s
	18	##' lfit <- function(x, y) {
	19	##' f <- rms::lrm.fit(x, y)
	20	##' k <- length(coef(f))
	21	##' c(coef(f)[k], vcov(f)[k, k])
	22	##' }
	23	##' gdat <- function(beta, n1, n2) {
	24	##' # Cell probabilities for a 7-category ordinal outcome for the control group
	25	##' p <- c(2, 1, 2, 7, 8, 38, 42) / 100
	26	##'
	27	##' # Compute cell probabilities for the treated group
	28	##' p2 <- pomodm(p=p, odds.ratio=exp(beta))
	29	##' y1 <- sample(1 : 7, n1, p, replace=TRUE)
	30	##' y2 <- sample(1 : 7, n2, p2, replace=TRUE)
	31	##' list(y1=y1, y2=y2)
	32	##' }
	33	##'
	34	##' # Assertion 1: log(OR) < 0 under prior with prior mean 0.1 and sigma 1 on log OR scale
	35	##' # Assertion 2: OR between 0.9 and 1/0.9 with prior mean 0 and sigma computed so that
	36	##' # P(OR > 2) = 0.05
	37	##' asserts <- list(list('Efficacy', '<', 0, mu=0.1, sigma=1),
	38	##' list('Similarity', 'in', log(c(0.9, 1/0.9)),
	39	##' cutprior=log(2), tailprob=0.05))
	40	##'
	41	##' set.seed(1)
	42	##' est <- estSeqSim(c(0, log(0.7)), looks=c(50, 75, 95, 100, 200),
	43	##' gendat=gdat,
	44	##' fitter=lfit, nsim=100)
	45	##' z <- gbayesSeqSim(est, asserts)
	46	##' head(z)
	47	##' attr(z, 'asserts')
	48	##'
	49	##' # Compute the proportion of simulations that hit targets (different target posterior
	50	##' # probabilities for efficacy vs. similarity)
	51	##'
	52	##' # For the efficacy assessment compute the first look at which the target
	53	##' # was hit (set to infinity if never hit)
	54	##' require(data.table)
	55	##' z <- data.table(z)
	56	##' u <- z[, .(first=min(p1 > 0.95])), by=.(parameter, sim)]
	57	##' # Compute the proportion of simulations that ever hit the target and
	58	##' # that hit it by the 100th subject
	59	##' u[, .(ever=mean(first < Inf)), by=.(parameter)]
	60	##' u[, .(by75=mean(first <= 100)), by=.(parameter)]
	61	##' }
	62	##' @md
	63	gbayesSeqSim <- function(est, asserts) {
	64	nas <- length(asserts)
	65	alabels <- character(nas)
	66	for(i in 1 : nas) {
	67	a <- asserts[[i]]
	68	nam <- names(a)
	69	if(nam[1] == '') nam[1] <- 'label'
	70	if(nam[2] == '') nam[2] <- 'dir'
	71	if(nam[3] == '') nam[3] <- 'cutoff'
	72	names(a) <- nam
	73	if((length(a$cutoff) == 2) != (a$dir == 'in'))
	74	stop('mismatch of direction and length of cutoff in asserts')
	75	if(any(c('mu', 'sigma') %nin% names(a))) {
	76	a$mu <- 0
	77	a$sigma <- - a$cutprior / qnorm(a$tailprob)
	78	if(a$sigma <= 0) stop('error in specification of cutoff or tailprob')
	79	}
	80	w <- format(round(a$cutoff, 3))
	81	a$assertion <- paste(if(a$dir %in% c('<','>')) a$dir,
	82	if(length(a$cutoff) == 2)
	83	paste0('[', w[1], ',', w[2], ']')
	84	else
	85	w[1])
	86	asserts[[i]] <- a
	87	alabels[i] <- a$label
	88	}
	89
	90	N <- nrow(est)
	91
	92	ests <- est$est
	93	vests <- est$vest
	94	## For each simulated parameter estimate and variance compute nas
	95	## posterior probabilities
	96
	97	for(i in 1 : nas) {
	98	a <- asserts[[i]]
	99	dir <- a$dir
	100	cutoff <- a$cutoff
	101	mu <- a$mu
	102	sigma2 <- (a$sigma) ^ 2
	103	var.post <- 1. / (1. / sigma2 + 1. / vests)
	104	mean.post <- (mu / sigma2 + ests / vests) * var.post
	105	sd.post <- sqrt(var.post)
	106
	107	pp <-
	108	switch(dir,
	109	'<' = pnorm(cutoff, mean.post, sd.post),
	110	'>' = pnorm(cutoff, mean.post, sd.post, lower.tail=FALSE),
	111	'in' = pnorm(cutoff[2], mean.post, sd.post) -
	112	pnorm(cutoff[1], mean.post, sd.post))
	113	label(pp) <- a$label
	114	est[[paste0('p', i)]] <- pp
	115	est[[paste0('mean', i)]] <- mean.post
	116	est[[paste0('sd', i)]] <- sd.post
	117	}
	118
	119	a <- asserts
	120	w <- NULL
	121	for(j in 1 : nas) {
	122	a <- asserts[[j]]
	123	a$dir <- a$cutoff <- NULL
	124	if(any(c('cutprior', 'tailprob') %nin% names(a)))
	125	a$cutprior <- a$tailprob <- NA
	126	w <- rbind(w, as.data.frame(a))
	127	}
	128	attr(est, 'asserts') <- w
	129	names(alabels) <- paste0('p', 1 : nas)
	130	attr(est, 'alabels') <- alabels
	131	est
	132	}
	133
	134	##' Simulate Comparisons For Use in Sequential Clinical Trial Simulations
	135	##'
	136	##' Simulates sequential clinical trials. Looks are done sequentially at observation numbers given in the vector `looks` with the earliest possible look being at observation 2. For each true effect parameter value, simulation, and at each look, runs a function to compute the estimate of the parameter of interest along with its variance. For each simulation, data are first simulated for the last look, and these data are sequentially revealed for earlier looks. The user provides a function `gendat` that given a true effect of `parameter` and the two sample sizes (for treatment groups 1 and 2) returns a list with vectors `y1` and `y2` containing simulated data. The user also provides a function `fitter` with arguments `x` (group indicator 0/1) and `y` (response variable) that returns a 2-vector containing the effect estimate and its variance. `parameter` is usually on the scale of a regression coefficient, e.g., a log odds ratio.
	137	##' @title estSeqSim
	138	##' @param parameter vector of true parameter (effects; group differences) values
	139	##' @param looks integer vector of observation numbers at which posterior probabilities are computed
	140	##' @param gendat a function of three arguments: true parameter value (scalar), sample size for first group, sample size for second group
	141	##' @param fitter a function of two arguments: 0/1 group indicator vector and the dependent variable vector
	142	##' @param nsim number of simulations (default is 1)
	143	##' @param progress set to `TRUE` to send current iteration number to the console
	144	##' @return a data frame with number of rows equal to the product of `nsim`, the length of `looks`, and the length of `parameter`.
	145	##' @author Frank Harrell
	146	##' @seealso `gbayesSeqSim()`
	147	##' @examples
	148	##' # Run 100 simulations, 5 looks, 2 true parameter values
	149	##' # Total simulation time: 2s
	150	##' lfit <- function(x, y) {
	151	##' f <- rms::lrm.fit(x, y)
	152	##' k <- length(coef(f))
	153	##' c(coef(f)[k], vcov(f)[k, k])
	154	##' }
	155	##' gdat <- function(beta, n1, n2) {
	156	##' # Cell probabilities for a 7-category ordinal outcome for the control group
	157	##' p <- c(2, 1, 2, 7, 8, 38, 42) / 100
	158	##'
	159	##' # Compute cell probabilities for the treated group
	160	##' p2 <- pomodm(p=p, odds.ratio=exp(beta))
	161	##' y1 <- sample(1 : 7, n1, p, replace=TRUE)
	162	##' y2 <- sample(1 : 7, n2, p2, replace=TRUE)
	163	##' list(y1=y1, y2=y2)
	164	##' }
	165	##'
	166	##' set.seed(1)
	167	##' est <- estSeqSim(c(0, log(0.7)), looks=c(50, 75, 95, 100, 200),
	168	##' gendat=gdat,
	169	##' fitter=lfit, nsim=100)
	170	##' head(est)
	171	##' @md
	172	estSeqSim <- function(parameter, looks, gendat, fitter, nsim=1,
	173	progress=FALSE) {
	174
	175	looks <- sort(looks)
	176	nlook <- length(looks)
	177	N <- max(looks)
	178	np <- length(parameter)
	179	nc <- nsim * nlook * np
	180	parm <- est <- vest <- numeric(nc)
	181	look <- sim <- integer(nc)
	182
	183	## For each simulation and each parameter value, simulate data for the
	184	## whole study
	185
	186	is <- 0
	187	for(isim in 1 : nsim) {
	188	if(progress) cat('Simulation', isim, '\r')
	189	for(param in parameter) {
	190	X <- sample(0 : 1, N, replace=TRUE)
	191	dat <- gendat(param, sum(X == 0), sum(X == 1))
	192	Y <- rep(NA, N)
	193	Y[X == 0] <- dat$y1
	194	Y[X == 1] <- dat$y2
	195
	196	## For each look compute the parameter estimate and its variance
	197
	198	for(l in looks) {
	199	f <- fitter(X[1 : l], Y[1 : l])
	200	is <- is + 1
	201	sim[is] <- isim
	202	parm[is] <- param
	203	look[is] <- l
	204	est[is] <- f[1]
	205	vest[is] <- f[2]
	206	} # end looks
	207	} # end param
	208	} # end sim
	209
	210	if(progress) cat('\n')
	211
	212	data.frame(sim=sim, parameter=parm, look=look,
	213	est=est, vest=vest)
	214	}

-2

R/ggfreqScatter.r less more

67	67	xlab(xlab) + ylab(ylab) +
68	68	guides(size = guide_legend(title='Frequency'))
69	69	else
70		ggplot(k, aes(x=x, y=y, alpha=Freq ^ 0.25, label=Freq,
	70	ggplot(k, aes(x=x, y=y, label=Freq, #alpha=Freq ^ 0.25,
71	71	color=Freq ^ 0.25)) +
72	72	geom_point(...) +
73	73	scale_color_gradientn(colors=fcolors) +

89	89	xlab(xlab) + ylab(ylab) +
90	90	guides(size = guide_legend(title='Frequency'))
91	91	else
92		ggplot(k, aes(x=x, y=y, alpha=fg, label=Freq,
	92	ggplot(k, aes(x=x, y=y, label=Freq, # alpha=fg,
93	93	color=if(few) k$Freq else as.integer(fg))) +
94	94	geom_point(...) +
95	95	scale_color_gradientn(colors=fcolors,

+128

-0

R/pairUpDiff.r less more

	0	##' Pair-up and Compute Differences
	1	##'
	2	##' This function sets up for plotting half-width confidence intervals for differences, sorting by descending order of differences within major categories, especially for dot charts as produced by [dotchartpl()]. Given a numeric vector `x` and a grouping (superpositioning) vector `group` with exactly two levels, computes differences in possibly transformed `x` between levels of `group` for the two observations that are equal on `major` and `minor`. If `lower` and `upper` are specified, using `conf.int` and approximate normality on the transformed scale to backsolve for the standard errors of estimates, and uses approximate normality to get confidence intervals on differences by taking the square root of the sum of squares of the two standard errors. Coordinates for plotting half-width confidence intervals are also computed. These intervals may be plotted on the same scale as `x`, having the property that they overlap the two `x` values if and only if there is no "significant" difference at the `conf.int` level.
	3	##' @title pairUpDiff
	4	##' @param x a numeric vector
	5	##' @param major an optional factor or character vector
	6	##' @param minor an optional factor or character vector
	7	##' @param group a required factor or character vector with two levels
	8	##' @param refgroup a character string specifying which level of `group` is to be subtracted
	9	##' @param lower an optional numeric vector giving the lower `conf.int` confidence limit for `x`
	10	##' @param upper similar to `lower` but for the upper limit
	11	##' @param minkeep the minimum value of `x` required to keep the observation. An observation is kept if either `group` has `x` exceeding or equalling `minkeep`. Default is to keep all observations.
	12	##' @param sortdiff set to `FALSE` to avoid sorting observations by descending between-`group` differences
	13	##' @param conf.int confidence level; must have been the value used to compute `lower` and `upper` if they are provided
	14	##' @return a list of two objects both sorted by descending values of differences in `x`. The `X` object is a data frame that contains the original variables sorted by descending differences across `group` and in addition a variable `subscripts` denoting the subscripts of original observations with possible re-sorting and dropping depending on `sortdiff` and `minkeep`. The `D` data frame contains sorted differences (`diff`), `major`, `minor`, `sd` of difference, `lower` and `upper` confidence limits for the difference, `mid`, the midpoint of the two `x` values involved in the difference, `lowermid`, the midpoint minus 1/2 the width of the confidence interval, and `uppermid`, the midpoint plus 1/2 the width of the confidence interval. Another element returned is `dropped` which is a vector of `major` / `minor` combinations dropped due to `minkeep`.
	15	##' @author Frank Harrell
	16	##' @export
	17	##' @md
	18	##' @examples
	19	##' x <- c(1, 4, 7, 2, 5, 3, 6)
	20	##' pairUpDiff(x, c(rep('A', 4), rep('B', 3)),
	21	##' c('u','u','v','v','z','z','q'),
	22	##' c('a','b','a','b','a','b','a'), 'a', x-.1, x+.1)
	23
	24	pairUpDiff <- function(x, major=NULL, minor=NULL, group, refgroup,
	25	lower=NULL, upper=NULL, minkeep=NULL,
	26	sortdiff=TRUE, conf.int=0.95) {
	27
	28	n <- length(x)
	29	major <- if(! length(major)) rep(' ', n) else as.character(major)
	30	minor <- if(! length(minor)) rep(' ', n) else as.character(minor)
	31	## Note: R will not let you use z[cbind(...)] if one of the dimensions
	32	## is of length 1 and equal to ''; needed to use ' '
	33	group <- as.character(group)
	34	glev <- unique(group)
	35	if(length(glev) != 2) stop('group must have two distinct values')
	36	if(refgroup %nin% glev) stop('refgroup must be one of the group values')
	37	altgroup <- setdiff(glev, refgroup)
	38
	39	mm <- c(major, minor)
	40	sep <- if(! any(grepl(':', mm))) ':'
	41	else
	42	if(! any(grepl('\|', mm))) '\|'
	43	else
	44	if(! any(grepl(';', mm))) ';'
	45	else
	46	if(! any(grepl('!', mm))) '!'
	47	else
	48	stop('major or minor contain all delimiters :\|;!')
	49
	50	m <- paste0(major, sep, minor)
	51	u <- unique(m)
	52	lu <- length(u)
	53
	54	lowup <- length(lower) * length(upper) > 0
	55	zcrit <- qnorm((1 + conf.int) / 2)
	56
	57	## See if any observations should be dropped
	58	dropped <- NULL
	59	if(length(minkeep)) {
	60	xa <- xb <- structure(rep(NA, lu), names=u)
	61	j <- group == refgroup
	62	xa[m[j]] <- x[j]
	63	j <- group == altgroup
	64	xb[m[j]] <- x[j]
	65	j <- ! is.na(xa + xb) & (xa < minkeep & xb < minkeep)
	66	if(any(j)) {
	67	dropped <- names(xa)[j]
	68	u <- setdiff(u, dropped)
	69	lu <- length(u)
	70	}
	71	}
	72
	73	xa <- xb <- sda <- sdb <- diff <- mid <- dsd <- dlower <-
	74	dupper <- dlowermid <- duppermid <- structure(rep(NA, lu), names=u)
	75	j <- (group == refgroup) & (m %nin% dropped)
	76	w <- m[j]
	77	xa[w] <- x[j]
	78	if(lowup) sda[w] <- 0.5 * (upper[j] - lower[j]) / zcrit
	79
	80	j <- (group == altgroup) & (m %nin% dropped)
	81	w <- m[j]
	82	xb[w] <- x[j]
	83	if(lowup) sdb[w] <- 0.5 * (upper[j] - lower[j]) / zcrit
	84
	85	diff[u] <- xb[u] - xa[u]
	86	if(lowup) {
	87	dsd[u] <- sqrt(sda[u] ^ 2 + sdb[u] ^ 2)
	88	dlower[u] <- diff[u] - zcrit * dsd[u]
	89	dupper[u] <- diff[u] + zcrit * dsd[u]
	90	}
	91	mid[u] <- (xa[u] + xb[u]) / 2.
	92	if(lowup) {
	93	dlowermid[u] <- mid[u] - 0.5 * zcrit * dsd[u]
	94	duppermid[u] <- mid[u] + 0.5 * zcrit * dsd[u]
	95	}
	96
	97	k <- strsplit(u, sep)
	98	ma <- sapply(k, function(x) x[[1]])
	99	mi <- sapply(k, function(x) x[[2]])
	100
	101	ww <- list(x, major, minor, group, lower, upper, subscripts=1:length(x))
	102
	103	Z <- if(lowup)
	104	data.frame(x, major, minor, group, lower, upper,
	105	subscripts=1 : length(x))
	106	else
	107	data.frame(x, major, minor, group, subscripts=1 : length(x))
	108
	109	if(length(dropped)) Z <- Z[m %nin% dropped, ]
	110	if(sortdiff) {
	111	m <- paste0(Z$major, sep, Z$minor)
	112	## diff[m] is a table lookup; difference will be same for both groups
	113	j <- order(Z$major, ifelse(is.na(diff[m]), -Inf, - diff[m]))
	114	Z <- Z[j, ]
	115	}
	116
	117	## Variables referenced below have already had observations dropped
	118	## due to minkeep
	119	D <- data.frame(diff=diff[u], major=ma, minor=mi, sd=dsd[u],
	120	lower=dlower[u], upper=dupper[u],
	121	mid=mid[u], lowermid=dlowermid[u], uppermid=duppermid[u])
	122	if(sortdiff) {
	123	j <- order(D$major, ifelse(is.na(D$diff), -Inf, -D$diff))
	124	D <- D[j, ]
	125	}
	126	list(X=Z, D=D, dropped=dropped)
	127	}

+50

-10

R/popower.s less more

61	61	if(length(x) && (length(x) != length(p)))
62	62	stop('p and x must have same length')
63	63	if(length(x) && any(diff(x) <= 0)) stop('x is not sorted or has duplicates')
	64
64	65	if(abs(sum(p) - 1) > .00001)
65	66	stop('probabilities do not sum to 1')
66	67

152	153	x <- d[[v[2]]]
153	154	xl <- label(x, default=v[2])
154	155	s <- sn <- NULL
	156	sl <- NULL
155	157	if(length(v) > 2) {
156	158	if(length(odds.ratio))
157		stop('odds ratio may not be specified when a stratification variable is include')
	159	stop('odds ratio may not be specified when a stratification variable is included')
158	160	s <- d[[v[3]]]
159	161	sl <- label(s, default=v[3])
160	162	s <- as.factor(s)
161	163	sn <- 's'
162	164	}
163	165	names(d) <- c('y', 'x', sn)
	166	## ggplot2 bar chart puts first category at the top
	167	## Let's put it at the bottom
	168	d$y <- factor(d$y, levels=rev(levels(as.factor(d$y))))
	169
	170
164	171	# For each x compute the vector of proportions of y categories
165	172	# Assume levels are in order
	173	# Put numerators and denominators into a character string so that
	174	# data.table [ operator can operate on one variable
	175	# The delimiter is a single space
166	176	g <- function(y) {
167	177	tab <- table(y)
168		tab / sum(tab)
169		}
	178	structure(paste(tab, rep(sum(tab), length(tab))), names=names(tab))
	179	}
	180	atxt <- function(d, strat=NULL, or=FALSE) {
	181	if(! or) {
	182	z <- d$prop
	183	num <- as.numeric(sub(' .*', '', z)) # get first number
	184	denom <- as.numeric(sub('.* ', '', z)) # get second number
	185	d$prop <- num / denom
	186	}
	187	d$txt <- paste0(sl, if(length(strat)) ': ', as.character(strat),
	188	if(length(strat)) '<br>',
	189	xl, ': ', as.character(d$x), '<br>',
	190	yl, ': ', as.character(d$y), '<br>',
	191	'Proportion: ', round(d$prop, 3),
	192	if(! or) '\u2003',
	193	if(! or) markupSpecs$html$frac(num, denom, size=90))
	194	d
	195	}
	196
170	197	d <- data.table(d)
171		if(! length(s)) {
	198
	199	if(! length(s)) { # stratification variable not present
172	200	p <- d[, as.list(g(y)), by=x]
173	201	pm <- melt(p, id=1, variable.name='y', value.name='prop')
	202	pm <- atxt(pm)
	203
174	204	plegend <- guides(fill=guide_legend(title=yl))
175	205	if(! length(odds.ratio)) {
176		gg <- ggplot(pm, aes(x=as.factor(x), y=prop, fill=factor(y))) +
	206	gg <- ggplot(pm, aes(x=as.factor(x), y=prop, fill=factor(y), text=txt)) +
177	207	geom_col() + plegend + xlab(xl) + ylab('Proportion')
178	208	return(gg)
179	209	}
180		} else {
181		p <- d[, as.list(g(y)), by=.(x,s)]
	210	} else { # stratification variable present; odds ratio must be absent
	211	p <- d[, as.list(g(y)), by=.(x, s)]
182	212	pm <- melt(p, id=c('x', 's'), variable.name='y', value.name='prop')
	213	pm <- atxt(pm, pm$s)
183	214	plegend <- guides(fill=guide_legend(title=yl))
184		gg <- ggplot(pm, aes(x=as.factor(x), y=prop, fill=factor(y))) +
	215	gg <- ggplot(pm, aes(x=as.factor(x), y=prop, fill=factor(y), text=txt)) +
185	216	facet_wrap(~ s, ncol=ncol, nrow=nrow) +
186	217	geom_col() + plegend + xlab(xl) + ylab('Proportion')
187	218	return(gg)
188	219	}
189	220
	221	## odds.ratio is present
	222
190	223	if(! length(ref)) ref <- p$x[1]
191		propfirstx <- as.matrix(p[x == ref, -1])
	224	pfx <- as.matrix(p[x == ref, -1])
	225	propfirstx <- as.numeric(sub(' .*', '', pfx)) /
	226	as.numeric(sub('.* ', '', pfx))
	227
192	228	g <- function(x) {
193	229	w <- pomodm(p=propfirstx, odds.ratio=odds.ratio(x))
194	230	names(w) <- levels(y)

196	232	}
197	233	pa <- d[, as.list(g(x)), by=x]
198	234	pma <- melt(pa, id=1, variable.name='y', value.name='prop')
	235	pma <- atxt(pma, or=TRUE)
199	236	w <- rbind(cbind(type=1, pm),
200	237	cbind(type=2, pma))
201	238	w$type <- factor(w$type, 1 : 2,
202	239	c('Observed', 'Asssuming Proportional Odds'))
203		ggplot(w, aes(x=as.factor(x), y=prop, fill=factor(y))) + geom_col() +
	240	ggplot(w, aes(x=as.factor(x), y=prop, fill=factor(y), text=txt)) +
	241	geom_col() +
204	242	facet_wrap(~ type, nrow=2) +
205	243	plegend + xlab(xl) + ylab('Proportion')
206	244	}

307	345	guides(fill=guide_legend(override.aes=list(shape=NA), order=1)) +
308	346	coord_flip() + labs(y=xlab, x='')
309	347	}
	348
	349	utils::globalVariables('txt')

-1

R/rcorr.s less more

24	24	nam <- dimnames(x)[[2]]
25	25	dimnames(h) <- list(nam, nam)
26	26	dimnames(npair) <- list(nam, nam)
27		P <- matrix(2 * (1 - pt(abs(h) * sqrt(npair - 2) / sqrt(1 - h * h),
	27	P <- matrix(2 * (1 - pt(abs(h) * sqrt(npair - 2) / sqrt(pmax(1 - h * h, 0.)),
28	28	npair - 2)), ncol=p)
29	29	P[abs(h) == 1] <- 0
30	30	diag(P) <- NA

+28

-14

R/scat1d.s less more

368	368
369	369
370	370	histSpike <-
371		function(x, side=1, nint=100, frac=.05, minf=NULL, mult.width=1,
	371	function(x, side=1, nint=100, bins=NULL, frac=.05, minf=NULL, mult.width=1,
372	372	type=c('proportion','count','density'),
373	373	xlim=range(x),
374		ylim=c(0,max(f)), xlab=deparse(substitute(x)),
	374	ylim=c(0, max(f)), xlab=deparse(substitute(x)),
375	375	ylab=switch(type, proportion='Proportion',
376	376	count ='Frequency',
377	377	density ='Density'),
378		y=NULL, curve=NULL, add=FALSE,
	378	y=NULL, curve=NULL, add=FALSE, minimal=FALSE,
379	379	bottom.align=type == 'density',
380	380	col=par('col'), lwd=par('lwd'), grid=FALSE, ...)
381	381	{
382	382	type <- match.arg(type)
383		if(! add && side != 1)
384		stop('side must be 1 if add=FALSE')
385
386		if(add && type == 'count')
387		warning('type="count" is ignored if add=TRUE')
	383
	384	if(! minimal) {
	385	if(! add && side != 1)
	386	stop('side must be 1 if add=FALSE')
	387
	388	if(add && type == 'count')
	389	warning('type="count" is ignored if add=TRUE')
	390	}
388	391
389	392	if(length(y) > 1) {
390	393	if(length(y) != length(x))

393	396	if(length(curve))
394	397	warning('curve ignored when y specified')
395	398
396		i <- !is.na(x + y)
	399	i <- ! is.na(x + y)
397	400	curve <- list(x=x[i], y=y[i])
398	401	}
399	402

409	412	f <- table(x)
410	413	x <- as.numeric(names(f))
411	414	} else {
412		ncut <- nint+1
413		bins <- seq(xlim[1], xlim[2], length = ncut)
414		delta <- (bins[2]-bins[1]) / 2
	415	ncut <- nint + 1
	416	if(! length(bins)) bins <- seq(xlim[1], xlim[2], length = ncut)
	417	delta <- (bins[2] - bins[1]) / 2
415	418	f <- table(cut(x, c(bins[1] - delta, bins)))
416	419
417	420	x <- bins

429	432	f <- den$y
430	433	}
431	434
432		if(!add) {
	435	if(! minimal && ! add) {
433	436	if(grid)
434	437	stop('add=FALSE not implemented for lattice')
435	438
436	439	plot(0, 0, xlim=xlim, ylim=ylim, xlab=xlab, ylab=ylab, type='n')
437	440	}
438
	441
439	442	if(side == 1 \|\| side == 3) {
440	443	l <- 1:2;
441	444	ax <- 1

447	450	f <- f / max(f)
448	451	if(length(minf)) f <- pmax(f, minf)
449	452
	453	if(minimal) {
	454	plot.new() # sets (0,0) to (1,1)
	455	del <- diff(xlim)
	456	usr <- c(xlim[1] - 0.04 * del, xlim[2] + 0.04 * del, 0, 1)
	457	par(mar=c(3.5, 0.5, 0.5, 0.5), mgp=c(2, 0.5, 0), usr=usr)
	458	axis(1, line=0.25)
	459	title(xlab=xlab)
	460	segments(x, rep(0, length(x)), x, f, lwd=lwd, col=col)
	461	return(invisible(xlim))
	462	}
	463
450	464	pr <- parGrid(grid)
451	465	usr <- pr$usr;
452	466	pin <- pr$pin;

+11

-5

R/summaryS.s less more

47	47	}
48	48	gg <- function(x) if(is.character(x) \|\| is.factor(x))
49	49	'categorical' else 'numeric'
50		xlabels <- sapply(X, label)
	50	## For some reason sapply is doubling names e.g. sbp.sbp
	51	sapply2 <- function(data, ...) {
	52	s <- sapply(data, ...)
	53	names(s) <- names(data)
	54	s
	55	}
	56	xlabels <- sapply2(X, label)
51	57	xlabels <- ifelse(xlabels == '', names(xlabels), xlabels)
52		ylabels <- sapply(Y, label)
	58	ylabels <- sapply2(Y, label)
53	59	ylabels <- ifelse(ylabels == '', names(ylabels), ylabels)
54	60
55	61	for(n in names(w)) # takes care of R change stringsAsFactors=FALSE

57	63
58	64	structure(w, class=c('summaryS', 'data.frame'),
59	65	formula=formula, fun=fun,
60		xnames=names(X), xlabels=xlabels, xunits=sapply(X, units),
	66	xnames=names(X), xlabels=xlabels, xunits=sapply2(X, units),
61	67	xtype=sapply(X, gg),
62		ynames=namY, ylabels=ylabels, yunits=sapply(Y, units),
	68	ynames=namY, ylabels=ylabels, yunits=sapply2(Y, units),
63	69	ylim=ylim, funlabel=funlabel)
64	70	}
65	71

411	417
412	418	yvarlev <- NULL
413	419	for(v in levels(X$yvar)) {
414		un <- yunits[v]
	420	un <- yunits[v]
415	421	l <- if(ylabels[v] == v && un == '') v else
416	422	labelPlotmath(ylabels[v], un, html=TRUE)
417	423	yvarlev <- c(yvarlev, l)

+42

-2

man/dotchartpl.Rd less more

30	30	These have the property of intersecting the two proportions if and
31	31	only if there is no significant difference at the \code{1 - conf.int}
32	32	level.
	33
	34	Specify \code{fun=exp} and \code{ifun=log} if estimates and confidence
	35	limits are on the log scale. Make sure that zeros were prevented in
	36	the original calculations. For exponential hazard rates this can be
	37	accomplished by replacing event counts of 0 with 0.5.
33	38	}
34	39	\usage{
35	40	dotchartpl(x, major=NULL, minor=NULL, group=NULL, mult=NULL,
36	41	big=NULL, htext=NULL, num=NULL, denom=NULL,
	42	numlabel='', denomlabel='',
	43	fun=function(x) x, ifun=function(x) x, op='-',
37	44	lower=NULL, upper=NULL,
38	45	refgroup=NULL, sortdiff=TRUE, conf.int=0.95,
39	46	minkeep=NULL, xlim=NULL, xlab='Proportion',
40	47	tracename=NULL, limitstracename='Limits',
41	48	nonbigtracename='Stratified Estimates',
42		width=800, height=NULL,
	49	dec=3, width=800, height=NULL,
43	50	col=colorspace::rainbow_hcl)
44	51	}
45	52	\arguments{

60	67	\code{num} is given, \code{x} is automatically added to hover text,
61	68	rounded to 3 digits after the decimal point.}
62	69	\item{denom}{like \code{num} but for denominators}
	70	\item{numlabel}{character string to put to the right of the numerator
	71	in hover text}
	72	\item{denomlabel}{character string to put to the right of the
	73	denominator in hover text}
	74	\item{fun}{a transformation to make when printing estimates. For
	75	example, one may specify \code{fun=exp} to anti-log estimates and
	76	confidence limites that were computed on a log basis}
	77	\item{ifun}{inverse transformation of \code{fun}}
	78	\item{op}{set to for example \code{'/'} when \code{fun=exp} and
	79	effects are computed as ratios instead of differences. This is used
	80	in hover text.}
63	81	\item{lower}{lower limits for optional error bars}
64	82	\item{upper}{upper limits for optional error bars}
65	83	\item{refgroup}{if \code{group} is specified and there are exactly two

87	105	\item{col}{a function or vector of colors to assign to \code{group}.
88	106	If a function it will be evaluated with an argument equal to the
89	107	number of distinct groups.}
	108	\item{dec}{number of places to the right of the decimal place for
	109	formatting numeric quantities in hover text}
90	110	\item{width}{width of plot in pixels}
91	111	\item{height}{height of plot in pixels; computed from number of strata
92	112	by default}

121	141	# state/region/group
122	142	d <- subset(d, city == 0)
123	143	with(d,
124		dotchartpl(x, major, minor, group, num=num, denom=denom, refgroup='Male')
	144	dotchartpl(x, major, minor, group, num=num, denom=denom,
	145	lower=lower, upper=upper, refgroup='Male')
125	146	)
126	147
127	148	n <- 500

154	175	big=region == 'All Regions', num=freq, denom=denom)
155	176	)
156	177	# Note these plots can be created by plot.summaryP when options(grType='plotly')
	178
	179	# Plot hazard rates and ratios with confidence limits, on log scale
	180	d <- data.frame(tx=c('a', 'a', 'b', 'b'),
	181	event=c('MI', 'stroke', 'MI', 'stroke'),
	182	count=c(10, 5, 5, 2),
	183	exposure=c(1000, 1000, 900, 900))
	184	# There were no zero event counts in this dataset. In general we
	185	# want to handle that, hence the 0.5 below
	186	d <- upData(d, hazard = pmax(0.5, count) / exposure,
	187	selog = sqrt(1. / pmax(0.5, count)),
	188	lower = log(hazard) - 1.96 * selog,
	189	upper = log(hazard) + 1.96 * selog)
	190	with(d,
	191	dotchartpl(log(hazard), minor=event, group=tx, num=count, denom=exposure,
	192	lower=lower, upper=upper,
	193	fun=exp, ifun=log, op='/',
	194	numlabel='events', denomlabel='years',
	195	refgroup='a', xlab='Events Per Person-Year')
	196	)
157	197	}
158	198	}
159	199	\keyword{hplot}

+61

-0

man/estSeqSim.Rd less more

	0	% Generated by roxygen2: do not edit by hand
	1	% Please edit documentation in R/gbayesSeqSim.r
	2	\name{estSeqSim}
	3	\alias{estSeqSim}
	4	\title{estSeqSim}
	5	\usage{
	6	estSeqSim(parameter, looks, gendat, fitter, nsim = 1, progress = FALSE)
	7	}
	8	\arguments{
	9	\item{parameter}{vector of true parameter (effects; group differences) values}
	10
	11	\item{looks}{integer vector of observation numbers at which posterior probabilities are computed}
	12
	13	\item{gendat}{a function of three arguments: true parameter value (scalar), sample size for first group, sample size for second group}
	14
	15	\item{fitter}{a function of two arguments: 0/1 group indicator vector and the dependent variable vector}
	16
	17	\item{nsim}{number of simulations (default is 1)}
	18
	19	\item{progress}{set to \code{TRUE} to send current iteration number to the console}
	20	}
	21	\value{
	22	a data frame with number of rows equal to the product of \code{nsim}, the length of \code{looks}, and the length of \code{parameter}.
	23	}
	24	\description{
	25	Simulate Comparisons For Use in Sequential Clinical Trial Simulations
	26	}
	27	\details{
	28	Simulates sequential clinical trials. Looks are done sequentially at observation numbers given in the vector \code{looks} with the earliest possible look being at observation 2. For each true effect parameter value, simulation, and at each look, runs a function to compute the estimate of the parameter of interest along with its variance. For each simulation, data are first simulated for the last look, and these data are sequentially revealed for earlier looks. The user provides a function \code{gendat} that given a true effect of \code{parameter} and the two sample sizes (for treatment groups 1 and 2) returns a list with vectors \code{y1} and \code{y2} containing simulated data. The user also provides a function \code{fitter} with arguments \code{x} (group indicator 0/1) and \code{y} (response variable) that returns a 2-vector containing the effect estimate and its variance. \code{parameter} is usually on the scale of a regression coefficient, e.g., a log odds ratio.
	29	}
	30	\examples{
	31	# Run 100 simulations, 5 looks, 2 true parameter values
	32	# Total simulation time: 2s
	33	lfit <- function(x, y) {
	34	f <- rms::lrm.fit(x, y)
	35	k <- length(coef(f))
	36	c(coef(f)[k], vcov(f)[k, k])
	37	}
	38	gdat <- function(beta, n1, n2) {
	39	# Cell probabilities for a 7-category ordinal outcome for the control group
	40	p <- c(2, 1, 2, 7, 8, 38, 42) / 100
	41
	42	# Compute cell probabilities for the treated group
	43	p2 <- pomodm(p=p, odds.ratio=exp(beta))
	44	y1 <- sample(1 : 7, n1, p, replace=TRUE)
	45	y2 <- sample(1 : 7, n2, p2, replace=TRUE)
	46	list(y1=y1, y2=y2)
	47	}
	48
	49	set.seed(1)
	50	est <- estSeqSim(c(0, log(0.7)), looks=c(50, 75, 95, 100, 200),
	51	gendat=gdat,
	52	fitter=lfit, nsim=100)
	53	head(est)
	54	}
	55	\seealso{
	56	\code{gbayesSeqSim()}
	57	}
	58	\author{
	59	Frank Harrell
	60	}

-0

man/gbayes.Rd less more

257	257	\cr
258	258	\email{fh@fharrell.com}
259	259	}
	260	\seealso{\code{\link{gbayesSeqSim}}}
260	261	\references{
261	262	Spiegelhalter DJ, Freedman LS, Parmar MKB (1994): Bayesian approaches to
262	263	randomized trials. JRSS A 157:357--416. Results for \code{gbayes} are derived from

+82

-0

man/gbayesSeqSim.Rd less more

	0	% Generated by roxygen2: do not edit by hand
	1	% Please edit documentation in R/gbayesSeqSim.r
	2	\name{gbayesSeqSim}
	3	\alias{gbayesSeqSim}
	4	\title{gbayesSeqSim}
	5	\usage{
	6	gbayesSeqSim(est, asserts)
	7	}
	8	\arguments{
	9	\item{est}{data frame created by \code{estSeqSim()}}
	10
	11	\item{asserts}{list of lists. The first element of each list is the user-specified name for each assertion/prior combination, e.g., \code{"efficacy"}. The other elements are, in order, a character string equal to "<", ">", or "in", a parameter value \code{cutoff} (for "<" and ">") or a 2-vector specifying an interval for "in", and either a prior distribution mean and standard deviation named \code{mu} and \code{sigma} respectively, or a parameter value (\code{"cutprior"}) and tail area \code{"tailprob"}. If the latter is used, \code{mu} is assumed to be zero and \code{sigma} is solved for such that P(parameter > 'cutprior') = P(parameter < - 'cutprior') = \code{tailprob}.}
	12	}
	13	\value{
	14	a data frame with number of rows equal to that of \code{est} with a number of new columns equal to the number of assertions added. The new columns are named \code{p1}, \code{p2}, \code{p3}, ... (posterior probabilities), \code{mean1}, \code{mean2}, ... (posterior means), and \code{sd1}, \code{sd2}, ... (posterior standard deviations). The returned data frame also has an attribute \code{asserts} added which is the original \code{asserts} augmented with any derived \code{mu} and \code{sigma} and converted to a data frame, and another attribute \code{alabels} which is a named vector used to map \code{p1}, \code{p2}, ... to the user-provided labels in \code{asserts}.
	15	}
	16	\description{
	17	Simulate Bayesian Sequential Treatment Comparisons Using a Gaussian Model
	18	}
	19	\details{
	20	Simulate a sequential trial under a Gaussian model for parameter estimates, and Gaussian priors using simulated estimates and variances returned by \code{estSeqSim}. For each row of the data frame \code{est} and for each prior/assertion combination, computes the posterior probability of the assertion.
	21	}
	22	\examples{
	23	\dontrun{
	24	# Simulate Bayesian operating characteristics for an unadjusted
	25	# proportional odds comparison (Wilcoxon test)
	26	# For 100 simulations, 5 looks, 2 true parameter values, and
	27	# 2 assertion/prior combinations, compute the posterior probability
	28	# Use a low-level logistic regression call to speed up simuluations
	29	# Use data.table to compute various summary measures
	30	# Total simulation time: 2s
	31	lfit <- function(x, y) {
	32	f <- rms::lrm.fit(x, y)
	33	k <- length(coef(f))
	34	c(coef(f)[k], vcov(f)[k, k])
	35	}
	36	gdat <- function(beta, n1, n2) {
	37	# Cell probabilities for a 7-category ordinal outcome for the control group
	38	p <- c(2, 1, 2, 7, 8, 38, 42) / 100
	39
	40	# Compute cell probabilities for the treated group
	41	p2 <- pomodm(p=p, odds.ratio=exp(beta))
	42	y1 <- sample(1 : 7, n1, p, replace=TRUE)
	43	y2 <- sample(1 : 7, n2, p2, replace=TRUE)
	44	list(y1=y1, y2=y2)
	45	}
	46
	47	# Assertion 1: log(OR) < 0 under prior with prior mean 0.1 and sigma 1 on log OR scale
	48	# Assertion 2: OR between 0.9 and 1/0.9 with prior mean 0 and sigma computed so that
	49	# P(OR > 2) = 0.05
	50	asserts <- list(list('Efficacy', '<', 0, mu=0.1, sigma=1),
	51	list('Similarity', 'in', log(c(0.9, 1/0.9)),
	52	cutprior=log(2), tailprob=0.05))
	53
	54	set.seed(1)
	55	est <- estSeqSim(c(0, log(0.7)), looks=c(50, 75, 95, 100, 200),
	56	gendat=gdat,
	57	fitter=lfit, nsim=100)
	58	z <- gbayesSeqSim(est, asserts)
	59	head(z)
	60	attr(z, 'asserts')
	61
	62	# Compute the proportion of simulations that hit targets (different target posterior
	63	# probabilities for efficacy vs. similarity)
	64
	65	# For the efficacy assessment compute the first look at which the target
	66	# was hit (set to infinity if never hit)
	67	require(data.table)
	68	z <- data.table(z)
	69	u <- z[, .(first=min(p1 > 0.95])), by=.(parameter, sim)]
	70	# Compute the proportion of simulations that ever hit the target and
	71	# that hit it by the 100th subject
	72	u[, .(ever=mean(first < Inf)), by=.(parameter)]
	73	u[, .(by75=mean(first <= 100)), by=.(parameter)]
	74	}
	75	}
	76	\seealso{
	77	\code{gbayes()}, \code{estSeqSim()}
	78	}
	79	\author{
	80	Frank Harrell
	81	}

-2

man/ggfreqScatter.Rd less more

42	42	2-vector, the first element corresponds to \code{x} and the second to
43	43	\code{y}.}
44	44	\item{g}{number of quantile groups to make for frequency counts. Use
45		\code{g=0} to use frequencies continuously for color and alpha
	45	\code{g=0} to use frequencies continuously for color
46	46	coding. This is recommended only when using \code{plotly}.}
47	47	\item{cuts}{instead of using \code{g}, specify \code{cuts} to provide
48	48	the vector of cuts for categorizing frequencies for assignment to colors}

55	55	\item{xlab,ylab}{axis labels. If not specified and variable has a
56	56	\code{label}, thatu label will be used.}
57	57	\item{fcolors}{\code{colors} argument to pass to
58		\code{scale_color_gradientn} to color code frequencies}
	58	\code{scale_color_gradientn} to color code frequencies. Use
	59	\code{fcolors=gray.colors(10, 0.75, 0)} to show gray
	60	scale, for example. Another good choice is
	61	\code{fcolors=hcl.colors(10, 'Blue-Red')}.}
59	62	\item{nsize}{set to \code{TRUE} to not vary color or transparency but
60	63	instead to size the symbols in relation to the number of points. Best
61	64	with both \code{x} and \code{y} are discrete. \code{ggplot2}

+58

-0

man/pairUpDiff.Rd less more

	0	% Generated by roxygen2: do not edit by hand
	1	% Please edit documentation in R/pairUpDiff.r
	2	\name{pairUpDiff}
	3	\alias{pairUpDiff}
	4	\title{pairUpDiff}
	5	\usage{
	6	pairUpDiff(
	7	x,
	8	major = NULL,
	9	minor = NULL,
	10	group,
	11	refgroup,
	12	lower = NULL,
	13	upper = NULL,
	14	minkeep = NULL,
	15	sortdiff = TRUE,
	16	conf.int = 0.95
	17	)
	18	}
	19	\arguments{
	20	\item{x}{a numeric vector}
	21
	22	\item{major}{an optional factor or character vector}
	23
	24	\item{minor}{an optional factor or character vector}
	25
	26	\item{group}{a required factor or character vector with two levels}
	27
	28	\item{refgroup}{a character string specifying which level of \code{group} is to be subtracted}
	29
	30	\item{lower}{an optional numeric vector giving the lower \code{conf.int} confidence limit for \code{x}}
	31
	32	\item{upper}{similar to \code{lower} but for the upper limit}
	33
	34	\item{minkeep}{the minimum value of \code{x} required to keep the observation. An observation is kept if either \code{group} has \code{x} exceeding or equalling \code{minkeep}. Default is to keep all observations.}
	35
	36	\item{sortdiff}{set to \code{FALSE} to avoid sorting observations by descending between-\code{group} differences}
	37
	38	\item{conf.int}{confidence level; must have been the value used to compute \code{lower} and \code{upper} if they are provided}
	39	}
	40	\value{
	41	a list of two objects both sorted by descending values of differences in \code{x}. The \code{X} object is a data frame that contains the original variables sorted by descending differences across \code{group} and in addition a variable \code{subscripts} denoting the subscripts of original observations with possible re-sorting and dropping depending on \code{sortdiff} and \code{minkeep}. The \code{D} data frame contains sorted differences (\code{diff}), \code{major}, \code{minor}, \code{sd} of difference, \code{lower} and \code{upper} confidence limits for the difference, \code{mid}, the midpoint of the two \code{x} values involved in the difference, \code{lowermid}, the midpoint minus 1/2 the width of the confidence interval, and \code{uppermid}, the midpoint plus 1/2 the width of the confidence interval. Another element returned is \code{dropped} which is a vector of \code{major} / \code{minor} combinations dropped due to \code{minkeep}.
	42	}
	43	\description{
	44	Pair-up and Compute Differences
	45	}
	46	\details{
	47	This function sets up for plotting half-width confidence intervals for differences, sorting by descending order of differences within major categories, especially for dot charts as produced by \code{\link[=dotchartpl]{dotchartpl()}}. Given a numeric vector \code{x} and a grouping (superpositioning) vector \code{group} with exactly two levels, computes differences in possibly transformed \code{x} between levels of \code{group} for the two observations that are equal on \code{major} and \code{minor}. If \code{lower} and \code{upper} are specified, using \code{conf.int} and approximate normality on the transformed scale to backsolve for the standard errors of estimates, and uses approximate normality to get confidence intervals on differences by taking the square root of the sum of squares of the two standard errors. Coordinates for plotting half-width confidence intervals are also computed. These intervals may be plotted on the same scale as \code{x}, having the property that they overlap the two \code{x} values if and only if there is no "significant" difference at the \code{conf.int} level.
	48	}
	49	\examples{
	50	x <- c(1, 4, 7, 2, 5, 3, 6)
	51	pairUpDiff(x, c(rep('A', 4), rep('B', 3)),
	52	c('u','u','v','v','z','z','q'),
	53	c('a','b','a','b','a','b','a'), 'a', x-.1, x+.1)
	54	}
	55	\author{
	56	Frank Harrell
	57	}

+10

-4

man/popower.Rd less more

26	26	proportions stratified by a grouping variable (and optionally a stratification variable), with an optional
27	27	additional graph showing what the proportions would be had proportional
28	28	odds held and an odds ratio was applied to the proportions in a
29		reference group.
	29	reference group. If the result is passed to \code{ggplotly}, customized
	30	tooltip hover text will appear.
30	31
31	32	\code{propsTrans} uses \code{\link{ggplot2}} to plot all successive
32	33	transition proportions. \code{formula} has the state variable on the

55	56	\arguments{
56	57	\item{p}{
57	58	a vector of marginal cell probabilities which must add up to one.
58		The \code{i}th element specifies the probability that a patient will be
59		in response level \code{i}, averaged over the two treatment groups.
	59	For \code{popower} and \code{posamsize}, The \code{i}th element specifies the probability that a patient will be
	60	in response level \code{i}, averaged over the two treatment groups. For
	61	\code{pomodm}, \code{p} is the vector of cell probabilities to be
	62	translated under a given odds ratio.
60	63	}
61	64	\item{odds.ratio}{
62	65	the odds ratio to be able to detect. It doesn't

66	69	apply to the refernce group to get all other group's expected proportions
67	70	were proportional odds to hold against the first group. Normally the
68	71	function should return 1.0 when its \code{x} argument corresponds to the
69		\code{ref} group.}
	72	\code{ref} group. For \code{pomodm} is the odds ratio to apply to
	73	convert the given cell probabilities.}
70	74	\item{n}{
71	75	total sample size for \code{popower}. You must specify either \code{n} or
72	76	\code{n1} and \code{n2}. If you specify \code{n}, \code{n1} and

198	202	d <- expand.grid(time=1:3, reps=1:30)
199	203	d$y <- sample(letters[1:5], nrow(d), replace=TRUE)
200	204	propsPO(y ~ time, data=d, odds.ratio=function(time) c(1, 2, 4)[time])
	205	# To show with plotly, save previous result as object p and then:
	206	# plotly::ggplotly(p, tooltip='text')
201	207
202	208	# Add a stratification variable and don't consider an odds ratio
203	209	d <- expand.grid(time=1:5, sex=c('female', 'male'), reps=1:30)

+14

-4

man/scat1d.Rd less more

118	118	# sc(a,b) means default to a if number of axes <= 3, b if >=50, use
119	119	# linear interpolation within 3-50
120	120
121		histSpike(x, side=1, nint=100, frac=.05, minf=NULL, mult.width=1,
	121	histSpike(x, side=1, nint=100, bins=NULL, frac=.05, minf=NULL, mult.width=1,
122	122	type=c('proportion','count','density'),
123	123	xlim=range(x), ylim=c(0,max(f)), xlab=deparse(substitute(x)),
124	124	ylab=switch(type,proportion='Proportion',
125	125	count ='Frequency',
126	126	density ='Density'),
127		y=NULL, curve=NULL, add=FALSE,
	127	y=NULL, curve=NULL, add=FALSE, minimal=FALSE,
128	128	bottom.align=type=='density', col=par('col'), lwd=par('lwd'),
129	129	grid=FALSE, \dots)
130	130

204	204	the curve. For \code{histSpike}, interpolated \code{y} values are
205	205	derived for binmidpoints.
206	206	}
	207	\item{minimal}{for \code{histSpike} set \code{minimal=TRUE} to draw a
	208	minimalist spike histogram with no y-axis. This works best when
	209	produce graphics images that are short, e.g., have a height of
	210	two inches. \code{add} is forced to be \code{FALSE} in this case
	211	so that a standalone graph is produced. Only base graphics are
	212	used.}
207	213	\item{bottom.align}{
208	214	set to \code{TRUE} to have the bottoms of tick marks (for
209	215	\code{side=1} or \code{side=3}) aligned at the y-coordinate. The

259	265	plotted. For \code{histSpikeg}, if \code{x} has no more than
260	266	\code{nint} unique values, all observed values are used, otherwise
261	267	the data are rounded before tabulation so that there are no more
262		than \code{nint} intervals.
263		}
	268	than \code{nint} intervals. For \code{histSpike}, \code{nint} is
	269	ignored if \code{bins} is given.
	270	}
	271	\item{bins}{for \code{histSpike} specifies the actual cutpoints to use
	272	for binning \code{x}. The default is to use \code{nint} in
	273	conjunction with \code{xlim}.}
264	274	\item{\dots}{
265	275	optional arguments passed to \code{scat1d} from \code{datadensity}
266	276	or to \code{histSpike} from \code{scat1d}. For \code{histSpikep}