Chapter26: Statistics
615
[FunctionFile]
gaminv
(
x
,
a
,
b
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
Gammadistributionwithshapeparameteraandscaleb.
[FunctionFile]
geopdf
(
x
,
p
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
geometricdistributionwithparameterp.
Thegeometric distributionmodelsthenumberoffailures (x-1)ofaBernoullitrial
withprobabilitypbeforethefirstsuccess(x).
[FunctionFile]
geocdf
(
x
,
p
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thegeometricdistributionwithparameterp.
Thegeometric distributionmodelsthenumberoffailures (x-1)ofaBernoullitrial
withprobabilitypbeforethefirstsuccess(x).
[FunctionFile]
geoinv
(
x
,
p
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
geometricdistributionwithparameterp.
Thegeometric distributionmodelsthenumberoffailures (x-1)ofaBernoullitrial
withprobabilitypbeforethefirstsuccess(x).
[FunctionFile]
hygepdf
(
x
,
t
,
m
,
n
)
Computetheprobabilitydensity function(PDF) at x ofthehypergeometricdistri-
butionwithparameterst,m,andn.
Thisistheprobabilityofobtainingxmarkeditemswhenrandomlydrawingasample
ofsizenwithoutreplacementfromapopulationoftotalsizetcontainingmmarked
items.
Theparameterst,m,andnmustbepositiveintegerswithmandnnotgreaterthan
t.
[FunctionFile]
hygecdf
(
x
,
t
,
m
,
n
)
Compute the cumulative distribution n function (CDF) at x x of f the e hypergeometric
distributionwithparameterst,m,andn.
Thisis theprobability ofobtainingnotmorethanx x markeditems s whenrandomly
drawing asample of size n without replacement from a population of totalsize t
containingmmarkeditems.
Theparameterst,m,andnmustbepositiveintegerswithmandnnotgreaterthan
t.
[FunctionFile]
hygeinv
(
x
,
t
,
m
,
n
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
hypergeometricdistributionwithparameterst,m,andn.
Thisistheprobabilityofobtainingxmarkeditemswhenrandomlydrawingasample
ofsizenwithoutreplacementfromapopulationoftotalsizetcontainingmmarked
items.
Theparameterst,m,andnmustbepositiveintegerswithmandnnotgreaterthan
t.
Pdf split file - Split, seperate PDF into multiple files in C#.net, ASP.NET, MVC, Ajax, WinForms, WPF
Explain How to Split PDF Document in Visual C#.NET Application
pdf link to specific page; how to split pdf file by pages
Pdf split file - VB.NET PDF File Split Library: Split, seperate PDF into multiple files in vb.net, ASP.NET, MVC, Ajax, WinForms, WPF
VB.NET PDF Document Splitter Control to Disassemble PDF Document
pdf split file; break pdf into single pages
616
GNUOctave
[FunctionFile]
kolmogorov_smirnov_cdf
(
x
,
tol
)
Returnthecumulativedistributionfunction(CDF)atxoftheKolmogorov-Smirnov
distribution.
Thisisdefinedas
Q(x)=
X1
k= 1
( 1)
k
exp( 2k
2
x
2
)
forx>0.
Theoptionalparametertolspecifies theprecisionuptowhichtheseries shouldbe
evaluated;thedefaultistol=eps.
[FunctionFile]
laplace_pdf
(
x
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
Laplacedistribution.
[FunctionFile]
laplace_cdf
(
x
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
theLaplacedistribution.
[FunctionFile]
laplace_inv
(
x
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
Laplacedistribution.
[FunctionFile]
logistic_pdf
(
x
)
Foreachelementofx,computethePDFatx ofthelogisticdistribution.
[FunctionFile]
logistic_cdf
(
x
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thelogisticdistribution.
[FunctionFile]
logistic_inv
(
x
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
logisticdistribution.
[FunctionFile]
lognpdf
(
x
)
[FunctionFile]
lognpdf
(
x
,
mu
,
sigma
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
lognormaldistributionwithparametersmuandsigma.
Ifarandomvariablefollows thisdistribution,its logarithmis normally distributed
withmeanmuandstandarddeviationsigma.
Defaultvaluesaremu=0,sigma=1.
[FunctionFile]
logncdf
(
x
)
[FunctionFile]
logncdf
(
x
,
mu
,
sigma
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thelognormaldistributionwithparametersmuandsigma.
Ifarandomvariablefollows thisdistribution,its logarithmis normally distributed
withmeanmuandstandarddeviationsigma.
Defaultvaluesaremu=0,sigma=1.
Online Split PDF file. Best free online split PDF tool.
Split PDF file. Just upload your file by clicking on the blue button or drag-and-drop your PDF file into the drop area. Then set your PDF file split settings.
break pdf file into parts; break apart pdf pages
VB.NET PDF File Compress Library: Compress reduce PDF size in vb.
Also able to uncompress PDF file in VB.NET programs. Offer flexible and royalty-free developing library license for VB.NET programmers to compress PDF file.
pdf specification; break pdf file into multiple files
Chapter26: Statistics
617
[FunctionFile]
logninv
(
x
)
[FunctionFile]
logninv
(
x
,
mu
,
sigma
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
lognormaldistributionwithparametersmuandsigma.
Ifarandomvariablefollows thisdistribution,its logarithmis normally distributed
withmeanmuandstandarddeviationsigma.
Defaultvaluesaremu=0,sigma=1.
[FunctionFile]
nbinpdf
(
x
,
n
,
p
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
negativebinomialdistributionwithparametersnandp.
WhennisintegerthisisthePascaldistribution.Whennisextendedtorealnumbers
thisisthePolyadistribution.
Thenumberoffailures inaBernoulliexperiment withsuccessprobability pbefore
then-thsuccessfollowsthisdistribution.
[FunctionFile]
nbincdf
(
x
,
n
,
p
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thenegativebinomialdistributionwithparametersnandp.
WhennisintegerthisisthePascaldistribution.Whennisextendedtorealnumbers
thisisthePolyadistribution.
Thenumberoffailures inaBernoulliexperiment withsuccessprobability pbefore
then-thsuccessfollowsthisdistribution.
[FunctionFile]
nbininv
(
x
,
n
,
p
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
negativebinomialdistributionwithparametersnandp.
WhennisintegerthisisthePascaldistribution.Whennisextendedtorealnumbers
thisisthePolyadistribution.
Thenumberoffailures inaBernoulliexperiment withsuccessprobability pbefore
then-thsuccessfollowsthisdistribution.
[FunctionFile]
normpdf
(
x
)
[FunctionFile]
normpdf
(
x
,
mu
,
sigma
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
normaldistributionwithmeanmuandstandarddeviationsigma.
Defaultvaluesaremu=0,sigma=1.
[FunctionFile]
normcdf
(
x
)
[FunctionFile]
normcdf
(
x
,
mu
,
sigma
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thenormaldistributionwithmeanmuandstandarddeviationsigma.
Defaultvaluesaremu=0,sigma=1.
[FunctionFile]
norminv
(
x
)
[FunctionFile]
norminv
(
x
,
mu
,
sigma
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
normaldistributionwithmeanmuandstandarddeviationsigma.
C# PDF File & Page Process Library SDK for C#.net, ASP.NET, MVC
Well-designed APIs are provided. Splitting PDF File. If you want to split PDF file into two or small files, you may refer to this online guide.
acrobat split pdf bookmark; break apart a pdf file
VB.NET PDF File Merge Library: Merge, append PDF files in vb.net
Professional VB.NET PDF file merging SDK support Visual Studio .NET. Merge PDF without size limitation. Append one PDF file to the end of another one in VB.NET.
add page break to pdf; break pdf into multiple files
618
GNUOctave
Defaultvaluesaremu=0,sigma=1.
[FunctionFile]
poisspdf
(
x
,
lambda
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
Poissondistributionwithparameterlambda.
[FunctionFile]
poisscdf
(
x
,
lambda
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thePoissondistributionwithparameterlambda.
[FunctionFile]
poissinv
(
x
,
lambda
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
Poissondistributionwithparameterlambda.
[FunctionFile]
stdnormal_pdf
(
x
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
standardnormaldistribution(mean=0,standarddeviation=1).
[FunctionFile]
stdnormal_cdf
(
x
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thestandardnormaldistribution(mean=0,standarddeviation=1).
[FunctionFile]
stdnormal_inv
(
x
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
standardnormaldistribution(mean=0,standarddeviation=1).
[FunctionFile]
tpdf
(
x
,
n
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)atxofthet
(Student)distributionwithndegreesoffreedom.
[FunctionFile]
tcdf
(
x
,
n
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
thet(Student)distributionwithndegreesoffreedom.
[FunctionFile]
tinv
(
x
,
n
)
Foreachelementofx,computethequantile(theinverseoftheCDF)atx ofthet
(Student)distributionwithndegreesoffreedom.
This function is analogous to o looking in a a table e for r the e t-value of a a single-tailed
distribution.
[FunctionFile]
unidpdf
(
x
,
n
)
For each h element t of x, , compute the probability density y function n (PDF) ) at x x of f a
discreteuniformdistributionwhichassumestheintegervalues1–nwithequalprob-
ability.
Warning: The e underlying g implementation uses the e double class and will l only be
accuratefornbitmax(2
53
1onIEEE754compatiblesystems).
[FunctionFile]
unidcdf
(
x
,
n
)
For each h element of x, compute the cumulative distribution function (CDF) ) at t x
ofadiscrete uniformdistributionwhichassumes the integer values1–nwithequal
probability.
C# PDF File Compress Library: Compress reduce PDF size in C#.net
Reduce image resources: Since images are usually or large size, images size reducing can help to reduce PDF file size effectively.
break a pdf file into parts; break a pdf file
C# PDF File Merge Library: Merge, append PDF files in C#.net, ASP.
Professional C#.NET PDF SDK for merging PDF file merging in Visual Studio .NET. Append one PDF file to the end of another and save to a single PDF file.
break apart pdf; break pdf password online
Chapter26: Statistics
619
[FunctionFile]
unidinv
(
x
,
n
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
discreteuniformdistributionwhichassumestheintegervalues1–nwithequalprob-
ability.
[FunctionFile]
unifpdf
(
x
)
[FunctionFile]
unifpdf
(
x
,
a
,
b
)
Foreachelementofx,computetheprobabilitydensityfunction(PDF)at x ofthe
uniformdistributionontheinterval[a,b].
Defaultvaluesarea=0,b=1.
[FunctionFile]
unifcdf
(
x
)
[FunctionFile]
unifcdf
(
x
,
a
,
b
)
Foreachelement ofx,computethecumulativedistributionfunction(CDF)atx of
theuniformdistributionontheinterval[a,b].
Defaultvaluesarea=0,b=1.
[FunctionFile]
unifinv
(
x
)
[FunctionFile]
unifinv
(
x
,
a
,
b
)
For eachelement of x,computethe quantile (the inverseofthe CDF) at x x of f the
uniformdistributionontheinterval[a,b].
Defaultvaluesarea=0,b=1.
[FunctionFile]
wblpdf
(
x
)
[FunctionFile]
wblpdf
(
x
,
scale
)
[FunctionFile]
wblpdf
(
x
,
scale
,
shape
)
Computetheprobabilitydensityfunction(PDF)atxoftheWeibulldistributionwith
scaleparameterscaleandshapeparametershape.
Thisisgivenby
shape
scale
shape
x
shape 1
e
(
x
scale
)
shape
forx0.
Defaultvaluesarescale =1,shape =1.
[FunctionFile]
wblcdf
(
x
)
[FunctionFile]
wblcdf
(
x
,
scale
)
[FunctionFile]
wblcdf
(
x
,
scale
,
shape
)
Computethecumulativedistributionfunction(CDF)atxoftheWeibulldistribution
withscaleparameterscale andshapeparametershape.
Thisisdefinedas
1 e
x
scale
)
shape
forx0.
[FunctionFile]
wblinv
(
x
)
[FunctionFile]
wblinv
(
x
,
scale
)
[FunctionFile]
wblinv
(
x
,
scale
,
shape
)
Computethequantile(theinverseoftheCDF)atxoftheWeibulldistributionwith
scaleparameterscaleandshapeparametershape.
Defaultvaluesarescale =1,shape =1.
C# Word - Split Word Document in C#.NET
C# DLLs: Split Word File. Add references: RasterEdge.Imaging.Basic.dll. using RasterEdge.XDoc.Word; Split Word file into two files in C#.
break pdf into multiple documents; can't select text in pdf file
C# PowerPoint - Split PowerPoint Document in C#.NET
File: Split PowerPoint Document. |. Home ›› XDoc.PowerPoint ›› C# PowerPoint: Split PowerPoint Document. Split PowerPoint file into two files in C#.
pdf splitter; cannot print pdf file no pages selected
620
GNUOctave
26.6 Tests
Octavecanperformmany differentstatisticaltests. . Thefollowingtable e summarizes the
availabletests.
Hypothesis
TestFunctions
Equalmeanvalues
anova,hotelling
test2,t
test
2,
welch
test,wilcoxon
test,z
test
2
Equalmedians
kruskal
wallis
test,sign
test
Equalvariances
bartlett
test,manova,var
test
Equaldistributions
chisquare
test
homogeneity,
kolmogorov
smirnov
test
2,u
test
Equalmarginalfrequencies
mcnemar
test
Equalsuccessprobabilities
prop
test
2
Independentobservations
chisquare
test
independence,
run
test
Uncorrelatedobservations
cor
test
Givenmeanvalue
hotelling
test,t
test,z
test
Observationsfromdistribution
kolmogorov
smirnov
test
Regression
f
test
regression,t
test
regression
The tests returnap-valuethat describestheoutcomeofthetest. . Assumingthatthe
testhypothesisistrue,thep-valueis theprobabilityofobtainingaworseresultthanthe
observedone. Solargep-valuescorrespondstoasuccessfultest.Usuallyatesthypothesis
isacceptedifthep-valueexceeds0.05.
[FunctionFile]
[pval, f, df_b, df_w] = anova
(
y
,
g
)
Performaone-wayanalysisofvariance(ANOVA).
Thegoalistotestwhetherthepopulationmeansofdatatakenfromkdifferentgroups
areallequal.
Data may y be e given in a single vector r y with groups specifiedby y a corresponding
vectorofgrouplabelsg(e.g.,numbersfrom1tok). Thisisthegeneralformwhich
doesnotimposeanyrestrictiononthenumber ofdata ineachgrouporthegroup
labels.
Ify isamatrixandgisomitted,eachcolumnofy istreatedasagroup. . Thisform
isonlyappropriateforbalancedANOVAinwhichthenumbersofsamplesfromeach
groupareallequal.
Underthenullofconstantmeans,thestatisticf followsanFdistributionwithdf
b
anddf
wdegreesoffreedom.
Thep-value(1minustheCDFofthisdistributionatf)isreturnedinpval.
Ifnooutputargumentisgiven,thestandardone-wayANOVAtableisprinted.
Seealso: [manova],page623.
[FunctionFile]
[pval, chisq, df] = = bartlett_test
(
x1
,...)
PerformaBartletttestforthehomogeneityofvariancesinthedatavectors x1,x2,
...,xk,wherek >1.
Chapter26: Statistics
621
Under the nullof equal l variances, the test t statistic chisq q approximately y follows s a
chi-squaredistributionwithdf degreesoffreedom.
Thep-value(1minustheCDFofthisdistributionatchisq)isreturnedinpval.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
[pval, chisq, df] = = chisquare_test_homogeneity
(
x
,
y
,
c
)
Giventwo samples x x andy, performa a chisquare test for r homogeneity y of the null
hypothesis that t x x and y come e from the same distribution, based onthe partition
inducedbythe(strictlyincreasing)entriesofc.
Forlargesamples,theteststatisticchisqapproximatelyfollowsachisquaredistribu-
tionwithdf =length(c)degreesoffreedom.
Thep-value(1minustheCDFofthisdistributionatchisq)isreturnedinpval.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
[pval, chisq, df] = = chisquare_test_independence
(
x
)
Performachi-squaretestforindependencebasedonthecontingencytablex.
Under the null hypothesis of f independence, , chisq approximately y has s a a chi-square
distributionwithdf degreesoffreedom.
Thep-value(1minustheCDFofthisdistributionatchisq)ofthetestisreturnedin
pval.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
cor_test
(
x
,
y
,
alt
,
method
)
Testwhethertwosamplesx andy comefromuncorrelatedpopulations.
The optional argument stringalt describes s the e alternative hypothesis, and can be
"!="or"<>"(nonzero),">"(greaterthan0),or"<"(lessthan0).Thedefaultisthe
two-sidedcase.
The optional argument string method specifies which correlationcoefficient touse
fortesting. Ifmethodis"pearson"(default),the(usual)Pearson’sprodutmoment
correlationcoefficientis used. . Inthis s case,thedata shouldcomefromabivariate
normaldistribution. Otherwise,theothertwomethodsoffernonparametricalterna-
tives.Ifmethodis"kendall",thenKendall’srankcorrelationtauisused.Ifmethod
is"spearman",thenSpearman’srankcorrelationrhoisused.Onlythefirstcharacter
isnecessary.
Theoutputisastructurewiththefollowingelements:
pval
Thep-valueofthetest.
stat
Thevalueoftheteststatistic.
dist
Thedistributionoftheteststatistic.
params
Theparametersofthenulldistributionoftheteststatistic.
alternative
Thealternativehypothesis.
method
Themethodusedfortesting.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
622
GNUOctave
[FunctionFile]
[pval, f, df_num, df_den] = = f_test_regression
(
y
,
x
,
rr
,
r
)
PerformanFtestforthenullhypothesisrr*b=rinaclassicalnormalregression
modely=X*b+e.
Underthenull,theteststatisticf followsanFdistributionwithdf
numanddf
den
degreesoffreedom.
Thep-value(1minustheCDFofthisdistributionatf)isreturnedinpval.
Ifnotgivenexplicitly,r=0.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
[pval, tsq] ] = = hotelling_test
(
x
,
m
)
For a sample x x from a a multivariate normaldistributionwithunknown mean and
covariancematrix,testthenullhypothesisthatmean(x)==m.
Hotelling’s T
2
is returned d intsq. . Under r the e null, (n p)T
2
=(p(n 1)) ) has s anF
distributionwithpandn pdegreesoffreedom,wherenandparethenumbersof
samplesandvariables,respectively.
Thep-valueofthetestisreturnedinpval.
Ifnooutputargumentisgiven,thep-valueofthetestisdisplayed.
[FunctionFile]
[pval, tsq] ] = = hotelling_test_2
(
x
,
y
)
Fortwosamplesx frommultivariatenormaldistributionswiththesamenumberof
variables (columns), , unknownmeans s andunknown n equalcovariance matrices, test
thenullhypothesismean(x)==mean(y).
Hotelling’stwo-sampleT
2
isreturnedintsq.Underthenull,
(n
x
+n
y
p 1)T
2
p(n
x
+n
y
2)
hasanFdistributionwithpandn
x
+n
y
p 1degreesoffreedom,wheren
x
and
n
y
arethesamplesizesandpisthenumberofvariables.
Thep-valueofthetestisreturnedinpval.
Ifnooutputargumentisgiven,thep-valueofthetestisdisplayed.
[FunctionFile]
[pval, ks] = kolmogorov_smirnov_test
(
x
,
dist
,
params
,
alt
)
PerformaKolmogorov-Smirnovtestofthenullhypothesisthatthesamplex comes
fromthe(continuous)distributiondist.
ifFandGaretheCDFscorrespondingtothesampleanddist,respectively,thenthe
nullisthatF==G.
Theoptionalargumentparamscontainsalistofparametersofdist.Forexample,to
testwhetherasamplex comesfromauniformdistributionon[2,4],use
kolmogorov_smirnov_test (x, , "unif", 2, 4)
distcanbeanystringforwhichafunctiondistcdf thatcalculatestheCDFofdistri-
butiondistexists.
Chapter26: Statistics
623
Withtheoptionalargumentstringalt,thealternativeofinterestcanbeselected. If
altis "!="or "<>",thenullis testedagainstthetwo-sidedalternativeF!=G. . In
thiscase,theteststatisticks followsatwo-sidedKolmogorov-Smirnovdistribution.
Ifalt is">", , theone-sidedalternativeF > > Gis considered. . Similarly y for "<", , the
one-sided alternative F > > Gis considered. . Inthis s case,the test statistic ks has a
one-sidedKolmogorov-Smirnovdistribution.Thedefaultisthetwo-sidedcase.
Thep-valueofthetestisreturnedinpval.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
[pval, ks, d] = kolmogorov_smirnov_test_2
(
x
,
y
,
alt
)
Performa2-sampleKolmogorov-Smirnovtestofthenullhypothesisthatthesamples
x andy comefromthesame(continuous)distribution.
IfFandGaretheCDFscorrespondingtothex andy y samples,respectively,then
thenullisthatF==G.
Withtheoptionalargumentstringalt,thealternativeofinterestcanbeselected. If
altis "!="or "<>",thenullis testedagainstthetwo-sidedalternativeF!=G. . In
thiscase,theteststatisticks followsatwo-sidedKolmogorov-Smirnovdistribution.
Ifalt is">", , theone-sidedalternativeF > > Gis considered. . Similarly y for "<", , the
one-sided alternative F < < Gis considered. . Inthis s case,the test statistic ks has a
one-sidedKolmogorov-Smirnovdistribution.Thedefaultisthetwo-sidedcase.
Thep-valueofthetestisreturnedinpval.
The third d returned d value, d, is s the test statistic, , the maximum m vertical distance
betweenthetwocumulativedistributionfunctions.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
[pval, k, df] = kruskal_wallis_test
(
x1
,...)
PerformaKruskal-Wallisone-factoranalysisofvariance.
Supposeavariableisobservedfork >1differentgroups,andletx1,...,xk bethe
correspondingdatavectors.
Underthenullhypothesisthattheranksinthepooledsamplearenotaffectedbythe
groupmemberships,theteststatistick isapproximatelychi-squarewithdf f =k k -1
degreesoffreedom.
Ifthedatacontainsties(somevalueappearsmorethanonce)kisdividedby
1-sum
ties/(n^3-n)
wheresum
ties isthesumoft^2-tovereachgroupoftieswheret isthenumber
oftiesinthegroupandnisthetotalnumberofvaluesintheinputdata. Formore
infoonthis adjustmentseeWilliamH.KruskalandW.AllenWallis,UseofRanks
inOne-CriterionVarianceAnalysis,JournaloftheAmericanStatisticalAssociation,
Vol. 47,No. 260(Dec1952).
Thep-value(1minustheCDFofthisdistributionatk)isreturnedinpval.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
manova
(
x
,
g
)
Performaone-waymultivariateanalysisofvariance(MANOVA).
624
GNUOctave
Thegoalistotestwhetherthep-dimensionalpopulationmeansofdatatakenfrom
k different t groups s are allequal. . All l data are assumed drawnindependently y from
p-dimensionalnormaldistributionswiththesamecovariancematrix.
The data matrix is given by y x. . As s usual, , rows are observations s and d columns s are
variables. Thevectorg specifiesthecorrespondinggrouplabels(e.g.,numbersfrom
1tok).
TheLRteststatistic(Wilks’Lambda)andapproximatep-valuesarecomputedand
displayed.
Seealso: [anova],page620.
[FunctionFile]
[pval, chisq, df] = = mcnemar_test
(
x
)
For a a square contingency table e x x of f data a cross-classified on n the row w and column
variables,McNemar’stestcanbeusedfortestingthenullhypothesisofsymmetryof
theclassificationprobabilities.
Under the null, , chisq is s approximately distributedaschisquare withdf f degrees s of
freedom.
Thep-value(1minustheCDFofthisdistributionatchisq)isreturnedinpval.
Ifnooutputargumentisgiven,thep-valueofthetestisdisplayed.
[FunctionFile]
[pval, z] = = prop_test_2
(
x1
,
n1
,
x2
,
n2
,
alt
)
Ifx1andn1arethecountsofsuccessesandtrialsinonesample,andx2andn2those
inasecondone,testthenullhypothesisthatthesuccessprobabilitiesp1andp2 are
thesame.
Underthenull,theteststatisticzapproximatelyfollowsastandardnormaldistribu-
tion.
Withtheoptionalargumentstringalt,thealternativeofinterestcanbeselected. If
altis"!="or"<>",thenullistestedagainstthetwo-sidedalternativep1 !=p2. If
altis">",theone-sidedalternativep1>p2isused.Similarlyfor"<",theone-sided
alternativep1 <p2isused. Thedefaultisthetwo-sidedcase.
Thep-valueofthetestisreturnedinpval.
Ifnooutputargumentisgiven,thep-valueofthetestisdisplayed.
[FunctionFile]
[pval, chisq] = run_test
(
x
)
Performachi-squaretestwith6degreesoffreedombasedontheupwardrunsinthe
columnsofx.
run_testcanbeusedtodecidewhetherx containsindependentdata.
Thep-valueofthetestisreturnedinpval.
Ifnooutputargumentisgiven,thep-valueisdisplayed.
[FunctionFile]
[pval, b, n] = sign_test
(
x
,
y
,
alt
)
For two matched-pair samples x x andy,perform asign testofthe e nullhypothesis
PROB(x>y)==PROB(x<y)==1/2.
Underthenull,theteststatisticbroughlyfollows abinomialdistributionwithpa-
rametersn=sum(x!=y)andp=1/2.
Documents you may be interested
Documents you may be interested