5/30/2016

24thFebruary2014

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

AssignmentStatisticsbehaviourofsikkimmanipal
Universitysemester1

Question:Astatisticalsurveyisascientificprocessofcollectionandanalysisof
numericaldata.Explainthestagesofstatisticalsurvey.Describethevarious
methodsforcollectingdatainastatisticalsurvey.

Answer:Definition of Statistical Survey:A Statistical Survey is a scientific process of collection and

analysisofnumericaldata.Statisticalsurveysareusedtocollectinformationaboutunitsinapopulation
anditinvolvesaskingquestionstoindividuals.Surveysofhumanpopulationsarecommoningovernment,
health,socialscienceandmarketingsectors.
StagesofStatisticalSurvey:Statistical surveys involve two stages namely Planning and Execution.
Figure2.1showsthetwobroadstagesofStatisticalSurvey
PlanningaStatisticalSurvey:Therelevanceandaccuracyofdataobtainedinasurveydependsupon
the care taken in planning.A properly planned investigation can lead to the best results with least cost
andtime.Figure2.2givestheexplanationofstepsinvolvedintheplanningstage.
Executionofstatisticalsurvey:Controlled methods should be adopted at every stage of carrying out
theinvestigationtochecktheaccuracy,coverage,methodsofmeasurements,analysisandinterpretation.
The collected data should be edited, classified, tabulated and presented in the form of diagrams and
graphs.Thedatashouldbecarefullyandsystematicallyanalysedandinterpreted.
CollectionofData:Collectionofdataisthefirstandmostimportantstageinanystatisticalsurvey.The
method for collection of data depends upon various factors such as objective, scope, nature of
investigation and availability of resources. Direct personal interviews, third party agencies, and
questionnairesaresomewaysthroughwhichdataiscollected.
Primarydata:Primarydataistheone,whichiscollectedbytheinvestigatorforthepurposeofaspecific
inquiryorstudy.Suchdataisoriginalincharacterandisgeneratedbyasurveyconductedbyindividuals
oraresearchinstitutionoranyorganisation.
ForExample:Ifaresearcherisinterestedtoknowtheimpactofanonmealschemeforschoolchildren,
he/she has to undertake a survey and collect data on the opinion of parents and children by asking
relevantquestions.Suchadatacollectediscalledprimarydata.
Collectionofprimarydataisdonebyasuitablemethodasperthefollowing:
1.Directpersonalobservation
2.Indirectoralinterview
3.Informationthroughagencies
4.Informationthroughmailedquestionnaires
5.Informationthroughaschedulefilledbyinvestigators.
Secondary data: Any information, that is used for the current investigation but is obtained from some
data,whichhasbeencollectedandusedbysomeotheragencyorpersoninaseparateinvestigation,or
survey,isknownassecondarydata.Theyareavailableinapublishedorunpublishedform.
Inpublishedform,secondarydataisavailableinresearchpapers,newspapers,magazines,government
publication, international publication, and websites. Secondary data is collected for different purposes.
Therefore, care should be exercised while using it. The accuracy, reliability, objectives and scope of
secondarydatashouldbeexaminedthoroughlybeforeuse.Secondarydatamaybecollectedeitherby
censusorbysamplingmethods.
PublishedsourcesThevarioussourcesofpublisheddataare:
Reportsandofficialpublicationsofinternationalandnationalorganisationsaswellascentraland
stategovernments
Publicationsofseverallocalbodiessuchasmunicipalcorporationsanddistrictboards
Financialandeconomicjournals
Annualreportsofvariouscompanies
http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

1/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

Publicationsbroughtoutbyresearchagenciesandresearchscholars [http://smumbaassignment.com/1a
statisticalsurveyisascientificprocessofcollectionandanalysisofnumericaldataexplainthestagesofstatisticalsurvey
describethevariousmethodsforcollectingdatainastatistical/]

Question2.AnalysisofdailywagesofworkersintwoorganisationsAandB
yieldedthefollowingresults:Organisation
A
No.ofworkers
10
Averagedailywages(Rs)
30
Variance
25

B
20
15
100

Obtaintheaveragedailywagesandthestandarddeviationofwagesofall
workersinthetwoorganizationstakentogether.Whichorganizationis
moreequitableinregardtowages?
Answer:LetX1&X2,n1,n2,1and2respectivelydenotemeannumberofworkerandstandard
deviationoforganizationsAandBthen
X1=30,x2=15,n1=10,n2=201=25=5,2=100=10
Combinedmeanx12=n1x1+n2x2
n1+n2
=10*30+20*15
30
=600=30
30
CombinedStandarddeviation=
Whend1=x12x1=2030=10
D2=x12x2=2015=5
12=10*25+20*100+10*100+20*25
10*20
=55Anorganisationmoreequitableinregardawageswheneverthevariabilityinwagesis
lesserthereforewehavetocomparethecoefficientofvariationhere.
CoefficientofvariationfororganisationA
=5/30*100=16.7%
CoefficientofvariationfororganisationB

=10/15*100=66.7%
SincecoefficientofvariationfororganizationAislessitmoreequitableinRegardstowages.
Question3.a.Statetheadditionandmultiplicationrulesofprobabilitygivinganexampleofeach
case.
b.InaboltfactorymachinesA,B,Cmanufacture25,35and40percentofthetotaloutput.Oftheir
totaloutput5,4and2percentaredefectiverespectively.Aboltisdrawnatrandomandisfound
tobedefective.WhataretheprobabilitiesthatitwasmanufacturedbymachinesA,BandC?
Answer:TheAdditionLawofProbability
IftwoeventsAandBaremutuallyexclusivethen
P(AB)=P(A)+P(B)
http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

2/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

ThisisthesimplifiedversionoftheAdditionLaw.However,whenAandBarenotmutually
exclusive,AB_=,itcanbeshownthatamoregenerallawapplies:
P(AB)=P(A)+P(B)P(AB)
OfcourseifAB=then,sinceP()=0thisgeneralexpressionreducestothesimpler
version.
Example:Abagcontains20marbles,3arecolouredred,6arecolouredgreen,4arecolouredblue,
2arecolouredwhiteand5arecolouredyellow.Oneballisselectedatrandom.Findthe
probabilitiesofthefollowingevents.
(a)theballiseitherredorgreen
(b)theballisnotblue
(c)theballiseitherredorwhiteorblue.(Hint:considerthecomplementary
event.)
Answer
Inthelastexample(part(c))wecouldalternativelyhaveusedanobviousextensionofthelaw
ofadditionformutuallyexclusiveevents:
P(RWB)=P(R)+P(W)+P(B)=3/20+2/20/+4/20=9/20.
TheMultiplicationLaw
IfAandBareindependenteventsthen
P(AB)=P(A)P(B)
InwordsTheprobabilityofindependenteventsAandBoccurringistheproductoftheprobabilitiesofthe
eventsoccurringseparately.
Example:Aistheeventcomponentaisoperating
Bistheeventcomponentbisoperating
PrevioustestinghasindicatedthatP(A)=0.99,andP(B)=0.98.Thecircuitfunctionsonly
ifaandbarebothoperatingsimultaneously.Thecomponentsareassumedtobeindependent.
Thentheprobabilitythatthecircuitisoperatingisgivenby
P(AB)=P(A)P(B)=0.990.98=0.9702
NotethatthisprobabilityissmallertheneitherP(A)orP(B).
B.LetE1,E2,E3betheeventsthataboltselectedatrandomismanufacturedby
themachinesA,B,CrespectivelyandletEdenotetheeventofitsbeingdefective.Now
P(E1)=0.25,P(E2)=0.35,P(E3)=0.40
TheprobabilityofdrawingadefectiveboltmanufacturedbymachineAisP(E/E1)
=0.05.Similarlywehave
P(E/E2)=0.04,P(E/E3)=0.02
Hencetheprobabilitythatadefectiveboltselectedatrandomismanufacturedby
machineAisgivenby

Question4.a.WhatisaChisquaretest?Pointoutitsapplications.Underwhatconditionsisthistest
applicable?

b.Discussthetypesofmeasurementscaleswithexamples.
http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

3/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

Answer:ChiSquaretest
TheChisquaretestisoneofthemostcommonlyusednonparametrictests
instatisticalwork.TheGreekLetter 2isusedtodenotethistest. 2
describethemagnitudeofdiscrepancybetweentheobservedandthe
expectedfrequencies.Thevalueof 2iscalculatedas:

Where,O1,O2,O3.OnaretheobservedfrequenciesandE1,E2,E3En
arethecorrespondingexpectedortheoreticalfrequencies.
ApplicationsofChiSquaretest
1Testsforindependenceofattributes
Inthetestforindependence,thenullhypothesisisthattherowandcolumn
variablesareindependentofeachother.Wehavestudiedearlier,thatthe
hypothesistestingisdoneundertheassumptionthatthenullhypothesisis
true.
Thefollowingarethepropertiesofthetestforindependence:
Thedataaretheobservedfrequencies
Thedataisarrangedintheformofacontingencytable
Thedegreesoffreedom canbecalculatedas:
Numberofrows 1
Numberofcolumns 1
where, isthedegreesoffreedom
ThetestforindependencehasaChiSquaredistributionandisalwaysa
righttailtest.
Theexpectedvalueiscomputedbytakingtherowtotal,multiplyingit
withthecolumntotalanddividingbythegrandtotal.Thatisgivenby:

Theteststatisticvaluedoesnotchange,iftheorderoftherowsor
columnsisinterchanged.Alsothevaluedoesnotchangeevenifthe
rowsandcolumnsareinterchanged.
2Testofgoodnessoffit
Thetestofgoodnessoffitofastatisticalmodelmeasureshowaccurately
thetestfitsasetofobservations.Thistestmeasuresandsummarisesthe
differencesifany,betweentheobservedandexpectedvaluesofthe
consideredstatisticalmodel.Thesetestresultsarehelpfultoknowwhether
thesamplesaredrawnfromidenticaldistributionsornot.Thedegreesof
freedomaren1andtheexpectedvalueisequaltotheaverageofthe
observedvalues.
10.3.3Testforcomparingvariance
Whenwehavetouse2 asatestofpopulationvariance,then,

http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

4/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

(n1)=degreesoffreedom,nbeingthenumberofitemsinthe
sample.
Thenbycomparingthecalculatedvaluewiththetablevalueof2 for(n1)
degreesoffreedomatagivenlevelofsignificance,wemayeitheracceptor
rejectthenullhypothesis.Ifthecalculatedof2 islessthanthetablevalue,
thenullhypothesisisaccepted,butifthecalculatedvalueisequalorgreater
thanthetablevaluethehypothesisisrejected.
ConditionsforapplyingtheChiSquaretest
ThefollowingaretheconditionsforusingtheChiSquaretest:
1.ThefrequenciesusedinChiSquaretestmustbeabsoluteandnotin
relativeterms.
2.Thetotalnumberofobservationscollectedforthistestmustbelarge.
3.Eachoftheobservationswhichmakeupthesampleofthistestmustbe
independentofeachother.
4.As 2testisbasedwhollyonsampledata,noassumptionismade
concerningthepopulationdistribution.Inotherwords,itisanon
parametrictest.
5. 2testiswhollydependentondegreesoffreedom.Asthedegreesof
freedomincrease,theChiSquaredistributioncurvebecomes
symmetrical.
6.Theexpectedfrequencyofanyitemorcellmustnotbelessthan5,the
frequenciesofadjacentitemsorcellsshouldbepolledtogetherinorder
tomakeitmorethan5.
7.Thedatashouldbeexpressedinoriginalunitsforconvenienceof
comparisonandthegivendistributionshouldnotbereplacedbyrelative
frequenciesorproportions.
8.Thistestisusedonlyfordrawinginferencesthroughtestofthe
hypothesis,soitcannotbeusedforestimationofparametervalue.
AnswerB.Therearefourmeasurementscales(ortypesofdata):nominal,ordinal,intervalandratio.
Thesearesimplywaystocategorizedifferenttypesofvariables.
Nominal
Letsstartwiththeeasiestonetounderstand.Nominalscalesareusedforlabelingvariables,without
anyquantitative[http://www.mymarketresearchmethods.com/quantitativevsqualitativeresearchwhatsthedifference/] value.
Nominalscalescouldsimplybecalledlabels.Herearesomeexamples,below.Noticethatallof
thesescalesaremutuallyexclusive(nooverlap)andnoneofthemhaveanynumericalsignificance.A
goodwaytorememberallofthisisthatnominalsoundsalotlikenameandnominalscalesarekindof
likenamesorlabels.

[http://www.mymarketresearchmethods.com/wpcontent/uploads/2012/11/exampleofnominalscales.png]

Ordinal
Withordinalscales,itistheorderofthevaluesiswhatsimportantandsignificant,butthedifferences
betweeneachoneisnotreallyknown.Takealookattheexamplebelow.Ineachcase,weknowthata
http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

5/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

#4isbetterthana#3or#2,butwedontknowandcannotquantifyhowmuchbetteritis.Forexample,
isthedifferencebetweenOKandUnhappythesameasthedifferencebetweenVeryHappyand
Happy?Wecantsay.
Ordinalscalesaretypicallymeasuresofnonnumericconceptslikesatisfaction,happiness,discomfort,
etc.
Ordinaliseasytorememberbecauseissoundslikeorderandthatsthekeytorememberwithordinal
scalesitistheorderthatmatters,butthatsallyoureallygetfromthese.
Advancednote:Thebestwaytodeterminecentraltendencyonasetofordinaldataistousethemodeor
medianthemeancannotbedefinedfromanordinalset.

[http://www.mymarketresearchmethods.com/wpcontent/uploads/2012/11/ordinalscales.png]

ExampleofOrdinalScales
Interval
Intervalscalesarenumericscalesinwhichweknownotonlytheorder,butalsotheexactdifferences
betweenthevalues.TheclassicexampleofanintervalscaleisCelsiustemperaturebecausethe
differencebetweeneachvalueisthesame.Forexample,thedifferencebetween60and50degreesisa
measurable10degrees,asisthedifferencebetween80and70degrees.Timeisanothergoodexample
ofanintervalscaleinwhichtheincrementsareknown,consistent,andmeasurable.
Intervalscalesarenicebecausetherealmofstatisticalanalysisonthesedatasetsopensup.For
example,centraltendencycanbemeasuredbymode,median,ormeanstandarddeviationcanalsobe
calculated.
Ratio
Ratioscalesaretheultimatenirvanawhenitcomestomeasurementscalesbecausetheytellusabout
theorder,theytellustheexactvaluebetweenunits,ANDtheyalsohaveanabsolutezerowhichallows
forawiderangeofbothdescriptiveandinferentialstatistics[http://www.mymarketresearchmethods.com/descriptive
inferentialstatisticsdifference/] tobeapplied.Attheriskofrepeatingmyself,everythingaboveaboutinterval
dataappliestoratioscales+ratioscaleshaveacleardefinitionofzero.Goodexamplesofratiovariables
includeheightandweight.
Question:5.ExplaintheComponentsofTimeseries.
Answer:TimeSeries[https://www.blogger.com/null]:Atimeseriesisasequenceofobservationswhich
areorderedintime(orspace).Ifobservationsaremadeonsomephenomenonthroughouttime,itismost
sensibletodisplaythedataintheorderinwhichtheyarose,particularlysincesuccessiveobservations
willprobablybedependent.Timeseriesarebestdisplayedinascatterplot.TheseriesvalueXisplotted
ontheverticalaxisandtimetonthehorizontalaxis.Timeiscalledtheindependentvariable(inthiscase
however,somethingoverwhichyouhavelittlecontrol).Therearetwokindsoftimeseriesdata:
1. Continuous,wherewehaveanobservationateveryinstantoftime,e.g.liedetectors,

electrocardiograms.WedenotethisusingobservationXattimet,X(t).Discrete,wherewehavean
observationat(usuallyregularly)spacedintervals.WedenotethisasXt.
ComponentsofTimeSeries
Thebehaviourofatimeseriesoverperiodsoftimeiscalledthemovement
ofthetimeseries.Thetimeseriesisclassifiedintothefollowingfour
components:
i)Longtermtrendorseculartrend
http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

6/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

ii)Seasonalvariations
iii)Cyclicvariations
iv)Randomvariations
i)Longtermtrendorseculartrend
Thisreferstothesmoothorregularlongtermgrowthordeclineoftheseries.Thismovementcanbe
characterisedbyatrendcurve.Ifthiscurveisastraightline,thenitiscalledatrendline.Ifthevariable
increasesoveralongperiodoftime,thenitiscalledanupwardtrend.Ifthevariabledecreasesoveralong
periodoftime,thenitiscalledadownwardtrend.Ifthevariablemovesupwardordownwardalonga
straightlinethenthetrendiscalledalineartrend,otherwiseitiscalledanonlineartrend.
ii)Seasonalvariations
Variationsinatimeseriesthatareperiodicinnatureandoccurregularlyovershortperiodsoftimeduring
ayeararecalledseasonalvariations.Thesevariationsarepreciseandcanbeforecasted.Thefollowing
areexamplesofseasonalvariationsinatimeseries.
i.Thepricesofvegetablesdropdownafterrainyseasonorinwinter
monthsandtheygoupduringsummer,everyyear.
ii.Thepricesofcookingoilsreduceaftertheharvestingofoilseedsand
goupaftersometime.
iii)Cyclicvariations
Thelongtermoscillationsthatrepresentconsistentriseanddeclineinthe
valuesofthevariablearecalledcyclicvariations.Sincethesearelongterm
oscillationsinthetimeseries,theperiodofoscillationisusuallygreaterthan
oneyear.Theoscillationsareeitheratrendcurveoratrendline.Theperiod
ofonecycleisthetimedistancebetweentwosuccessivepeaksortwo
successivetroughs.
iv)Randomvariations
Randomvariationsarecalledirregularmovements.Movementsthatoccur
usuallyinbriefperiodsoftime,withoutanypatternandwhichare
unpredictableinnaturearecalledirregularmovements.Thesemovements
donothaveanyregularperiodortimeofoccurrences.Forexample,the
effectofnationalstrikes,floods,earthquakes,etc.Itisverydifficulttostudy
thebehaviourofsuchatimeseries.
Question:6a.Whatisanalysisofvariance?Whataretheassumptionsofthetechnique?
b.Threesamplesbelowhavebeenobtainedfromnormalpopulationswithequalvariances.Testthehypothesis
at5%levelthatthepopulationmeansareequal.
Answer:AnalysisofVariance(ANOVA)
AnalysisofVariance(ANOVA)isusefulinsuchsituationsascomparingthemileageachievedbyfivedifferentbrandsof
gasoline,testingwhichoffourdifferenttrainingmethodsproducethefastestlearningrecord,orcomparingthefirstyear
earningsofthegraduatesofhalfadozendifferentbusinessschools.Ineachofthesecases,wewouldcomparethe
meansofmorethan
twosamples.Hence,inmostofthefields,suchasagriculture,medical,finance,banking,insurance,education,etc.,
theconceptofANOVAisused.Instatisticalterms,thedifferencebetweentwostatisticaldataisknownasvariance.
Whentwodataarecomparedforanypracticalpurpose,theirdifferenceisstudiedthroughthetechniquesofANOVA.
Withtheanalysisofvariancetechnique,wecantestthenullhypothesisandthealternative
hypothesis.Nullhypothesis,H0:Allsamplemeansareequal.AlternateHypothesis,H1:Allsamplemeansarenot
equaloratleastoneofthesamplesmeansdiffer.
http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

7/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

AssumptionsforstudyofANOVA
TheunderlyingassumptionsforthestudyofANOVAare:
i)Eachofthesamplesisasimplerandomsample
ii)Populationfromwhichthesamplesareselectedarenormally
distributed
iii)Eachofthesamplesisindependentoftheothersamples
iv)Eachofthepopulationhasthesamevariationandidenticalmeans
v)Theeffectofvariouscomponentsareadditive
LetH0:Thereisnosignificantdifferenceinthemeansofthreesamples

T=Sumofallobservations=150

Posted24thFebruary2014byDevendrakachhi
0 Viewcomments

http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

8/9

5/30/2016

AssignmentStatisticsbehaviourofsikkimmanipalUniversitysemester1

No comments yet

Add a comment

http://devendratips61.blogspot.in/2014/02/assignmentstatisticsbehaviourof.html

9/9