ISYE 6644 - FINAL PREP
Whatbgisbgabgpossiblebggoalbgofbganbgindifference-zonebgnormalbgmeansbgselectionbgtechnique?bg-
bganswers--
Findbgthebgnormalbgpopulationbghavingbgthebglargestbgmean,bgespeciallybgifbgthebglargestbgmeanbgisbg≫
bgthebgsecond-largest.
TRUEbgorbgFALSE?
bgThebgBechhoferbgprocedurebgforbgselectingbgthebgnormalbgpopulationbgwithbgthebglargestbgmeanbgspe
cifiesbgthebgappropriatebgnumberbgofbgobservationsbgtobgtakebgfrombgeachbgcompetingbgpopulation,bga
ndbgsimplybgselectsbgthebgcompetitorbghavingbgthebglargestbgsamplebgmean.bg-bganswers--True
TRUEbgorbgFALSE?bgSometimesbgabgsingle-
stagebgprocedurebglikebgBechhofer'sbgisbginefficient.bgInbgfact,bgit'sbgpossiblebgtobgusebgcertainbgsequenti
albgproceduresbgthatbgtakebgobservationsbgone-at-a-
timebg(insteadbgofbgallbgatbgoncebginbgabgsinglebgstage)bgtobgmakebggoodbgselectionbgdecisionsbgusingbgf
ewerbgobservations.bg-bganswers--True
Forbgwhichbgscenarios(s)bgbelowbgmightbgitbgbebgappropriatebgtobgusebgabgBernoullibgselectionbgproced
ure?
a)bgFindbgthebginventorybgpolicybghavingbgthebglargestbgprofit.
b)bgFindbgthebgdrugbggivingbgthebgbestbgchancebgofbgabgcure.
c)bgFindbgthebgmaintenancebgpolicybghavingbgthebglowestbgfailurebgprobability.
d)bgFindbgthebgschedulingbgrulebgthatbgthatbghasbgthebgbestbgchancebgofbgmakingbganbgon-
timebgdelivery.bg-bganswers--Allbgthreebgofbg(b),bg(c),bgandbg(d).
SupposebgthatbgabgBernoullibgselectionbgprocedurebgtellsbgyoubgtobgtakebg100bgobservationsbgfrombgeac
hbgofbgtwobgpopulations,bgAbgandbgB.bgItbgturnsbgoutbgthatbgAbggetsbg85bgsuccessesbgandbgBbggetsbg46bg
successes.bgWhatbgdobgyoubgthink?bg-bganswers--
1)bgAbgalmostbgcertainlybghasbgabghigherbgsuccessbgprobabilitybgthanbgB.
,2)bgWebgcould'vebgprobablybgstoppedbgsamplingbgabgbitbgearlierbg(i.e.,bgwithbgfewerbgthanbg100bgobserv
ations)bgbecausebgAbgwasbgsobgfarbgaheadbgofbgB.
Forbgwhichbgscenarios(s)bgbelowbgmightbgitbgbebgappropriatebgtobgusebgabgmultinomialbgselectionbgproc
edure?bg-bganswers--Findbgthebgmost-popularbgpoliticalbgcandidate.
SupposebgthatbgwebgwantbgtobgknowbgwhichbgofbgCoke,bgPepsibgandbgDr.pepperbgisbgthebgmostbgpopular
.bgWebgwouldbglikebgtobgmakebgthebgcorrectbgselectionbgwithbgprobabilitybgofbgatbgleastbgP*=0.90bginbgt
hebgeventbgthatbgthebgrationbgofbgthebghighest-to-second-
highestbgpreferencebgprobabilitiesbghappensbgtobgbebgatbgleastbg0*=1.4.bgHowbgmanybgpeoplebgdoesbgt
hebgsingle-stagebgprocedurebgMbembgrequirebgusbgtobginterview?bg-bganswers--126
Gobgtobgthebgtablebginbgthebgnotesbgandbgpickbgoffbgthebgentrybgforbgk=3,bgP^⋆=0.90,bgandbgθ^⋆=1.4.
Whichbgofbgthebgfollowingbgparametersbgcanbgyoubggetbgconfidencebgintervalsbgfor?
Means
Variances
Quantiles
Differencesbgbetweenbgthebgmeansbgofbgtwobgsystemsbg-bganswers--Allbgofbgthebgabove
Ifbgwebghavebganbgiidbgnormalbgsamplebgofbgobservations,bgX1,bgX2,...Xn,bgwhatbgprobabilitybgdistributio
nbgisbgmost-commonlybgusedbgtobgobtainbgcofidencebgintervalsbgforbgthebgmean?bg-bganswers--t
TRUEbgorbgFALSE?
bgThebgpairedbgCIbgforbgthebgdifferencesbginbgtwobgmeansbgisbgdesignedbgtobgworkbgespeciallybgwellbgifbg
allbgofbgthebgobservationsbgfrombgthebgfirstbgpopulationbgarebgcompletelybgindependentbgofbgallbgofbgth
ebgobservationsbgfrombgthebgsecondbgpopulation.bg-bganswers--
FALSE.bg{Inbgfact,bgit'sbgeasierbgtobgdistinguishbgbetweenbgthebgtwobgmeansbgifbgXibgisbgpositivelybgcorrel
atedbgwithbgYi.bgThinkbgaboutbgmybgparallelbgparkingbgexamplebginbgthebgclassbgnotes.}
TRUEbgorbgFALSE?
bgYoubgcanbgusebgabgversionbgofbgindependentbgreplicationsbgtobgobtainbgconfidencebgintervalsbgforbgthe
bgdifferencebginbgthebgmeansbgfrombgtwobgsimulationbgmodels.bg-bganswers--TRUE
, TRUEbgorbgFALSE?
bgThebgcommonbgrandombgnumbersbgtechniquebgintentionallybginducesbgpositivebgcorrelationbgbetwee
nbgtwobgsystemsbg-bgmuchbglikebgabgpaired-𝑡confidencebginterval.bg-bganswers--TRUE
CRNbgdependsbgonbgsomeone'sbgabilitybgtobgmanipulatebgthebgunderlyingbgpseudo-
randombgnumbersbg-
bge.g.,bgusebgthebgsamebgarrivalbgtimesbgwhenbgsimulatingbgtwobgcompetingbgsimulatedbgsystems.bgSobg
whobgultimatelybgcontrolsbgthosebgPRNs??bg-bganswers--Youbgdobg-bgyoubgarebgpowerful!
TRUEbgorbgFALSE?
bgThebgantitheticbgrandombgnumbersbgtechniquebgintentionallybginducesbgnegativebgcorrelationbgbetwe
enbgtwobgrunsbgofbgthebgsamebgsystembg-
bgthisbgallowsbgusbgtobgbetterbgestimatebgthebgmeanbgofbgthebgsystem.bg-bganswers--TRUE
TRUEbgorbgFALSE?bgThebgcontrolbgvariatesbgtechniquebgprovidesbgunbiased,bglow-
variancebgestimatesbgusingbgabgmethodbgreminiscentbgofbgregression.bg-bganswers--TRUE
Statisticalbgrankingbgandbgselectionbgtechniquesbghavebgbeenbgdesignedbgtobgaddressbgabgvarietybgofbgc
omparisonbgproblems.bgWhichbgonesbgfrombgthebgfollowingbglist?
Findbgthebgpopulationbghavingbgthebglargestbgmean.
Findbgthebgsystembgwithbgthebgsmallestbgvariance.
Findbgthebgalternativebgwithbgthebghighestbgsuccessbgprobability.
Findbgthebgmost-popularbgcandidate.
Allbgofbgthebgabove.bg-bganswers--Allbgofbgthebgabove.
Supposebgwebgarebgdealingbgwithbgi.i.d.bgnormalbgobservationsbgwithbgunknownbgvariance.bgWhichbgofb
gthebgfollowingbgisbgtruebgaboutbgabg95%bgconfidencebgintervalbgforbgthebgmeanbgμ?bg-bganswers--
Webgarebg95%bgsurebgthatbgourbgCIbgwillbgactuallybgcontainbgthebgunknownbgvaluebgofbgμ.
Webgarebgstudyingbgthebgwaitingbgtimesbgarisingbgfrombgtwobgqueueingbgsystems.bgSupposebgwebgmake
bg4bgindependentbgreplicationsbgofbgbothbgsystems,bgwherebgthebgsystemsbgarebgsimulatedbgindepende
ntlybgofbgeachbgother.bg
replicationbgsystembg1bgsystem2
1bg10bg25
Whatbgisbgabgpossiblebggoalbgofbganbgindifference-zonebgnormalbgmeansbgselectionbgtechnique?bg-
bganswers--
Findbgthebgnormalbgpopulationbghavingbgthebglargestbgmean,bgespeciallybgifbgthebglargestbgmeanbgisbg≫
bgthebgsecond-largest.
TRUEbgorbgFALSE?
bgThebgBechhoferbgprocedurebgforbgselectingbgthebgnormalbgpopulationbgwithbgthebglargestbgmeanbgspe
cifiesbgthebgappropriatebgnumberbgofbgobservationsbgtobgtakebgfrombgeachbgcompetingbgpopulation,bga
ndbgsimplybgselectsbgthebgcompetitorbghavingbgthebglargestbgsamplebgmean.bg-bganswers--True
TRUEbgorbgFALSE?bgSometimesbgabgsingle-
stagebgprocedurebglikebgBechhofer'sbgisbginefficient.bgInbgfact,bgit'sbgpossiblebgtobgusebgcertainbgsequenti
albgproceduresbgthatbgtakebgobservationsbgone-at-a-
timebg(insteadbgofbgallbgatbgoncebginbgabgsinglebgstage)bgtobgmakebggoodbgselectionbgdecisionsbgusingbgf
ewerbgobservations.bg-bganswers--True
Forbgwhichbgscenarios(s)bgbelowbgmightbgitbgbebgappropriatebgtobgusebgabgBernoullibgselectionbgproced
ure?
a)bgFindbgthebginventorybgpolicybghavingbgthebglargestbgprofit.
b)bgFindbgthebgdrugbggivingbgthebgbestbgchancebgofbgabgcure.
c)bgFindbgthebgmaintenancebgpolicybghavingbgthebglowestbgfailurebgprobability.
d)bgFindbgthebgschedulingbgrulebgthatbgthatbghasbgthebgbestbgchancebgofbgmakingbganbgon-
timebgdelivery.bg-bganswers--Allbgthreebgofbg(b),bg(c),bgandbg(d).
SupposebgthatbgabgBernoullibgselectionbgprocedurebgtellsbgyoubgtobgtakebg100bgobservationsbgfrombgeac
hbgofbgtwobgpopulations,bgAbgandbgB.bgItbgturnsbgoutbgthatbgAbggetsbg85bgsuccessesbgandbgBbggetsbg46bg
successes.bgWhatbgdobgyoubgthink?bg-bganswers--
1)bgAbgalmostbgcertainlybghasbgabghigherbgsuccessbgprobabilitybgthanbgB.
,2)bgWebgcould'vebgprobablybgstoppedbgsamplingbgabgbitbgearlierbg(i.e.,bgwithbgfewerbgthanbg100bgobserv
ations)bgbecausebgAbgwasbgsobgfarbgaheadbgofbgB.
Forbgwhichbgscenarios(s)bgbelowbgmightbgitbgbebgappropriatebgtobgusebgabgmultinomialbgselectionbgproc
edure?bg-bganswers--Findbgthebgmost-popularbgpoliticalbgcandidate.
SupposebgthatbgwebgwantbgtobgknowbgwhichbgofbgCoke,bgPepsibgandbgDr.pepperbgisbgthebgmostbgpopular
.bgWebgwouldbglikebgtobgmakebgthebgcorrectbgselectionbgwithbgprobabilitybgofbgatbgleastbgP*=0.90bginbgt
hebgeventbgthatbgthebgrationbgofbgthebghighest-to-second-
highestbgpreferencebgprobabilitiesbghappensbgtobgbebgatbgleastbg0*=1.4.bgHowbgmanybgpeoplebgdoesbgt
hebgsingle-stagebgprocedurebgMbembgrequirebgusbgtobginterview?bg-bganswers--126
Gobgtobgthebgtablebginbgthebgnotesbgandbgpickbgoffbgthebgentrybgforbgk=3,bgP^⋆=0.90,bgandbgθ^⋆=1.4.
Whichbgofbgthebgfollowingbgparametersbgcanbgyoubggetbgconfidencebgintervalsbgfor?
Means
Variances
Quantiles
Differencesbgbetweenbgthebgmeansbgofbgtwobgsystemsbg-bganswers--Allbgofbgthebgabove
Ifbgwebghavebganbgiidbgnormalbgsamplebgofbgobservations,bgX1,bgX2,...Xn,bgwhatbgprobabilitybgdistributio
nbgisbgmost-commonlybgusedbgtobgobtainbgcofidencebgintervalsbgforbgthebgmean?bg-bganswers--t
TRUEbgorbgFALSE?
bgThebgpairedbgCIbgforbgthebgdifferencesbginbgtwobgmeansbgisbgdesignedbgtobgworkbgespeciallybgwellbgifbg
allbgofbgthebgobservationsbgfrombgthebgfirstbgpopulationbgarebgcompletelybgindependentbgofbgallbgofbgth
ebgobservationsbgfrombgthebgsecondbgpopulation.bg-bganswers--
FALSE.bg{Inbgfact,bgit'sbgeasierbgtobgdistinguishbgbetweenbgthebgtwobgmeansbgifbgXibgisbgpositivelybgcorrel
atedbgwithbgYi.bgThinkbgaboutbgmybgparallelbgparkingbgexamplebginbgthebgclassbgnotes.}
TRUEbgorbgFALSE?
bgYoubgcanbgusebgabgversionbgofbgindependentbgreplicationsbgtobgobtainbgconfidencebgintervalsbgforbgthe
bgdifferencebginbgthebgmeansbgfrombgtwobgsimulationbgmodels.bg-bganswers--TRUE
, TRUEbgorbgFALSE?
bgThebgcommonbgrandombgnumbersbgtechniquebgintentionallybginducesbgpositivebgcorrelationbgbetwee
nbgtwobgsystemsbg-bgmuchbglikebgabgpaired-𝑡confidencebginterval.bg-bganswers--TRUE
CRNbgdependsbgonbgsomeone'sbgabilitybgtobgmanipulatebgthebgunderlyingbgpseudo-
randombgnumbersbg-
bge.g.,bgusebgthebgsamebgarrivalbgtimesbgwhenbgsimulatingbgtwobgcompetingbgsimulatedbgsystems.bgSobg
whobgultimatelybgcontrolsbgthosebgPRNs??bg-bganswers--Youbgdobg-bgyoubgarebgpowerful!
TRUEbgorbgFALSE?
bgThebgantitheticbgrandombgnumbersbgtechniquebgintentionallybginducesbgnegativebgcorrelationbgbetwe
enbgtwobgrunsbgofbgthebgsamebgsystembg-
bgthisbgallowsbgusbgtobgbetterbgestimatebgthebgmeanbgofbgthebgsystem.bg-bganswers--TRUE
TRUEbgorbgFALSE?bgThebgcontrolbgvariatesbgtechniquebgprovidesbgunbiased,bglow-
variancebgestimatesbgusingbgabgmethodbgreminiscentbgofbgregression.bg-bganswers--TRUE
Statisticalbgrankingbgandbgselectionbgtechniquesbghavebgbeenbgdesignedbgtobgaddressbgabgvarietybgofbgc
omparisonbgproblems.bgWhichbgonesbgfrombgthebgfollowingbglist?
Findbgthebgpopulationbghavingbgthebglargestbgmean.
Findbgthebgsystembgwithbgthebgsmallestbgvariance.
Findbgthebgalternativebgwithbgthebghighestbgsuccessbgprobability.
Findbgthebgmost-popularbgcandidate.
Allbgofbgthebgabove.bg-bganswers--Allbgofbgthebgabove.
Supposebgwebgarebgdealingbgwithbgi.i.d.bgnormalbgobservationsbgwithbgunknownbgvariance.bgWhichbgofb
gthebgfollowingbgisbgtruebgaboutbgabg95%bgconfidencebgintervalbgforbgthebgmeanbgμ?bg-bganswers--
Webgarebg95%bgsurebgthatbgourbgCIbgwillbgactuallybgcontainbgthebgunknownbgvaluebgofbgμ.
Webgarebgstudyingbgthebgwaitingbgtimesbgarisingbgfrombgtwobgqueueingbgsystems.bgSupposebgwebgmake
bg4bgindependentbgreplicationsbgofbgbothbgsystems,bgwherebgthebgsystemsbgarebgsimulatedbgindepende
ntlybgofbgeachbgother.bg
replicationbgsystembg1bgsystem2
1bg10bg25