Scribd
Upload
65 views3 pages

Mid Term Solutionscz

This document discusses probability plots, probability distributions, and calculating percentiles from different distributions. It provides examples of: 1) Constructing a probability plot to test if data comes from a uniform distribution. 2) The slope and intercept of a probability plot if data comes from a normal distribution. 3) Situations that are or are not possible based on constraints of different distributions. 4) Using the binomial distribution to test if a geostatistical modeling method is working properly. 5) Calculating the probability of two different reservoir interpretations given well data using Bayes' theorem. 6) Calculating the 2.5th, 50th, and 84th percentiles for data from normal, lognormal, uniform

Uploaded by

Zoha Ahmed
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
65 views3 pages

Mid Term Solutionscz

This document discusses probability plots, probability distributions, and calculating percentiles from different distributions. It provides examples of: 1) Constructing a probability plot to test if data comes from a uniform distribution. 2) The slope and intercept of a probability plot if data comes from a normal distribution. 3) Situations that are or are not possible based on constraints of different distributions. 4) Using the binomial distribution to test if a geostatistical modeling method is working properly. 5) Calculating the probability of two different reservoir interpretations given well data using Bayes' theorem. 6) Calculating the 2.5th, 50th, and 84th percentiles for data from normal, lognormal, uniform

Uploaded by

Zoha Ahmed
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

1.

Probabilityplots
a. SupposeIwanttoseeifasetofmeasurementsmightbefromauniformPDF.Definethe
variables(iethehorizontalandverticalaxes)neededfortheplot.
Oneaxiswillbethemeasurementvalueswhiletheotheraxisistheprobabilityvalueofeach
measurement.Thelatteraxisisbecause,forthestandardU(0,1)uniformdistribution,F1(p)=p.
b.

SupposeIhaveprobabilitypaperwhichteststhesamplePDFofXagainststandarddeviatesof
theN(0,1)normaldistribution.IfX~N(a,b2),givetheslopeandinterceptoftheresultingline.
TheslopeisgivenbythenumberofstandarddeviationsofXoccurwithonestandarddeviation
changeofthestandardnormal.Inthiscase,wehavebSDschange.theinterceptisthevalueof
Xwhenthestandardnormalhaszerodeviates.thatvalueisa,themedianofX.

2.

Stateifandwhyorwhynotitispossibletohavethefollowingsituations.
a. PermeabilitywithnormalPDFandCv=1.
notpossible.permisnonnegativeand,ifCv=1,thereisa16%chancethepermisnegative.
b.

PorositywithlognormalPDFandCv=1.
notpossible.porositymustbelessthanorequalto1andforthelognormalwith>0,Prob(X>
1)>0.Cv=1means=0.8

c.

WatersaturationwithuniformdistributionandCv=1.
Notpossible.CvofauniformRV<1/sqrt(3)=0.6

d.

ReservesdistributioninabasinwithaParetoPDF.
Possible.therecanbemany,smallaccumulationsandjustafewbigpools.

3.

Areservoirhastworocktypes,AandB,whicharetobegeostatisticallymodelledwithProb(A)=0.6and
Prob(B)=0.4.Totestwhetherthemodelisworkingcorrectly,someonesamples10wellseparatedpoints
ofonerealisationandfindsthat8fallontypeAand2ontypeB.Giveaquantitativeargumenttoshow
whethermethodisworkingproperly.
wehaveabinaryvariablehere,sothebinomialPDFisappropriate.LetProb(A)=p=0.6,soProb(B)=
Prob(Ac)=1p=0.4.TheProb(8Asand2Bs)=10C8p8(1p)2=[10!/(8!2!)](0.6)8(0.4)2=45x0.017x
0.16=0.12
sothereisa12%probthat,withthegivenprobsforAandB,wewouldget8Asand2Bs.fromthis
evidence,thereisnobasistosuspectthemethodisnotworkingproperly.

4.

Analysis of seismic data indicates a potential reservoir, but the interpretation is uncertain. Version A
suggestsalimitedstructurewhichmightbemarginallycommercial.VersionBisamuchlargerstructure
with high potential for commercial amounts of hydrocarbons. The geophysicist suggests Prob(A) = 0.4
andProb(B)=0.6.

One exploration well (EW) is drilled to help clarify which situation exists. Under similar conditions
elsewhere,experiencehasshownthatsuchwellshavethefollowingperformance:Prob(EW=A|A)=0.8
andProb(EW=B|B)=0.6.

CalculatetheprobabilityofversionBifthewellindicatesBexists.
TheaprioriprobsareProb(A)=0.4andProb(B)=0.6.
Wealsohave
Prob(EW=A|A)=0.8andProb(EW=B|A)=0.2
Prob(EW=B|B)=0.6andProb(EW=A|B)=0.4

WeseekProb(B|EW=B).

FromBayes,Prob(B|EW=B)=Prob(EW=B|B)xProb(B)/Prob(EW=B)

Now,Prob(EW=B)=Prob(EW=B|B)xProb(B)+Prob(EW=B|A)xProb(A),so

Prob(B|EW=B)=0.6x0.6/(0.6x0.6+0.2x0.4)=0.36/0.44=0.82
5.

Forthefollowingsituations,statethevaluesofthe2.5th,50th,and84thpercentilesoftherandomvariable
X.
Answerstoa.andb.arebasedonthevaluesdiscussedinclassforthenormalPDF.
a.

b.

c.

d.

X~N(3,22)
i. 2=32x2=1
ii. =3
iii. +=3+2=5
log10(X)~N(0,1)
i. 102=1002x1=102=1/100=0.01
ii. 10=100=1
iii. 10+=100+1=10
X~U(1,3)
i. 0.025x2+1=1.05
ii. 0.5x2+1=2
iii. 0.84x2+1=2.68
XhasanexponentialPDFwithmeanof2,soF(X)=1exp(x/2)
i. 0.025=1exp(X0.025/2)givesX0.025=0.05
ii. X0.5=1.38
iii. X0.84=3.7

You might also like