lecture 2 & 3. basic concepts and review of probabilityyxie77/isye2028/lecture2_final.pdf•...
TRANSCRIPT
![Page 1: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/1.jpg)
Lecture 2 & 3. Basic Concepts and Review of
Probability
• Logistics
• Basic concepts
• Review of probability
• Use of R
Prof. Yao Xie, ISyE 2028, Basic Statistical Methods, Georgia Tech
![Page 2: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/2.jpg)
• TA: Caglar Caglayan, office hour: W 12:00-1:00pm, Main Building 321
• Grading
– Class Attendance 3%– Submitting Teaching Evaluation 2%– Homework - 15%– Computer Example 1 - 10%, Computer Exam 2 - 10%– Midterm 1 - 15%, Midterm 2 - 15%– Final - 30%
• Homework grades: {0, 1, 2}. Homeworks are aimed for practice.Students are responsible to check details with solutions.
Prof. Yao Xie, ISyE 2028, Basic Statistical Methods, Georgia Tech 1
![Page 3: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/3.jpg)
Winning the number guessing
0 10 20 30 40 50 60 70 80 90 1000
0.5
1
1.5
2
2.5
3
Winner: Mallory Weaver, Number: 3
Prof. Yao Xie, ISyE 2028, Basic Statistical Methods, Georgia Tech 3
![Page 4: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/4.jpg)
List of movies
• Fight club
• Harry Potter and Hunger Games
• Game of thrones
• Dirty Dancing
• The Shawshank Redemption
• Coach Carter
• Money Ball
• 500 Days of Summer
• The Catcher in the Rye
• The Prestige
• The silence of the Lambs
• The Royal Tenenbaums
• The Shack, 1984
• Fahrenheit 451
• The Girl with the Dragon Tattoo
• Crazy Stupid Love
• Jurassic Park
• The Great Gatsby
• Forrest Gump
Prof. Yao Xie, ISyE 2028, Basic Statistical Methods, Georgia Tech 4
![Page 5: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/5.jpg)
• The Giver
• Forrest Gump
• Howl’s Moving Castle
• The Truman Show
• Equilibrium
• The Sherlock
• Infinite Jest
Prof. Yao Xie, ISyE 2028, Basic Statistical Methods, Georgia Tech 5
![Page 6: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/6.jpg)
The'field'of'Sta2s2cs'deals'with'the'collec%on,'presenta%on,'analysis,'and'use'of'data'to'model'systems,'make'decisions,'solve'problems,'and'design'products'and'processes.'
///////////////////////////////////////Sta%s%cs/is/the/science/of/data/
Examples:'Sta2s2cs'helps'us'in'• 'sports'(ref:'movie'``Money'Ball'')'• 'stock'market/finance'• 'weather'forecast'• 'machine'learning/computer/internet'• 'poli2cs'• 'biology/medicine'
Statistics
Data'
Decision/Inference'
Sta%s%cs/'
![Page 7: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/7.jpg)
Where'data'come'from'
• Retrospec2ve'study'using'historical'data''e.g.'historical'temperature'record'of'Atlanta'
'• Observa2onal'study''e.g.'measurements'of'digital'thermometer'
'• A'designed'experiments''e.g.'to'test'if'a'drug'is'effec2ve'
![Page 8: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/8.jpg)
• 'Sta2s2cal'methods'are'useful'for'describing'and'understanding'variability.'
• ''By'variability,'we'mean'successive'observa2ons'of'a'system'or'phenomenon'do'not'produce'exactly'the'same'result.'
• 'Sta2s2cs'gives'us'a'framework'for'describing'this'variability'and'for'learning'about'poten2al'sources'of'variability.'
Large'Variability'
Small'Variability'
'Sta2s2cal'Methods'
� ������ ��
���� ����������� ��������
![Page 9: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/9.jpg)
'Sta2s2cal'Methods'
• Point'es2mator''''''• Confidence'Interval'''''• Hypothesis'test''''''• Two'sample'/'ANOVA'
• Linear'regression'''''''
![Page 10: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/10.jpg)
'Sta2s2cal'Methods'• Point'es2mator'''''e.g.'es2ma2ng'temperature'from'5'digital'thermometer'measurements'
• Confidence'Interval'''''e.g.'giving'an'interval'where'true'temperature'is'mostly'likely'to'be'within'
• Hypothesis'test'''''e.g.'tes2ng'good/bad'the'quality'of'a'batch'of'laptops.'• Two'sample'test'/'ANOVA'''''e.g.'test'effec2veness'of'a'new'drug'
• Linear'regression'''''e.g.'fibng'a'linear'model'from'input/output''measurements''''
![Page 11: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/11.jpg)
'Sta2s2cal'Methods'• Point'es2mator'''''estimate parameters of a distribution based on a random sample'e.g.'es2ma2ng'temperature'from'5'digital'thermometer'measurements'
• Confidence'Interval'''''e.g.'giving'an'interval'where'true'temperature'is'mostly'likely'to'be'within'
''''
?)?,(~ 2 == σµNX
2ˆ
ˆ
σ
µpopulation
sample
Point estimates
( )( )22 ˆ,ˆ
ˆ,ˆ
UL
UL
σσ
µµ
Interval estimates
![Page 12: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/12.jpg)
'Sta2s2cal'Methods'• Hypothesis'test'''''Make a decision about a population based on a random sample e.g.'tes2ng'good/bad'the'quality'of'a'batch'of'laptops.'
• Two'sample'test'/'ANOVA'''''e.g.'test'effec2veness'of'a'new'drug''''
Popula2on'(Lot)' Sample'
Accept
Reject'
![Page 13: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/13.jpg)
'Sta2s2cal'Methods'• Linear'regression'''''Predict'a'response'variable'based'on'one'or'more'predictor'variables'e.g.'fibng'a'linear'model'from'input/output'measurements,'say'voltage'and'current'of'a'resistor''''
![Page 14: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/14.jpg)
An'engineer'is'someone'who'solves'problems'of'interest'to'society'by'the'efficient'applica2on'of'scien2fic'principles'through'
• 'Refining'exis2ng'methods'or'products'
• 'Designing'new'methods'or'products''
Figure 1.1. The engineering method
Role of Engineers
![Page 15: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/15.jpg)
Defini2on'1:'(MerriameWebster'dic2onary)'
something'(as'a'similar'object'or'a'construct)'used'to'help'visualize'or'explore'something'else'(as'the'living'human'body)'that'cannot'be'directly'observed'or'experimented'on.'
Defini2on'2:'(Law'and'Kelton)'
A'representa2on'of'the'system'that'is'studied'as'a'surrogate'of'the'actual'system''
A.'Law'and'D.'Kelton,'Simula2on'Modeling'and'Analysis,'3/e,'McGraweHill,''2000'
Model
![Page 16: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/16.jpg)
Probability Vs. Statistics
Popula2on' Sample'
Probability'
Inferen2al''Sta2s2cs'
Probability:'given'the'informa2on'in'the'pail,'what'is'in'your'hand?'''''Sta2s2cs:'given'the'informa2on'in'your'hand,'what'is'in'the'pail?''
![Page 17: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/17.jpg)
Review'of'Probability'
• Key'words'
• 'Popula2on,'Random'Sample,'Sample'Space'
• 'Law'of'Total'Probability,'Bayes'Theorem'
• 'Condi2onal'Probability,'Independence'
• 'Events,'Compliments,'Intersec2ons,'Unions'
• 'Outcomes,'Outcome'Probability'
![Page 18: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/18.jpg)
Population Vs. Sample
• Population: a finite well-defined group of ALL objects which, although possibly large, can be enumerated in theory (e.g. Atlanta population, Georgia Tech students).
• Sample: A sample is a SUBSET of a population (e.g. select 50 out of 1,000 bearings manufactured today). '
'Popula2on''''''''
Sample''
Observa2on'
![Page 19: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/19.jpg)
Other'terms'! Random'Sample'='randomly'selected'sample'
! Sample'Size'='number'of'object'in'the'sample'
! Experiment'='measure'characteris2cs'of'the'sample/popula2on''
! Outcome'='possible'values'of'the'measurements'
! Sample'Space'=''space'of'all'possible'outcomes'
![Page 20: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/20.jpg)
Example'
! Experiment:'Roll'two'dies'simultaneously'! Sample'space:'{(1,1),(1,2),..,(1,6),..., (5,6),
(6,6) } ! Outcome:'(6,6)'! What'is'the'popula2on?'
36 possible outcomes: 6 for each die
{(1,1),(1,2),..,(1,6),..., (5,6),(6,6) }
![Page 21: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/21.jpg)
Outcomes'and'Events'! An'Event'is'any'collec2on'of'sample'outcomes'! Simple'Event'='Collec2on'of'outcomes''! Compound'Event'='Collec2on'of'outcomes'
described'as'unions'or'intersec2ons'of'other'events'
36 possible outcomes: 6 for each die
{ Sum = 11 } = { (6,5), (5,6) }
![Page 22: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/22.jpg)
Outcome'probability'
• Probability'='likelihood'of'the'occurrence'of'an'outcome'in'the'sample'space''
{ } { }ii
nn
poPppooS
=
→=
)(,...,,..., 11
1yprobabilit0 ≤≤
![Page 23: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/23.jpg)
Example(Example: Roll a die twice. 1. The number of outcomes when at least one�6��
occurs in the combination of 2 dices: A. 1 outcome B. 6 outcomes C. 11 outcomes 2. The probability for the event that {at least one�6��
occurs in the combination of 2 dices}: A. 1/6 B. 11/36 C. 1/36
![Page 24: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/24.jpg)
Unions and Intersections
P(AUB)'=P(Outcome'contained'in'Event'A'OR'Event'B)'
P(AUB) = P(A) + P(B) – P(AB)
P(AB)'='P(Outcome'contained'in'both'Event'A'AND'Event'B)'
BAAB =
![Page 25: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/25.jpg)
Conditional Probability
P(A | B) = P(AB) / P(B)
Probability'that'event'A'happens'GIVEN'that''event'B'is'known'to'happen'
![Page 26: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/26.jpg)
Combination of Events: Example
At Georgia Tech, two major credit card providers want to investigate how many students have a Visa credit card and how many students have a MasterCard. From our survey we find that the proportion of GT students holding a Visa credit card is 50%, the proportion of GT students holding a MasterCard is 40% and that the proportion of students holding both credit cards is 25%. Let A be the event that a selected student has a Visa credit card and let B be the event that a selected student has a MasterCard.
![Page 27: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/27.jpg)
Combination of Events: Example
1. What is the probability that the selected individual has at least one of the two types of cards?
2. What is the probability that the selected student has neither type of card?
3. What is the probability that the selected student has a Visa card but not a Master-Card?
4. What is the probability that the selected student has a MasterCard, given that we know already that he/she has a Visa credit card?
![Page 28: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/28.jpg)
Combination of Events: Example
A = event that a selected student has a Visa Card (P(A)=.5) B = event that a selected student has a Master Card (P(B)=.4) Additional Information: P(AB) = .25 • P(AUB) • P((AUB)�) • P(A|B) • P(B|A)
Use Venn Diagram
![Page 29: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/29.jpg)
Independence
P( King ) = 4/52 = 1/13, and
P(King | Red) = 1/13.
Informa2on'about'the'card'color'did'not'effect'the'probability'of'drawing'a'King.'This'is'the'idea'of'independent'events.'
'''''''''$
Eg. A card is drawn at random from a pack of cards. Calculate the following probabilities:
Events$A$and$B$are$independent$if$$$
$$$$$$$$$$$$$$$$$$$$$$$P($A$|$B$)$=$P($A$)$
$$$$$$$$$$$$$$$$$$$$$$$P($AB$)$=$P($A$)P(B)$
![Page 30: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/30.jpg)
Independence - Definition
∏==
=""#
$%%&
'=∩∩∩
n
ii
n
iin APAPAAAP
1121 )()...(
If'events'A1,…,An'are'all'independent'of'each'other,''
then''
![Page 31: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/31.jpg)
Bayes� Theorem
∑=
== n
kkk
iiii
ABPAP
ABPAPBPBAPBAP
1)|()(
)|()()()()|(
EXAMPLE: For the GT student credit card example
1. Calculate P(A|B) by
2. Calculate P(B|A) using P(A|B) calculated in 1 and Bayes’ theorem
P(B | A) = P(A | B)P(B)P(A)
![Page 32: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/32.jpg)
General Multiplication Rule
)...|()...|()|()()...(
121213121
21
−
=∩∩∩
nn
n
AAAAPAAAPAAPAPAAAP
Extension'to'n'events'A1,…,An'
Mul2plica2on'Law:'P(AB)$=$P(A)P(B$|$A)$$Extension'to'three'events:'''P(ABC)$=$P(AB)P(C|AB)$=$P(A)P(B$|$A)P(C|AB)$
![Page 33: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/33.jpg)
Law of Total Probability
If {A1,…,An } represents a Partition of the whole space, then these sets are
mutually exclusive and
1)()...(11
21 ==!!"
#$$%
&=∪∪∪ ∑
==
n
ii
n
iin APAPAAAP
P(B)$=$$P(A1$)P($B$|$A1$)$+$P(A2$)P($B$|$A2$)$+$…$+$P(An$)P($B$|$An$)$$
so'that'
![Page 34: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/34.jpg)
Random Variables
In'an'experiment,'only'certain'factors'and'outcomes'are'of'interest'to'us.'
A'Random$Variable'is'the'number'assigned'to'the'(random)'outcome'of'an'experiment'in'order'to'summarize'its'meaning.'
'Example:'A'researcher'may'test'a'sample'of'components'from'a'produc2on'line'and'record'only'the/number/of/components'that'have'failed'within'12'hours,'rather'than'record'individual'component'failure'2mes.'
![Page 35: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/35.jpg)
Random Variables
DefiniBon:'A'Random$Variable'is'an'associa2on'rule'or'func2on'from'the'sample/space'to'the'state/(range)/space:'
'''''''''''''''''''RV:'S'''''''''''R'
where'the'characteris2c'measurement'of'outcome'o'is'x.'
Example'cont.:'For'each'component,'we'observe'0�s'(failed)'and'1�s'(not'failed).''What'are'possible'outcomes?'What'are'the'possible'values'of'the'random'variable?''
![Page 36: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/36.jpg)
Discrete Random Variables
Random'variables'(RV)'have'a'finite'or'countably/infinite'range,'and'we'can'write'the'outcomes'as'{x1,x2,….}.''
'
The'set'of'probabili2es'associated'with'the'outcomes'
' 'pi'='P(X'='xi)'for'each'xi'
Probability$Mass$FuncBon'(PMF)'for'the'discrete'random'variable'X:'
'''''''''''''''''''0 ≤ pi ≤1, pi =1∑
![Page 37: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/37.jpg)
Cumulative Distribution Function
The'CumulaBve$DistribuBon$FuncBon'(CDF)'of'a'random'variable'is'defined'as'
Note'that''0'≤'F(x)'≤'1'and'F(x)'is'increasing'in'x.'In'the'dice'example:'''
' 'F(x)'='x/6'for'x=1,2,3,4,5,6.'
'
∑≤
=xx
ii
xXP )(FX(x)'or'F(x)'='P('X'≤'x')'='
![Page 38: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/38.jpg)
Discrete vs. Continuous RVs
Random Variables
Discrete: X takes discrete values
PMF: pi = P(X = xi)
Continuous: X can be any (real) value in a given interval, so outcomes are not �countable�
Probability Density Function (PDF) or density function for continuous random variables X
-4 -2 2 4
0.1
0.2
0.3
0.4
-4 -2 2 4
0.1
0.2
0.3
0.4
F(x) = P(X ≤ x)
f (x) = ddxF(x),F(x) = f (t)dt
−∞
x
∫ f (x) ≥ 0
f (t)dt∫ =1
![Page 39: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/39.jpg)
Example: Continuous Density
F(x) = x2/16, 0 ≤ x ≤ 4 P( X ≤ 2 ) = F( 2 ) = 4/16 = ¼ P( 1 ≤ X ≤ 3 ) = P(X ≤ 3)-P(X ≤ 1)
= F( 3 ) – F( 1 ) = 9/16 - 1/16 = 1/2
f (x) = ∂
∂xFX (x) = x / 8
![Page 40: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/40.jpg)
Expectation
We'are'interested'in'summary'values'to'measure'the'typical'value'of'a'RV'(called'the'mean)'or'how'much'it'varies'(called'variance).'
∑ ==i
ii xXPxXE )()(
∫∞
∞−
= dxxxfXE )()(
'1.'Discrete'Case'
'
'2.'Con2nuous'Case'''
Expecta2on/mean'of'X'is'E(X)'='m:'
![Page 41: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/41.jpg)
Variance
( )( ) 2222 )()()()( XEXEXEXEXVar X −=−=≡σ
Standard Deviation is σx (the square root of the variance)
![Page 42: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/42.jpg)
Linear'Transform'of'RV'
Given'a'RV'X:'with'' 'E(X)'='μ'' 'Var(X)'='σ2'
''Let'another'RV:'''''''''''''''''''''Y'='aX'+'b'''Then'''''''''''''''''''''' 'E(Y)'='aμ'+'b''''''''''''' 'Var(Y)'='a2σ2'''
![Page 43: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/43.jpg)
X = outcome of a roll of 6-sided die
pi = 1/6, 1 ≤ i ≤ 6, and 5.32)16(6
61
61)(
6
1
=!"
#$%
& +==∑
=iiXE
Var(X) = σ2 = 15.167 – (3.5)2 = 2.9167, σ = 1.71
167.1561376
61
6)()(
6
1
26
1
22 =!"
#$%
& ⋅⋅==== ∑∑
== iiii
ixXPxXE
Discrete'RV'
![Page 44: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/44.jpg)
F(x) = x2/16, 0 < x < 4
94.089.0
,89.0)667.2(8)()(
8328
)(
,667.2248
)(
,8
)(
2222
4
0
4
0
432
4
0
4
0
32
==
=−=−=
="#
$==
="#
$==
=
∫
∫
σ
σ XEXE
xdxxXE
xdxxXE
xxf
0 1 2 3 4
f(x)
Continuous RV
![Page 45: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/45.jpg)
-4 -2 2 4
0.1
0.2
0.3
0.4
-4 -2 2 4
0.1
0.2
0.3
0.4
Quartiles
• xp is called the pth quartile if F(xp) = p.
• x0.5 = Median
• Quartiles are useful in describing what values we might see for X.
area = p
area = 1-p
xp
![Page 46: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/46.jpg)
Example continued
To solve for the 0.95 quantile, find t such that F(t) = 0.95:
t2/16 = 0.95 means t = 3.90
x.25'='lower'quar2le:'(x.25')2/16'='0.25''''''''x.25'=''2'
x.5'''='median:'(x.5')2'/16'='0.5'''''''''''''''''''''x.5'=''2.82'
x.75'='upper'quar2le:'(x.75')2'/16'='0.75'''''''x.75'='3.46'
Quar%les:''
![Page 47: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/47.jpg)
Review: Discrete Distributions
• Binomial$DistribuBon$
• 'Geometric'Distribu2on'
• 'Nega2ve'Binomial'Distribu2on'
• 'Hypergeometric'Distribu2on'
• 'Poisson'Distribu2on'
• 'Mul2nomial'Distribu2on'
![Page 48: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/48.jpg)
Binomial Distribution
Binomial random variables occur frequently in nature:
!Assign!X!=!number!of!successes!out!of!n!trials!
!We!say!that!X!follows!a!Binomial(distribu/on.!
!
!
• !P(success)=p!is!the!same!in!every!trial!
• !Two!outcomes!in!every!trial!(e.g.,!success!vs.!failure)!
• !n!independent!trials!or!events,!n!fixed!
• Write as X~B(n,p) or X~Bin(n,p)
• e.g. a batch of n = 12 laptops, the probability of each one of them being defective is p = 0.1, independent of each other, what’s the probability that having X = 2 defective ones in this batch?
![Page 49: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/49.jpg)
Binomial pmf
for!k=0,1,2,…,n!knk ppkn
kXP −−""#
$%%&
'== )1()(
If P(success) = p, the P(X = k) = P(k successes and n-k failures). Each outcome like this has probability pk(1-p)n-k.
From counting rules, we know there are ways of choosing k successes from a group of k successes and n-k failures. As a result, we have
!!"
#$$%
&
kn
![Page 50: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/50.jpg)
Binomial – Expected Value and VAR
knkn
k
n
kpp
knknkkXkPXE −
==
−"#
$%&
'−
=== ∑∑ )1(!!
!)()(11
npppjn
nppknk
nn jnjn
j
knkn
k=−""
#
$%%&
' −=−"
#
$%&
'−−
−= −−+
−
=
−
=∑∑ 111
01)1(
1)1(
!!1!1
E(X) = np
VAR(X) = n(1-p)p
![Page 51: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/51.jpg)
Binomial Example With'a'true/false'test'that'has'20'ques2ons,'let'X'='no.'correctly'answered'out'of'20.'For'a'grade'of'A,'one'needs'X'≥'19.'What'is'the'probability'a'student'who'guesses'all'the'answers'scores'an'A'on'this'exam?'
( ) ( ) ( ) ( ) 00002.02112
12020
2112
11920 020119
≈−##$
%&&'
(+−##
$
%&&'
(=
In this case E(X) = np = 10, Var(X) = np(1-p) = 5
P(get'an'A)'='P(X'≥'19)'='P(X=19)'+'P(X=20)'where'X~B(n=20,'p='½')''
≥≤'
![Page 52: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/52.jpg)
Review: Continuous Distributions
• Uniform Distribution
• Exponential Distribution
• Gamma Distribution
• Weibull Distribution
• Beta Distribution
• Normal Distribution
![Page 53: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/53.jpg)
X is distributed as a normal random variable with mean µ and variance σ2; i.e., X ~N(µ , σ2)
∞<<∞=−
−x
x
- ,e2
1 f(x)
2)(21
2σµ
πσ
Normal Distribution
-4 -2 2 4
0.1
0.2
0.3
0.4
Standard normal random variable Z: µ =0, σ2 =1
![Page 54: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/54.jpg)
Example :
Φz(1) =P(Z ≤1) =0.8413
)x(du)u(f )x P(X F(x)x
-x∫
∞
Φ≡=≤=
-3 -2 -1 1 2 3
0.1
0.2
0.3
0.4
CDF for Normal Distribution
![Page 55: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/55.jpg)
Properties of Standard Normal Distribution
• Φz(0) =1/2, Φz(∞) =1
• Φz(t) =1 - Φz(-t)
P(Z ≤t)=1-P(Z ≤ -t)
• P(Z ≥ t)=P(Z ≤ -t) -3 -2 -1 1 2 3
0.1
0.2
0.3
0.4
Example : P(Z ≥-1) =P(Z ≤1)
=0.8413
![Page 56: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/56.jpg)
-3 -2 -1 1 2 3
0.1
0.2
0.3
0.4
If'X'~N(µ',σ2)'then''Z'=[(Xeµ)/σ]'~N(0,'1)''Example:'Suppose'X'~N(100,'9),'so'σ'=3'P(X≤105)'=P(Xeµ'≤105'eµ)'='
' ''''''=P((Xeµ)/σ'≤(105'eµ)/σ)'=P(Z'≤(105'–100)/3)'' ''''''=P(Z'≤5/3)'=P(Z'≤1.67)'=0.9525'
Calculating Normal Probabilities
![Page 57: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/57.jpg)
-1 0 1 2
0.1
0.2
0.3
0.4
Suppose again X ~N(100, 9): P(99< X <103) = P((99 –100)/3 ≤(X - µ)/σ ≤(103 –100)/3) = P(-1/3 <Z <1) = P(Z<1) -P(Z<–1/3) = 0.8413 – 0.3707 = 0.4706
![Page 58: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/58.jpg)
-3 -2 -1 1 2 3
0.1
0.2
0.3
0.4
Computing Normal Quartiles
With%X%~N(100,%9),%what%is%the%value%X*%such%that%%P(X%>X*)%=0.01%?%We%will%first%find%P(X<X*)(=%1BP(X>X*))%
Φ(2.33) =0.99
P(X%<X*)%=%P((X%Bµ)/σ%<(X*%–100)/3)%=%P(Z%<Z*)%%=%0.99%
P(Z%>2.33)%=0.01%and%therefore%%
Z*%=(X*B100)/3%=2.33.%%%
Solve%for%X*%to%get%3(2.33)%+100%=106.99%
![Page 59: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/59.jpg)
Joint Discrete Distributions Random'variables'can'be'mul1variate'–'example:'consider'a'bivariate'random'variables'(X,Y)'
Joint&probability&mass&func2on&of&(X,Y): &'
∑ ∑∈ ∈
===X Yx y
yYxXPR R
1),(
),(),(),(
]1,0[:::
),(
),(
yYxXPyxpyxp
pYX
YX
YXYX
YY
XX
====
→×
→
→
RR
RS
RS
![Page 60: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/60.jpg)
Joint Discrete Distributions
∑∑≤ ≤
==xs yt
tYsXP ),(F(x,y)'='P('X'≤'x,Y'≤'y)='
0'≤'F(x,y)'≤'1'
Cumula&ve)distribu&on)func&on: ''
Marginal)Probability)func&ons:)''
∑
∑
�
�
),()(
),()(
X
Y
xY
yX
yxpyp
yxpxp
R
R
=
=
![Page 61: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/61.jpg)
Random Vector Example
(x,y) = (1,1) means 12% of calls are for places with one AC, service time = 1 hr.
Phone call center, for air conditioner maintenance:
X = service time (hours) spent for a request
Y = number of ACs in the unit being requested for service
0.12 0.08 0.07 0.05 0.08 0.15 0.21 0.13 0.01 0.01 0.02 0.07 0.21 0.24 0.30 0.25
X=1 X=2 X=3 X=4
Y = 1 Y = 2 Y = 3
Y = 4
![Page 62: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/62.jpg)
Example, continued
AC Maintenance example:
• P[(X,Y)=(4,2)] = p42 = 0.13
• P[(X,Y) ≤(2,2)] = p11 + p12 + p21 + p22 = 0.43
• P[X ≤ 2,Y=3] = p13 + p23 = 0.01 + 0.01 = 0.02
• P[X ≤ 1] = p11+p12+p13 = 0.21, P[X ≤ 2] = 0.45, P[X ≤ 3] = 0.75, P[X ≤ 4] = 1
By#�averaging(over�#Y,#we#have#the#CDF#for#X##(called#the#marginal(distribu.on)#
![Page 63: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/63.jpg)
Joint Continuous Distributions
If#X#and#Y#are#two#con/nuous#random#variables,#define#the#joint#probability#density#func/on#of#(X,Y): #!
∫ ∫
∫ ∫
=
=≤≤≤≤
∞×
X Y
dxdyyxf
dydxyxfdYcbXaP
fYX
b
a
d
c
YXYX
YYXX
R R
RR
RSRS
1),(
),(),(
),0[→:→: ,→:
),(
![Page 64: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/64.jpg)
Linear'Func+ons'of'Normal'Random'Variables'are's+ll'Normal'Random'Variables.'''This'is'not$true'with'other'distribu+ons;'if'X'~Exponen+al,''Y'~Exponen+al,'that'does'NOT'mean'X'+Y'~Exponen+al'(it�s'not!)'
Linear Functions of Normal RVs
• If'X'~'N(µx,'σx2),'then''aX'+'b'~'N(aµx+b,'a2σx
2)'
• If'Y'~'N(µy,'σy2)'is'independent'of'X,'then''
'X+Y'~'N(µx'+µy,'σx2'+σy
2)'
![Page 65: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/65.jpg)
Example: Investing We#have#$1000#to#invest.##
We#put#$Y#in#Mutual#Fund#I#and#the#rest#(1000:Y)#into#Mutual#Fund#II.##
The#returns'are#random#and#unknown,#characterized#with#the#normal#distribuDon:#
R1##~#N(x1,#0.0002x12)# R2#~#N(1.05x2,#0.0003x22)#
Invest $x1=y
Invest $x2= (1000-y)
![Page 66: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/66.jpg)
##E(T)#=#E(R1 + R2)#=#x1#+1.05x2##=#y#+1.05(1000#:#y)##=#1050#–#0.05y#
Investment example, cont.
What%is%the%expected%value%a0er%investment?%
What%is%the%variance%of%the%final%amount?%Var(T)#=#Var(R1)#+Var(R2)##=#0.0002x1+#0.0003#x2#=#0.0002y2#+0.0003(1000:y)2#
Let#T#be#the#amount#we#end#up#with,#then##T#=##R1#+#R2#
y#
y#
E(T)#
VAR(T)#
![Page 67: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/67.jpg)
If X1,….,Xn are independent and E(Xi)=µ, Var(Xi) = σ2, then if n is large
nXVXE
Xn
Xn
ii
21
)(,][
1
σµ ==
= ∑=
is approximately Normally distributed, with
Central Limit Theorem
-4 -2 2 4
0.1
0.2
0.3
0.4
We do NOT require that the Xis are normally distributed. This approximation usually works well if n ≥ 30
![Page 68: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/68.jpg)
Example - CLT
When a lithium batteries is prepared, the amount of a particular impurity in the battery is a random variable with mean value of 4.0 g and standard deviation 1.5 g. If 50 batteries are independently prepared, what is the (approximate) probability that the sample average amount of impurity is between 3.5 and 3.8 g?
The sample size n = 50 is large enough to use CLT to approximate the distribution of the average with a normal distribution. From CLT
We want to find the following probability
)045./,0.4( 2 ==≈ nNX σµ
1645.)36.2()94.(045.48.3
045.45.3
=−Φ−−Φ=#$
%&'
( −≤≤
−≈ ZP
)8.35.3( ≤≤ XP
![Page 69: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/69.jpg)
Introduction to R
• Youtube video: An Introduction to R – A Brief Tutorial for Rhttp://www.youtube.com/watch?v=LjuXiBjxryQ
• An Introduction to R:http://cran.r-project.org/doc/manuals/R-intro.html
• installation: http://cran.us.r-project.org/
• set directory: setwd(”/directory/data” )
• get help about a particular function: help(functionName)
• data < � read.table(file = ”Deflection.txt”)
Prof. Yao Xie, ISyE 2028, Basic Statistical Methods, Georgia Tech 6
![Page 70: Lecture 2 & 3. Basic Concepts and Review of Probabilityyxie77/isye2028/Lecture2_final.pdf• Basic concepts • Review of probability • Use of R Prof. Yao Xie, ISyE 2028, Basic Statistical](https://reader035.vdocuments.site/reader035/viewer/2022071411/61063d5b42518d0c0f7dc46a/html5/thumbnails/70.jpg)
• index = data[,1]
deflection = data[,2]
plot(deflection, index,type=’p’)
% type = ’p’: plot points; type = ’l’: plot a line
Prof. Yao Xie, ISyE 2028, Basic Statistical Methods, Georgia Tech 7