(TKG-233)
KULIAH PERTEMUAN KE-3
Teknik Geologi
2014
PELUANG
• If an event can happen in s ways (success) and fail to happen in f
ways (fail), and if each of these s+f ways is EQUALLY LIKELY
TO OCCUR, (mempunyai kesempatan muncul / terjadi yang
sama) the probability of success (the event happening, peristiwa
yang diharapkan, terjadi) in a single trial is
s
p
s f
And the probability of failure is
f
q
s f
In a toss of coin, a head can occur only in s=1 way and a tail in f=1
way, and both way are equally likely to occur; thus
1 1
p
11 2
s f s f
pq 1 ( 100%)
s f s f s f
Conditional Probability
• The probability of the occurrence of an event B, given that an
event A has occurred, is called the conditional probability of the
occurrence of the event B . P (B|A)
• The general formula:
P(BA )
P(B | A)
P(A)
Copper A B C
Low (L) 30 19 8
Medium (M) 10 20 17
High (H) 15 18 13
Question
1. Is it same, P(H|A) and P(A|H) ?
Bayes's Theorem
• ByThomas Bayes
• Conditional probability P(BA )
P(B | A)
P(A)
• Multiplication rule P(BA) = P(A|B) P(B) so
P( Bi ) P( A | Bi )
P( Bi | A)
P(B) = prior probability of B
P( A)
P(B|A) = posterior probability of B
P(A|B) = likelihood of A
Where P(A) = P(A|B)P(B) + .... + P(A|Bk)P(Bk)
Bayes's Theorem
• Measurements are made on a fossil which is known to be a
specimen of one of three species. The following table gives the
probabilities of obtaining such a set of characteristics in each
species Species P(characteristic A | species Bi)
B1 0.8
B2 0.3
B3 0.6
• The probabilities of obtaining a specimen of each species are
Species Probability of occurance
B1 0.1
B2 0.4
B3 0.5
120
100
80
60
40
20
0 1 2 3 4 5 6 7 8 9 X
Histogram for expected (binomial) frequencies in 512 tosses of 9 coins
Binomial Distribution
The value of Xi fi , Xi2 fi and their sums, needed in this calculation are shown in
the next slide table, then:
h
X f
i i
2304
m i 1
4.50
N 512
And the standard deviation:
h
( X i f i ) 2
h
2304
2
X fi i 1
2
i 11520
N 512 1152
i 1
N 512 512
2.25 1.50
Binomial Distribution
Frequency table for calculation of mean and standard deviation of
binomial distribution ( for the example)
Xi fi Xi fi Xi 2 fi
0 1 0 0
1 9 9 9
2 36 72 144
3 84 252 736
4 126 504 2016
5 126 630 3130
6 84 504 3024
7 36 252 1764
8 9 72 576
9 1 9 81
512 2304 11520
Binomial Distribution
m = n.p
σ = √npq
16
16
8
12 10 10
2
where :
m mean of the binomial distribution np
standar deviation of the binomial distribution npq,
e base of the natural logarithms 2.71828...,
3.14159.....,
Y frequency with which any X occurs .
The relative frequency (or probability) denoted by P, then P = Y/N , and we can written :
1 ( X m ) 2 / 2 2 ,
P e
2
the graph of which is called the normal probability curve.
Y (or f )
80
60
40
20
X
0 1 2 3 4 5 6 7 8 9
NORMAL DISTRIBUTION
If we multiply both sides of this equation by σ and let y = Pσ thereby
making σ the unit of measurement on the vertical axis, we can rewrite this
equation as:
1 ( X m )2 / 2 2 ,
P e
2
0.2420
P 0.1613
1.5
P
0.30
0.25
0.20
0.15
0.10
0.05
JAWAB
Untuk mendapatkan jumlah total 9, dari komposisi 6+3, 3+6, 5+4, 4+5 tampak ada
4 kemungkinan sehingga
(a) = 4/36 = 1/9
(b) = 1-1/9 = 8/9
PR
1. Sebanyak 50 sayatan tipis batuan dimana tiga darinya mengandung
mineral apatit. Kita mengambil secara random (acak) berapa peluang
bahwa yang diambil adalah batuan yang mengandung mineral apatit