Sie sind auf Seite 1von 23

Analytic Hierarchy Process (AHP)

MCDM Technique developed by Thomas L Saaty

Nisha S. Simon
Need ?
 Standard Questionnaire
 Relative Comparison among elements
 Attempt complex hierarchical models
 Consistency Measure
 Related to
 Structural Equation Modeling – No Relative
Comparison
 Conjoint Analysis – Higher and complex
combinations are difficult
Process

 Input – Subjective Opinion, Actual Measurement (Price,


Weights); provide judgments about the relative importance of
each criterion and then specify a preference for each decision
alternative on each criterion
 Output – Prioritized ranking indicating the overall preference
for each of the decision alternatives, Consistency Index
 Process –
 Build hierarchical model (Crit, Sub Crit, Alt)
 Obtain pair wise comparison matrix (for each element)
 Obtain synthesized matrix
 Find priority vector and normalized ranking vector
 Calculate Eigen max and consistency index
 Check consistency ratio
Questionnaire
 Rank the relative importance of following factors of
investing in stock market
 Returns Investing in
Stock Market
 Risk
 Liquidity
Returns Risk Liquidity

 Build Hierarchical model

Relative Choice Criteria Scale


Return or Risk Return 3
Risk or Liquidity Risk 5
Return or Liquidity Return 7
 No of comparisons =n(n-1)/2

 Scale

Relative Choice Criteria Scale


Return or Risk Risk 3
Return or Liquidity Risk 5
Risk or Liquidity Return 7
Relative Choice Criteria Scale
Return or Risk Risk 3
Return or Liquidity Return 5
Risk or Liquidity Risk 7

Factor Factor Weighting Score Factor


More Importance than Equal Less Importance than

Return 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Risk
Return 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Liquidity
Risk 9 8 7 6 5 4 3 2 1 2 3 4 5 6 7 8 9 Liquidity

Return Risk Liquidity


Return 1 1/3 5
Risk 1 7
Liquidity 1
Complete matrix

aij = 1/aji
Pairwise Comparison Matrix and Synthesized Matrix
 Pairwise Comparison Matrix
Return Risk Liquidity
Return 1 .33 5
Risk 3 1 7
Liquidity .20 .14 1
Total 4.2 1.48 13 Normalized with 1 as
the most imp factor
 Synthesized Matrix
Return Risk Liquidity Total Priority Ranking
Vector Vector
Return 1/4.2 .33/1.48 5/13 .85 .28 .44
Risk 3/4.2 1/1.48 7/13 1.93 .64 1
Liquidity .2/4.2 .14/1.48 1/13 .22 .07 .11
Total 1 1 1 1
Consistency Index and Ratio

λ
max

1 0.33 5 0.28 0.21 0.35 0.84 .84/.28 3.00


0.28 3 + 0.64 1 + 0.07 7 = 0.84 + 0.64 + 0.49 = 1.97 1.97/.64 3.08
0.2 0.14 1 0.06 0.09 0.07 0.22 .22/.07 3.08
Total 9.16
PV 9.16/3 3.05
Consistency Index and Ratio
 Consistency Index = (λ max – n)/(n-1)
= (3.1 -3)/2 = 0.0484
Consistency is closely related to transitive property
B>A; A>C = B>C
Consistency Index is compared with random
consistency index developed by Saaty.
n 1 2 3 4 5 6 7 8 9 10
R 0 0 .58 .9 1.12 1.24 1.32 1.41 1.45 1.49
I

CR = CI/RI = 0.0484/0.58 = 0.083


Inference
 Risk is considered to be the most important
criterion followed by return and liquidity when
investing in a stock market
 Since consistency ratio is smaller than10%,
subjective evaluation of preference for stock
market investment is consistent.
Complex Models

Investment Choice

Other
Fin Tech Env
s
2 level Model
 Pair wise matrix, synthesized matrix, priority and ranking vectors for
criteria – Others
Pairwise Comparison Matrix Synthesised Matrix Priority Ranking
O1 O2 O3 O4 O5 O1 O2 O3 O4 O5 Total Vector Vector
O1 1.00 1.60 1.00 1.90 0.65 0.21 0.21 0.21 0.21 0.21 1.07 0.21 0.65
O2 0.63 1.00 0.63 1.19 0.41 0.13 0.13 0.13 0.13 0.13 0.67 0.13 0.41
O3 1.00 1.60 1.00 1.90 0.65 0.21 0.21 0.21 0.21 0.21 1.07 0.21 0.65
O4 0.53 0.84 0.53 1.00 0.34 0.11 0.11 0.11 0.11 0.11 0.56 0.11 0.34
O5 1.54 2.46 1.54 2.92 1.00 0.33 0.33 0.33 0.33 0.33 1.64 0.33 1.00
4.69 7.50 4.69 8.91 3.05 1.00 1.00 1.00 1.00 1.00 5.00 1.00

 Construct matrices and compute vectors for all criteria – fin, tech and
env.
Pairwise Comparison Matrix Synthesised Matrix Priority Ranking
F T E O O1 O2 O3 O4 Total Vector Vector
F 1.00 0.65 0.83 1.07 0.21 0.21 0.21 0.21 0.85 0.21 0.64
T 1.54 1.00 1.28 1.65 0.33 0.33 0.33 0.33 1.32 0.33 1
E 1.20 0.78 1.00 1.29 0.26 0.26 0.26 0.26 1.03 0.26 0.79
O 0.93 0.61 0.78 1.00 0.20 0.20 0.20 0.20 0.80 0.20 0.61
4.67 3.04 3.89 5.01 1.00 1.00 1.00 1.00 4.00 1.00
Contd.
 Global Priorities
 GP of O1 = PV of O1 * PV of Others = .21 *.20 = .42
Factors Relative Imp Rank
F1 .02 14
F2 .96 1
F3 .10 13
F4 .23 11
T1 .25 9
T2 .58 4
E1 .13 12
E2 .78 3
E3 .85 2
O1 .42 7
O2 .36 8
O3 .56 5
O4 .45 6
O5 .25 9
Another Model (with alternatives)

Select Best Smart Phone

Display
Batter Internal
Cost Resolut Storage
ion y Life
Pairwise Comparison Matrix Synthesised Matrix
Cost S1 S2 S3 S1 S2 S3 Total Priority Vector
S1 1.00 0.33 1.00 0.20 0.21 0.17 0.58 0.19
S2 3.00 1.00 4.00 0.60 0.63 0.67 1.90 0.63
S3 1.00 0.25 1.00 0.20 0.16 0.17 0.52 0.17
5.00 1.58 6.00 1.00 1.00 1.00 3.00 1.00
Pairwise Comparison Matrix Synthesised Matrix
Display S1 S2 S3 S1 S2 S3 Total Priority Vector
S1 1 2 0.5 0.29 0.33 0.27 0.89 0.30
S2 0.5 1 0.33 0.14 0.17 0.18 0.49 0.16
S3 2 3 1 0.57 0.50 0.55 1.62 0.54
3.5 6 1.83 1.00 1.00 1.00 3.00 1.00

Pairwise Comparison Matrix Synthesised Matrix


Battery S1 S2 S3 S1 S2 S3 Total Priority Vector
S1 1 3 2 0.55 0.43 0.60 1.58 0.53
S2 0.33 1 0.33 0.18 0.14 0.10 0.42 0.14
S3 0.5 3 1 0.27 0.43 0.30 1.00 0.33
1.83 7 3.33 1.00 1.00 1.00 3.00 1

Pairwise Comparison Matrix Synthesised Matrix


Storage S1 S2 S3 S1 S2 S3 Total Priority Vector
S1 1 1 0.25 0.17 0.17 0.17 0.50 0.17
S2 1 1 0.25 0.17 0.17 0.17 0.50 0.17
S3 4 4 1 0.67 0.67 0.67 2.00 0.67
6 6 1.5 1 1 1 3 1
Pairwise Comparison Matrix Synthesised Matrix
C D B S C D B S Total Priority Vector
C 1 0.33 0.25 0.5 0.10 0.17 0.05 0.06 0.39 0.10
D 3 1 3 4 0.30 0.52 0.66 0.47 1.95 0.49
B 4 0.33 1 3 0.40 0.17 0.22 0.35 1.14 0.29
S 2 0.25 0.33 1 0.20 0.13 0.07 0.12 0.52 0.13
10 1.91 4.58 8.5 1.00 1.00 1.00 1.00 4.00 1.00

Global Priorities
S1 Cost .19*.10 0.02
S2 Cost .63*.10 0.06 S1 0.34
S3 Cost .17*.10 0.02 S2 0.20
S1 Display .30*.49 0.14 S3 0.46
S2 Display .16*.49 0.08
S3 Display .54*.49 0.26 C C B BC Ratio
S1 Battery .53*.29 0.15 S1 549 0.35 0.34 0.95
S2 Battery .14*.29 0.04 S2 450 0.29 0.20 0.71
S3 Battery .33*.29 0.10 S3 550 0.36 0.46 1.30
S1 Storage .17*.13 0.02 1549
S2 Storage .17*.13 0.02
S3 Storage .67*.13 0.09
Completed Model

Select Best Smart Phone 1

Display Batter Internal


Cost Resoluti y Life Storage
= .10 on = .49 = .29 = .13
Aggregation
 Aggregateindividual judgments to form an
average pairwise comparison matrix using
geometric mean
Definition
 AHP makes judgments on pairs of elements with
respect to a controlling element to derive ratio
scales that are then synthesized throughout the
structure to select the best alternative.
Softwares

 Excel Templates
 Superdecisions
 Expert Choice
 Priest
 MakeItRational
What else?

 RII(relative Imp Index)


 Fuzzy AHP
 Agreement Analysis