Statistik dan Probabilitas Unlambjb..ppt

menjelaskan tentang teori statistik probabilitas

1. Statistik dan Probabilitas
Nur Iriawan, PhD.

2. Research dan Problem
Research Problem

3. What is a problem ?
IDEA
THEORY
MODELGAP

4. Fakta/data/praktek
Conjecture/hypothesis/theory/model
Design
experiment
Problem
formulation
Testing

5. Data Metode analisis
•Pembedaan atas skala data
–Kwalitatif
–Kwantitatif
•Pembedaan atas kepastian perubahan
–Probabilistik
–Deterministik
•Pembedaan atas waktu kemunculan datanya
–Diskrit
–Kontinyu
•Tinjauan pengambilan keputusan
–Valid
–Reliable
–Konsisten

6. Valid vs Reliable vs Konsisten
reliable
konsisten
Tidak valid

7. Data Kwalitatif
•Skala data
–Nominal
–Ordinal
•Berikan contoh data yang berskala seperti
tersebut di atas!

8. Data Kwantitatif
•Skala data
–Interval
–Ratio
•Berikan contoh data yang berskala seperti
tersebut di atas!
•Bagaimana dengan
–Suhu
–Ph
–Kadar gula
–Berat besi
–korelasi

9. Perbandingan kwalitatif dan
kwantitatif
•Data mana yang mempunyai bayangan metode
analisis lebih mudah?
kwantitatif
•Dapatkah data kwalitatif dikwantitatifkan?
•Metode apa yang biasa digunakan dalam
menganalisis data:
–Kwantitatif?
–Kwalitatif?
•Bagaimana jika metode kwantitatif digunakan
dalam kasus data kwalitatif?

10. Probabilistik vs deterministik
•Kasus dunia nyata (environmental
problem) lebih banyak akan bersifat
probabilistik, multidimensi, dan kompleks
probabilistik deterministik

11. Agriculture
Age
Industrial
Age
Information
Age
Wealth
definition
Food
Food &
Things
Knowledge
People work
as
Slaves
/serfs
Employees Partners
People work
in Hierarchies
Bureaucracies
Teamnets
Production
system
One-piece
Customization
Mass
Production
Paradigm
Mass
customization
Paradigm
Scarcity of resource Abundance of information
Bio

12. 1960 1970 1980 1990 2000
year

13. Education and self
development
Training and
experimentation
Facilitation and on
the job training
Coaching and
mentoring
Learning Process
competence
Culture

14. Core
Capabilities
Implementing
and
Integrating
Experimenting
Importing
Knowledge
Problem
Solving
EXTERNAL
PRESENT
INTERNAL
FUTURE

15. Metode analisis data deskriptif
•Titik tengah
–Mean
–Median
–Modus
–Tream mean
•Penyimpangan
–Variance
–Range
Interval
konfidensi
Skewness?

16. Interval (konvensional) vs HPD
Interval (Konvensional)
HPD
(Highest Probability Distribution)

17. Interval Kepercayaan (1)

18. Interval kepercayaan (2)

19. HPD (Highest Probability Distribution)Peta Kendali
(1-)
x
100%
Batas
Kendali
Bawah
Batas
Kendali
Atas




95,0






71,3953




109,4810



97,5






64,4857



110,9149




99,0




55,3356



112,7754

20. Analytical Models
•Qualitative Approach (for qualitative data)
–Independence Analysis
–Proportion Analysis
–Analytical Hierarchical Process (AHP)
•Quantitative Approach (for quantitative data)
–Forecasting (Smoothing, AR, MA, ARIMA)
–Clustering Analysis
–Regression Analysis
–ANOVA
–MRP (Material Requirement Planning)
•Simulation

21. MIS Versus DSSMIS DSS
 Decision support
provided
Provide information
about the performance
of the organization
Provide information
and decision support
techniques to analyze
specific problems or
opportunities
 Information form and
frequency
Periodic, exception,
demand, and push
reports and responses
Interactive inquiries
and responses
 Information formatPrespecified, fixed
format
Ad hoc, flexible, and
adaptable format
 Information
processing
methodology
Information produced
by extraction and
manipulation of
business data
Information produced
by analytical modeling
of business data

22. Basic Challenges
Building information systems that can actually fulfill
management (decision maker) information requirements.
•Internal versus external data.
•Structured versus unstructured data.
•Anticipated versus unanticipated information.

23. Challenges
Integrating DSS and ESS with existing systems:
•Consistency of data
•Availability of data
•Timeliness of data

24. Issues
•Models validity
–Uncertainty
–Time and space scales
–Establishing a coherent dialogue between
models
•Design of a policy assessment process
–Define objectives and constraints
–Identify the possible controls
–Identify the system reaction to controls
–Identify the cost/benefits for control policies
–Find efficient solutions

25. IT practices
Information
Management practices
Information behavior
And values
Information
orientation :
A comprehensive
of high level idea
of how effective
a company is in
using information
Business
Performance

26. Types of Decisions
Structured: Repetitive; definite procedure; have a high level
of certainty; could even be called routine.
Semi-structured: One or more factors are not structured;
the more unstructured, the higher the risk.
Unstructured: Unique; non-routine; has definite
uncertainty; requires experience and judgment.

27. Unstructured Structured
•Ad hoc
•Unscheduled
•Summarized
•Infrequent
•Forward looking
•External
•Wide Scope
•Prespecified
•Scheduled
•Detailed
•Frequent
•Historical
•Internal
•Narrow Focus

28. Ways to Solve Business
Problems
Absolution: Ignore it and hope it will go away.
Dissolution: Redesign to eliminate the problem.
Ressolution: Do something that yields an outcome that is good
enoughemphasizing past experience.
Solution: Involves research and relies heavily on
experimentation, quantitative analysis and both
common and uncommon sense.

29. Speed
To marketcompetitive, market
position, market leadership
To decision consensus, commitment,
responsive
To task completion productivity

30. How Are Decisions Made?
Big Deal:
•Research and get as much data as possible.
•Evaluate as long as time permits.
•Consensus decision by key participants.
Little Deal:
•Routine based on past experience (habit).
•Quick decision since consequences are minimal.
•Individual versus consensus decision.
Depends on the significance of the decision.

31. Solution Selection Criteria
1. Risk: including the odds.
2. Economy of effort: greatest results with least
effort or needed change with least disturbance.
3. Timing: based on urgency which is difficult to
systematize.
4. Limitation of Resources: relative to those
that must carry out the decision. No decision is
better than the people that must carry it out.

32. Parameter vs Statistik
Populasi
Sampel
Parameter
Statistik
Uji kesamaan

33. Konsep Variabel OBJEK
•Variabel: suatu peubah yang dibuat untuk dapat
diisi dengan suatu nilai
•Nilai sebuah variabel merupakan identitas
demensi pandang pada suatu objek
•Objek dapat diterangkan dengan jelas oleh
suatu susunan variabel yang lebih banyak
•Atau semakin sedikit variabel dari obej yang
diamati semakin tidak jealas objek yang
dimaksudkan.

34. Univariate vs Multivariate
•Univariate merupakan istilah analisis data statistik yang
hanya memandang permasalahan hanya dengan
demensi terbatas, yaitu satu demensi.
•Multivariate mengandalkan cara pandang permasalahan
statistik dengan multi demensi, dan setiap demensi
diduga akan berkorelasi. (jika tiap dimensi tidak
berkorelasi, maka akan hanya digolongkan sebagai
permasalahan yang multivariabel saja).
•Jadi suatu permasalahan yang dipandang secara
multivariabel belum tentu merupakan kasus multivariate,
sedangkan kasus multivariate selalu mdipandang secara
multivariabel.

35. Konsep Derajat bebas
•Derajat bebas (db) memberikan informasi
tentang tingkat kebebasan suatu data akan
dapat terambil secara random dari kelompoknya
•Semakin besar db maka akan semakin besar
kepercayaan bahwa data yang dapat diambil
adalah dapat mewakili populasinya
•Semakin besar db semakin besar tingkat
kepercayaan dalam pengambilan keputusannya,
karena keputusannya didasarkan pada prinsip
STRONG LAW LARGE NUMBER THEOREM.
•Semakin besar db berarti semakin bervariasinya
suatu sisi pandang demensi analisisnya.