Upload
fadhl-alakwaa
View
227
Download
0
Embed Size (px)
Citation preview
8/8/2019 Probability Lecture ONE
1/49
Probability & Statistics forBME
(Third Year)2009-2010
Assistant Prof \
Fadhl M. Alakwaa
8/8/2019 Probability Lecture ONE
2/49
This course provides the foundational skills for
advanced statistical analysis of biological signals.The
basic analytical concepts of probability theory,
statistical design of experiments and data analysis andrepresentation of biological variables as random
processes are demonstrated and practiced through
computer based analysis of biologically relevant data
sets provided throughout the course. All computationalhomework assignments are carried out using
MATLAB.
Course Description
8/8/2019 Probability Lecture ONE
3/49
Solve simple probability problems with electrical and computer
engineering applications using the basic axioms of probability.
Describe the fundamental properties of probability density
functions with applications to single and multivariate randomvariables.
Describe the functional characteristics of probability density
functions frequently encountered in life science research such as
the Binomial, Uniform, Gaussian and Poisson.
Determine the first through fourth moments of any probabilitydensity function using the moment generating function.
Calculate confidence intervals and levels of statistical
significance using fundamental measures of expectation and
variance for a given numerical data set.
Course Objective
Upon completion of the course, students should be able to
8/8/2019 Probability Lecture ONE
4/49
Discern between random variables and random processes for
given mathematical functions and numerical data sets.
Determine the power spectral density of a random process for
given mathematical functions and biological data sets.
Determine whether a random process is ergodic or nonergodicand demonstrate an ability to quantify the level of correlation
between sets of random processes for given mathematical
functions and biological data sets.
Model complex families of signals by means of random
processes. Determine the random process model for the output of a linear
system when the system and input random process models are
known.
Course Objective
Upon completion of the course, students should be able to
8/8/2019 Probability Lecture ONE
5/49
1. Intuitive Probability and Random Processes using MATLAB
by Steven Kay (2006), Springer. ISBN: 0-387-24157-4.
2. Applied Statistics and Probability for Engineers by
Montgomery, Runger 4th Edition
3. Probability and Statistics for Engineering and the Sciences
by Jay L. Devore 6th Edition 2004 ISBN: 0534399339
References
8/8/2019 Probability Lecture ONE
6/49
Essential Book
8/8/2019 Probability Lecture ONE
7/49
Recommended Book
8/8/2019 Probability Lecture ONE
8/49
CourseL
abMatlab
Excel
Minitab
labview
8/8/2019 Probability Lecture ONE
9/49
8/8/2019 Probability Lecture ONE
10/49
Course Evaluation
8/8/2019 Probability Lecture ONE
11/49
Course Projects: Project1
8/8/2019 Probability Lecture ONE
12/49
Course Projects: Project2
Understand Manual service for medical
instrument
8/8/2019 Probability Lecture ONE
13/49
Course Presentation
Bayes theorem
Probability mass function PMF
Binomial distribution
Hypergometric distribution
Poisson distribution
Normal distribution
Joint probability distribution
Marginal probability distribution
Conditional probability distributionCovariance and correlation
Box plots
Time sequence plots
Histograms
Central limit thermo
Point estimation of parameters
Maximum likelihood
Confidence interval
t distribution
Hypothesis testing
P-value
Type I and type II error
Statistical inference
F distribution
Least square estimate
Linear regression
Design engineering experiment
ANOVA
Cumulative distribution functions
Means and variance of discrete variable
Means and variance of continuous variable
Choose one topics from the below and do a power point
presentation:
8/8/2019 Probability Lecture ONE
14/49
Important comments from the
previous course Not Excuses
Not degree explanation (fair assessment)
In time policy (one day late=one degree
loss)
Join a group (mandatory)
Update your attendance and results daily.
8/8/2019 Probability Lecture ONE
15/49
Group Activity: BME_UST
http://www.facebook.com/search/?q=BME_UST&init=quick#!/grou
p.php?gid=325135515239&ref=search&sid=1096082202.17723631
20..1
8/8/2019 Probability Lecture ONE
16/49
8/8/2019 Probability Lecture ONE
17/49
8/8/2019 Probability Lecture ONE
18/49
An engineeris someone who solves problems of
interest to society by the efficient application ofscientific principles by
Refining existing products
Designing new products or processes
1-1 The Engineering Method and
Statistical Thinking
8/8/2019 Probability Lecture ONE
19/49
1-1 The Engineering Method and
Statistical Thinking
Figure 1.1 The engineering method
1- Develop a clear and concise description of the problem
2- Identify, the important factors that affect this problem3- Propose a model for the problem using scientific or
engineering knowledge of the phenomenon being studied. State
any limitations or assumptions of the model
8/8/2019 Probability Lecture ONE
20/49
1-1 The Engineering Method and
Statistical Thinking..continue
4- Conduct appropriate experiments and collect data to test or
validate the model.
5- Refine the model on the basis of the observed data.
6- Manipulate the model to assist in developing solution to the
problem.7- Conduct an appropriate experiment to confirm that the
proposed solution to the problem is both effective and efficient.
8- Draw conclusions or make recommendation based on the
problem solution.
8/8/2019 Probability Lecture ONE
21/49
The field ofstatistics deals with the collection,
presentation, analysis, and use of data to
Make decisions
Solve problems
Design products and processes
1-1 The Engineering Method and
Statistical Thinking
8/8/2019 Probability Lecture ONE
22/49
Statistical techniques are useful for describing and
understanding variability.
By variability, we mean successive observations of asystem or phenomenon do notproduce exactly the same
result.
Statistics gives us a framework for describing thisvariability and for learning about potential sources of
variability.
1-1 The Engineering Method and
Statistical Thinking
8/8/2019 Probability Lecture ONE
23/49
Engineering Example
An engineer is designing a nylon connector to be used in an
automotive engine application. The engineer is consideringestablishing the design specification on wall thickness at 3/32
inch but is somewhat uncertain about the effect of this decision
on the connector pull-off force. If the pull-off force is too low, the
connector may fail when it is installed in an engine. Eightprototype units are produced and their pull-off forces measured
(in pounds): 12.6, 12.9, 13.4, 12.3, 13.6, 13.5, 12.6, 13.1.
1-1 The Engineering Method and
Statistical Thinking
8/8/2019 Probability Lecture ONE
24/49
Engineering Example
The dot diagram is a very useful plot for displaying a small
body of data - say up to about 20 observations.
T
his plot allows us to see easily two features of the data; thelocation, or the middle, and the scatterorvariability.
1-1 The Engineering Method and
Statistical Thinking
3/32 inch
8/8/2019 Probability Lecture ONE
25/49
Engineering Example
The engineer considers an alternate design and eight prototypes
are built and pull-off force measured.
T
he dot diagram can be used to compare two sets of data
Figure 1-3 Dot diagram of pull-off force for two
wall thicknesses.
1-1 The Engineering Method and
Statistical Thinking
average13.0
13.4
8/8/2019 Probability Lecture ONE
26/49
How do you know that another sample of prototypes will not
give different results?
If we use the test results obtained so far to conclude that
increasing the wall thickness increases the strength, what risks
are associated with this decision?
For example, is it possible that the apparent increase in
pull-off force observed in the thicker prototypes is only
due to the inherent variability in the system and that
increasing the thickness of the part (cost) really has no
effect on the pull-off force?
1-1 The Engineering Method and
Statistical Thinking
8/8/2019 Probability Lecture ONE
27/49
Engineering Example
Since pull-off force varies or exhibits variability, it is a
random variable.
A random variable, X, can be model by
X = Q + I
where Q is a constant and I a random disturbance.
Due to environment, test equipment, different in
individual parts themselves.
1-1 The Engineering Method and
Statistical Thinking
8/8/2019 Probability Lecture ONE
28/49
1-1 The Engineering Method and
Statistical Thinking
8 connectors
prototypes
Connectors that
will be sold to
customers
Sampling error, sample size
8/8/2019 Probability Lecture ONE
29/49
Three basic methods for collecting data:
A retrospective study using
historical data
An observational study
A designed experiment
1-2 Collecting Engineering Data
8/8/2019 Probability Lecture ONE
30/49
1-2.2 Retrospective Study
The retrospective study may involve a lot of data, but
that data may contain relatively little useful information
about the problem.
Some of the relevant data may be missing
There may be recording errors resulting in outliers
(unusual values)
Other important factors may not have been collected
1-2.3 Observational Study
8/8/2019 Probability Lecture ONE
31/49
1-2.4 Designed Experiments
8/8/2019 Probability Lecture ONE
32/49
1-2.4 Designed Experiments
Figure 1-5 The factorial design for the distillation column
8/8/2019 Probability Lecture ONE
33/49
1-2.4 Designed Experiments
Figure 1-6 A four-factorial experiment for the distillation column
8/8/2019 Probability Lecture ONE
34/49
1-2.5 Observing Processes Over Time
Whenever data are collected over time it is important to plot
the data over time. Phenomena that might affect the system
or process often become more visible in a time-oriented plot
and the concept of stability can be better judged.
Figure 1-8 The dot diagram illustrates variation but doesnot identify the problem.
8/8/2019 Probability Lecture ONE
35/49
1-2.5 Observing Processes Over Time
Figure 1-9 A time series plot of concentration provides
more information than a dot diagram.
8/8/2019 Probability Lecture ONE
36/49
1-2.5 Observing Processes Over Time
Figure 1-10 Demings funnel experiment.
8/8/2019 Probability Lecture ONE
37/49
1-2.5 Observing Processes Over Time
Figure 1-11 Adjustments applied to random disturbancesover control the process and increase the deviations from
the target.
8/8/2019 Probability Lecture ONE
38/49
1-2.5 Observing Processes Over Time
Figure 1-12 Process mean shift is detected at observationnumber 57, and one adjustment (a decrease of two units)
reduces the deviations from target.
8/8/2019 Probability Lecture ONE
39/49
1-2.6 Observing Processes Over Time
Figure 1-13 A control chart for the chemical process
concentration data.
8/8/2019 Probability Lecture ONE
40/49
1-2.6 Observing Processes Over Time
Figure 1-14 Enumerative versus analytic study.
8/8/2019 Probability Lecture ONE
41/49
1-3 Mechanistic and Empirical Models
A mechanistic model is built from our underlying
knowledge of the basic physical mechanism that relates
several variables.
Example: Ohms Law
Current = voltage/resistance
I=
E/R
I=E/R + I
8/8/2019 Probability Lecture ONE
42/49
1-3 Mechanistic and Empirical Models
An empirical model is built from our engineering and
scientific knowledge of the phenomenon, but is not
directly developed from our theoretical or first-
principles understanding of the underlying mechanism.
8/8/2019 Probability Lecture ONE
43/49
1-3 Mechanistic and Empirical Models
Example
Suppose we are interested in the number averagemolecular weight (Mn) of a polymer. Now we know that Mn
is related to the viscosity of the material (V), and it alsodepends on the amount of catalyst (C) and the temperature(T) in the polymerization reactor when the material ismanufactured. The relationship between Mn and thesevariables is
Mn = f(V,C,T)say, where the form of the function fis unknown.
where the Fs are unknown parameters.
8/8/2019 Probability Lecture ONE
44/49
8/8/2019 Probability Lecture ONE
45/49
1-3 Mechanistic and Empirical Models
In general, this type of empirical model is called a
regression model.
The estimated regression line is given by
8/8/2019 Probability Lecture ONE
46/49
Figure 1-15 Three-dimensional plot of the wire and pullstrength data.
8/8/2019 Probability Lecture ONE
47/49
Figure 1-16 Plot of the predicted values of pull strengthfrom the empirical model.
8/8/2019 Probability Lecture ONE
48/49
1-4 Probability and Probability Models
Probability models help quantify the
risks involved in statistical inference, that
is, risks involved in decisions made everyday.
Probability provides the frameworkfor
the study and application of statistics.
8/8/2019 Probability Lecture ONE
49/49