Topic 161

click here to download ppt

Overview

• Here, we will understand the recommendations for the model authors.

Recommendations for the model authors

• Research paper should be well specified.

• Complete source code for the model may not sufficient.
• Model authors must make their source code publicly available or at least the
“pseudo code”.
• To what extent the model developer presents a sensitivity analysis.
• Small modification to the original model can effect results drastically.
• These sensitive differences should be published by the authors.
• Original author may not be the original implementer.
• Implementation will not be veridical.
• Model authors implement their own models using “ low-threshold ” and toolkits.
• Authors to examine their conceptual models through the lens of a potential model
replicator.

Summary

• We have understood the recommendations for model authors in modelling.

• Credits: Uri Wilensky book

162

click here to download ppt

Overview

• Here, we will understand the basics about complex adaptive systems (CAS).

Complex Adaptive Systems

• Complexity
• Adaptation
• Systems
• The systems approach is an old concept.
• The approach stands on the assumption that breaking down of a complex concept
into simple easy to understand units helps in better understanding of the complexity.
• Ludwig von Bertalanffy first proposed it as 'General System Theory'
• A system is a delineated part of the universe which is distinguished from the rest by
an imaginary boundary. (NECSI)
• Properties of the system, the properties of the universe excluding the system which
affect the system and the interactions / relationships.
• An adaptive system (or a complex adaptive system, CAS) is a system that changes its
behavior in response to its environment.
• The adaptive change that occurs is often relevant to achieving a goal or objective.
(NECSI)
• Complexity is:
• ...the (minimal) length of a description of the system.
• ...the (minimal) amount of time it takes to create the system.
• Here the length of a description is measured in units of information.

Examples
 Summary

• We have learnt about the basic ideas of Complex Adaptive Systems.

• Credits: Uri Wilensky Book.

163

click here to downlod ppt

Overview

• Here, we will understand about the historical perspective of CAS

History

• Started with the concepts of General Systems Theory

• Biologists focused primarily on individual cells
• Social Scientists focused on connections
 • John Holland and cas/CAS

Summary

• We have learnt about the historical perspective of CAS.

• Credits: Niazi & Hussain book.

164

click here to download ppt

Overview

• Here, we will understand about the basic ideas of complexity

Complexity

• Disambiguation from Computational Complexity

• Nonlinearity
• Large number of variables
• Chaos theory
• The term "chaos" is popularly used to refer to disorder or confusion.
• In science, chaos is an important conceptual paradox that has a precise
mathematical meaning:
• A chaotic system is a deterministic system that is difficult to predict.
• A deterministic system is defined as one whose state at one time completely
determines its state for all future times.
• So what does it mean for a chaotic system to be difficult to predict?
• Butterfly effect

Summary

• We have learnt about the basics of Complexity in CAS.

• Credits: Niazi and Hussain book.

165

click here to download ppt

Overview

• Here, we will understand about adaptation in CAS.

Adaptation

• Adaptive changes in interactions

• Who changes
• What changes
• When changes
• How changes

Summary

• We have learnt about the adaptation in CAS.

• Credits: Niazi and Hussain book.

166

Overview

• Here, we will understand about the Systems approach.

Systems Approach

• Concept of a System - boundaries

• Many systems can be correlated
• Example: Life/Humans/Society
• Whole vs. parts
• Reductionist vs. Wholistic approach

Summary

• We have learnt about the system

167

click here to download ppt

Overview

• Here, we will understand about the problems encountered in modeling of CAS.

Modeling of CAS

• Cognitive Agent-based (CABC) Framework

Summary

• We have learnt about the modeling of CAS using CABC framework

• Credits: Niazi and Hussain book.

168

click here to download ppt

Overview

• Here, we will understand about the agent-based approach to modeling

Agent-based Approach

• Agent – something that acts

• There is a concept of agent in domains as distinct as:
o AI/CS/Robotics
o Sociology (Individual)
o Biology
o Ecology
o Etc.
• CABC framework has 3 levels for Agent-based modeling
• Exploratory Agent-based modeling (EABM)
• Descriptive Agent-based Modeling (DREAM)
• Validated Agent-based Modeling (VOMAS-based Modeling or VABM)
• Exploratory ABM
• DREAM
• Virtual Overlay Multiagent System (VOMAS)
• Validated Agent-based Modeling

Summary

• We have learnt about various approaches for modeling using Agents and the CABC
framework.
• Credits: Niazi and Hussain book.

169

click here to download ppt

Overview
 • Here, we will understand about complex networks/graphs and modeling using the
CABC Level 1

Complex Network-based Modeling

• What is a graph?
• Graph/Network
• Same entity different terms in different domains
• Network = weighted graph
• Different aspects of modeling using networks
• Labels
• arcs/lines
• Directed/Undirected
• Centrality-based
• Ego
• Sentiment graphs
• Community Detection
• Key benefits
• Very well-defined
• Lot of free tools
• Lot of community support
• Well-defined mathematical models
• Cons
• Difficult to figure out
• Difficulty in graph mining – lack of expertise in the community
• Missing data can be very difficult
• Requires complete data
• Complexity in deciding data boundaries

Summary

• We have learnt about complex-network modeling using the CABC framework

• Credits: Niazi and Hussain book.

170

click here to download ppt

Overview

• Here, we will understand how to put modeling and simulation to use employing the
CABC framework.

Cognitive Agent-based Computing Framework

 • Data and Models
• How much data is needed
• Why choose Agent-based Modeling?
• Why choose Complex network modeling?
• When to choose EABM?
• When to choose DREAM?
• When to choose VOMAS?

Summary

• We have learnt about putting the CABC framework to work.

• Credits: Niazi and Hussain book.

171

click here to download ppt

Overview

• Here, we will learn about the useful statistical models.

Useful Statistical Model

• The models that are useful in the case of limited data are:
• Queueing systems.
• Inventory and supply-chain systems.
• Reliability and maintainability.
• Limited data.
• Other distributions.

Queueing Systems

• Simulation solved the waiting line problems.

• Interval and service time patters were given in queuing examples.
• In these examples service and arrival time was probabilistic.
• But it is possible to have constant arrival time and constant service time.
• “Arrivals” can occur due to many reasons like machine break downs.
• Exponential distribution used to simulate the random exponential distribution of
services.
• For special cases truncated normal distribution can be utilized.

Inventory and supply-chain systems

• Three random variables

 • Number of units demanded per order or per time period.
• Time between demands.
• Lead time.
• Simple inventory systems have constant demand and lead time constant or zero.
• Mathematical tractability based demand and lead time in inventory theory text
could be invalid.
• Variety of demand patterns are satisfied by geometric, Poisson and negative
binomial distributions.

Reliability and Maintainability

• Failure time is modeled with many distributions.

• When failure occurs randomly model distribution is considered as exponential.
• Modeling standby redundancy required for gamma distribution.

Limited Data

• Three distributions.
• Uniform distribution
• Triangular distribution
• Beta distribution
• Uniform distribution is just a special case of beta distribution

Summary

• Here, we discussed about useful statistical model, queuing systems, Inventory and
supply-chain systems, Limited data, Reliability, and maintainability.
• Credits: Jerry Banks book.

172

click here to download ppt

Overview

• Here, we will learn about the Bernoulli distribution.

Bernoulli Distribution

• Bernoulli trail is one of the simplest experiment you can do in statistic and
probability which has one out of two possible outcome.
• Bernoulli distribution is a discrete probability distribution for Bernoulli trail which
has one out of two possible outcomes success or fail.
 • Let us consider an example which consists of n trails and each of which can be a
success or a failure.
• Xj = 1 if the jth experiment is success
• Xj = 0 if its failure.
• N trails and each trail has only two out comes success or failure.

• Success here happens with probability (p) and failure at (p – 1).

• X is a discrete random variable. It takes value 1 in case of success and 0 in case of
failure.
• Below is the expected value of random variable X is calculated a

• Below is the variance of the random variable X is calculated as:

Summary

• We have understand about the Bernoulli distribution.

• Credits: Jerry Banks book.

173

click here to download ppt

Overview

• Here, we will learn about the binomial distribution.

Binomial Distribution

• Binomial distribution is the discrete probability distribution with parameters n and p

of the number of successes in the sequence of n independent experiments.
• Each n and p with its own boolean-value.
• Random variable.
• Bernoulli trial’s random variable X has binomial distribution:

• This equation is for calculating the probability of a specific outcome with all
successes.

• S = success.
• In first X trials.
• n-x =f (failures).

• q=1-p
• This equation shows the required number of successes and failures.

• Again this equation can be used to calculate mean and variance of binomial
distribution.
 • X = sum of independent Bernoulli random variables. Then:

• Expected value E(X)

• n = total number of experiments.

• p = probability of each experiment getting successful result.
• X = binomially random variable.
• Variance V(X)

• n = total number of experiments.

• p = probability of each experiment getting successful result.
• X = binomially random variable.
• q = probability of failure in each experiment.

Summary

• We have understand about the Binomial distribution.

• Credits: Jerry Banks book.

174

click here to download ppt

Overview

• Here, we will learn about the Poisson distribution.

Poisson Distribution
 • A discrete frequency distribution that gives the probability of independent events
occurring in a fixed interval of time is known as the Poisson distribution.

• X = 0,1,2,3,….
• e= 2.71828
• a= mean number of successes in the given time interval or region of space.
• a>0

• This is the important property of Poisson distribution that is mean and variance is
equal to a .

• This is the cumulative distribution function.

Summary

• We have understand the Poisson distribution.

• Credits: Jerry Banks book.

175

click here to dowload ppt

Overview

• Here, We will understand “Uniform Distribution” also known as “rectangular

distribution”.

Uniform Distribution
• The uniform distribution is a continuous distribution.
• It takes values within a specified range.
• Probability density function for a uniform distribution always take a random number
on [0,1].

• Consider a random variable X on the interval [a, b] distributed uniformly. The,

equation is given by:
• X=a+b-aR
• For the derivative of function.
• First, compute the cdf of variable X .
• Second, set Fx=R on the range of X .
• Third, solve the equation Fx in terms of R for X.
• Step 1:
o The cdf for the equation is :

• If X takes only discrete values 0 and 1 then:

• Step 2:
• Set
 • Step 3:
• Solve for X in terms of R which proves.

X=a+b-aR

Summary

• Here, we understood about the fundamentals of Uniform Distribution.

• Credits, Discrete Event System Simulation, Jerry Banks.

176

click here to download ppt

Overview

• Here, we will understand About “Exponential Distribution”.

Exponential Distribution

• The Exponential Distribution is a continuous valued probability distribution.

• Takes positive real values and λ which controls the shape of the distribution.
• The probability density function

• This can also be defined as right-continuous Heaviside step function.

• Here λ is called the rate parameter.

• The cumulative distribution function
 • This can also be defined as heaviside step function

• Here λ is the number of occurrences per time unit.

The inverse transform technique is used when the form of cdf is so simple that its f-1 can be
computed easily.

This technique for exponential distribution consist of four steps defined next.

• Step 1:
• Compute the cdf of random variable x given

• Step 2
• Set Fx=R on the range of x which becomes as.

• Step 3
• Solve the equation FX=R for X in terms of R as follows:
• Here, X is called random-variate generator.
• Step 4
• Generate uniform random numbers to compute desired random variates.

• For exponential case.

 • If i=1,2,3… replace 1-Ri with Ri as follows:

• Which proves that both Ri and 1-Ri are uniformly distributed on [0, 1].

Summary

• We understand about “Exponential Distribution”

• Credits, Discrete Event System Simulation, Jerry Banks.

177

click here to download ppt

Overview

• Here, we will discuss about “Triangular Distribution”.

Triangular Distribution

• A triangular distribution is a continuous probability distribution with a probability

density function shaped like a triangle defined by
• The minimum value a
• The maximum value b
• The peak value c
• Where a<b and a≤c≤b
• Probability density function with variable X

• Where endpoints are 0,2 and mode at 1 .

• Triangular Distribution
• Its cumulative distribution function is given as :
 • For 0≤X≤1:

R=x22

• Which implies that x=2R

• For 1≤X≤2:

R=1-2-X22

• Which implies that x=2-21-R

• Thus X is generated as:

Summary

• Here, we have understood the basics of Triangular Distribution, its pdf, and cdf.
178

click here to download ppt

Overview

• Here, we will learn about characteristics of queueing systems.

CHARACTERISTICS OF QUEUEING SYSTEMS

• Customers and servers are the key elements of a queueing system.

• A Customer is anything that visits a facility and requires some service.
• Servers the service provider.
• Not every time the customer visits the service facility.

Summary

• We have learnt about the characteristics of the Queueing systems.

• Credits: Jerry Banks books.
179

click here to download ppt

Overview

• Here, we will learn about the calling population.

The Calling Population

• Calling population is the population of potential customers.

• Finite or infinite.
• Example of the bank having five machines and each machine is a customer and
worker is its server.
• Infinite calling population when the system has a large population of potential.
• Examples are banks and restaurants.
• Arrival rate is the main difference between the finite and infinite population models.
• Infinite population system is not affected by the arrival rate.
• The Finite system depends on the no of customers waiting and being served.

Summary

• We have learnt about the calling population and what are effects of finite and
infinite population.
• Credits: Jerry Banks book.

180

click here to download ppt

Overview

• Here, we will learn about the system capacity.

System Capacity

• Every queueing system has some capacity.

• They have different number of customers entering the system.
• There is a limit for every queueing system to hold the number of customers.
• If the entering system finds the capacity full it will return to the calling population.
• Some systems have unlimited capacity
 • Like tickets sold for a concert to students and many students can wait to purchase
them.
• In limited systems, there is a distinction in arrival time and effective arrival time.

Summary

• We have learnt about the system capacity.

• Credits: Jerry Banks book.

181

click here to download ppt

Overview

• Here, we will learn about the arrival process.

Arrival Process

• For infinite population models the arrival process characterized in terms of

interarrival times of successive customers.
• Probability distribution is used to characterize the interarrival times when arrival is
occurred in random time.
• Poisson arrival process is the most important for random arrivals.
• If An = inter-arrival
• Between n-1(customer) and n(customer) then
• An is exponentially distributed with mean 1/λ time units.

• The arrival rate is λ customers per time unit.

• So the number of arrivals in a time interval of length t has Poisson distribution with
mean λt customers.
• In restaurants and banks the Poisson arrival process has been employed
successfully.
• Scheduled arrival is the second arrival class.
• Inter-arrival could be either constant or constant plus or minus a small random
amount to represent early or late arrivals.
• Third situation occurs when one customer is always present in the queue.
• So we do not have an idle server.
• For finite population the arrival process is different.
 • We define the customer as pending when he is out of the queue system and a
member of potential calling population.
• Arrival Process

• Important application of finite population model is machine repair problem.

Summary

• We have learnt about the finite and infinite arrival processes.

• Credits: Jerry Banks book.

182

click here to download ppt

Overview

• Here, we will learn about the Queue behavior and Queue discipline.

Queue Behavior

• The actions of the customer while waiting in the queue system for his service is
known as the queue behaviour.
• The customers can baulk, renege or jockey.
Queue Discipline

• The logical ordering of the customers in a queue and determining that which
customer should come for service when the server is free.
• FIFO, LIFO, SIRO, SPT and PR are some common queue disciplines.

Summary

• We have learnt about the Queue behavior and Queue discipline.

• Credits: Jerry Banks book.

183

click here to download ppt

Overview

• Here, we will learn about the Service Times and the Service Mechanism.

Service Times and the Service Mechanism

• Service times of successive arrivals are denoted by S1 S2…..

• Their duration can be random or constant.
• This latter case is characterized as a sequence of random or identically distributed
variables.
• The exponential, Weibull, gamma, lognormal and truncated normal distributions are
used for service times.
• Sometimes the services are identically distributed for all same customers. Same
type customers can have different service time.
• Number of service centers and interconnecting queues are present in a queue
system.
• Parallel working servers for each service centre.
• Server = c
• Single server (c = 1)
• Multiple servers (1<c<∞ )
• Unlimited servers(c = ∞ )

Service Times and the Service Mechanism

Service Times and the Service Mechanism
Service Times and the Service Mechanism

Summary

• We have learnt about the Service times and the Service mechanism.
• Credits: Jerry Banks book.

184
click here to download ppt

Overview

• Here, we will learn about the Queueing notation.

Queueing Notation

• Proposed by Kendall [1953]

• For parallel server systems.
• A/B/c/N/K based on this format.
• “A” represents the inter-arrival-time distribution.
• “B” represents the service-time distribution.
• “C” represents the number of parallel servers.
• “N” represents the system capacity.
• “K” represents the size of the calling population.
• Common symbols for A and B include M (exponential or Markov)
• D (constant or deterministic).
• Ek (Erlang of order k).
• PH (phase-type).
• H (hyperexponential).
• G (arbitrary or general)
• Gl (general independent).

Queueing Notation
Summary

• We have learnt about the queueing notation.

• Credits: Jerry Banks book.

185

click here to download ppt

Overview

• Here, we will learn about the long-run measures of performance of queueing

systems.

LONG-RUN MEASURES OF PERFORMANCE

 • The primary long-run measures of performance of queueing systems are:
• Long-run time-average number of customers in the system (L).
• In the queue (LQ).
• Long-run average time spent in the system (w).
• In the queue (wQ) per customer.
• Server utilization.
• A Proportion of time that a server is busy (p).
• Waiting line and mechanism is a system.
• System can also be sub-system of queue system.
• “queue” means waiting line alone.
• Other measurements include long-run proportion of customer who is delayed
longer than t o time units.
• Customers turned away due to capacity constraints.
• Waiting line containing more than ko customers.
• G/G/c/N/K is defined.
• Their performance relationships and simulation run.
• The two types of estimators are:
• An ordinary sample average.
• A time-integrated sample average.

Summary

• We have learnt about the long-run measures of performance of queueing systems.

• Credits: Jerry Banks book.

186

click here to download ppt

Overview

• Here, we are going to learn about properties of random numbers.

Properties of Random Numbers

• Uniformity and independence are two important statistical properties.

• Each sample must be drawn from continuous uniform distribution between 0 and 1.

• The density function and the expected value of Ri is:

 • And the variance is:

• Generated by a digital computer.

• Individual computation is expensive so computationally efficient method should be
used.
• Routine should be portable for different computers and programming languages.
• Routine having a long cycle.
• Replicable random numbers.
• Random numbers should closely approximate the ideal statistical properties.
• Techniques to generate random is easy to develop.
• Techniques that produce sequences that are independent and uniformly distributed
are difficult.

Summary

• We have learnt about the properties of the random numbers.

• Credits: Jerry Banks book.

187

clcik here to download ppt

Overview

• Here, we will learn about the linear congruential method.

Linear Congruential Method

• Widely used technique.

• Yields sequences with longer periods.
• Produces the sequence of integers.
• Between zero and m-1 .
 • xo = seed.
• a= multiplier.
• c= increment (c≠0) .
• m= modulus.
• This is called mixed congruential method.
• When c = 0 then known as multiplicative congruential.
• Interval [0,1] is divided into n classes than expected observation for each interval
is Nn .
• N is the total number of observations.
• Observing a value in a particular interval is independent.

Summary

• We have learnt about the linear congruential method.

• Credits: Jerry Banks book.

188

clcik here to download ppt

Overview

• Here, we will learn about the generation of pseudo-random numbers.

Generation Of Pseudo-random Numbers

• False number generation.

• Removes the potential of true randomness.
• Set of random numbers can be replicated.
• Producing numbers between 0 and 1 that simulates the properties of uniform
distribution and independence.
• Some errors or departures are:
• Not uniformly distributed random numbers.
• Not too high or too low variance.
• Generated numbers might not be continuous valued but are discrete valued.
• Dependence: Autocorrelation between numbers.
• Numbers successively higher or lower than adjacent numbers.
• Several numbers above the mean followed by the several numbers below the mean.
 • Departures from uniformity and independence for a particular generation scheme
often can be detected by tests.
• If detected then dropped in the favour of an acceptable generator.

Summary

• We have learnt about the generation of pseudo-random numbers.

• Credits: Jerry Banks book.

189

click here to download ppt

Overview

• Here, we will understand about the Tests for Random numbers.

Tests for Random numbers

• Two categories:
• 1. Uniformity.

• Ho denotes numbers distributed uniformly on the interval [0, 1].

• 2. Independence.

• Ho denotes the independent numbers.

• A level of significance α must be stated for each test.
• Where α is probability of rejecting the null hypothesis.
 • Software not specifically developed for simulation comes with random number
generator pre-installed

Additional Tests:

• Good’s serial test, the median-spectrum test, the run’s test and a variance
heterogeneity test.

Summary

• Here, we understood about the tests for random numbers.

• Credits: Jerry banks book.

190

click here to download ppt

Overview

• Here, we will understand about the “Random Variate Generation”.

Random Variate

• A random variate is a generated variable using uniformly distributed pseudorandom

numbers.
• A random variate can be uniformly or nonuniformly distributed.
• Modeling activities is unpredictable or indeterminate.
• Random Variate Generations explain and illustrate some techniques for generating
random
• Sometimes there is no availability of random-variate-generation libraries.
• It is up to the modeller to create an acceptable routine.
• This topic includes:
o inverse-transform technique
o Acceptance-rejection technique
• Special properties
• The composition method

Summary

• Here, we understood about Random Variate Generation.

 • Credits: Jerry banks book.

191

click here to download ppt

Overview

• Here, we will understand about the “Inverse Transform Technique”.

Inverse Transform Technique

• The inverse-transform technique can be used to sample from the exponential.

• The uniform.
• The Weibull.
• The triangular and empirical distributions.
• The basic principle is to find the inverse function of F , F-1 such that FF-1=I

• Where F-1 denotes the solution of the equation r=Fx in terms of γ .

Inverse Transform Technique

Inverse Transform Technique

• This technique explains:

o Exponential Distribution
o Uniform Distribution
o Weibull Distribution
o Triangular Distribution
o Empirical Continuous Distribution
o Continuous Distribution
o Discrete Distribution

Summary

• Here, we understood about Inverse Transform Technique.

• Credits: Jerry banks book.

192

click here to download ppt

Overview

• Here, We will understand “Acceptance Rejection Technique”.

Acceptance Rejection Technique

• Acceptance Rejection Technique is a method used to generate random variate using

some steps explained next.
• Step 1. Generate a Random Number R .
• Step2a. If R≥14 Accept R .
• Step2b. If R≤14 Reject R .
• If another uniform random variate on [1/4, 1] is needed, repeat the procedure.
• The random variate generated using above methods is indeed uniformly distributed
over [1/4, 1].
• Take any 14≤a<b≤1 then

• Proves that probability for uniform distribution is on [1/4, 1].

• The efficiency of an acceptance-rejection technique rest on reducing the total of
rejections.
• Which depends on computer being used.
• The skills of programmer.
• Rejected numbers.
• The Distributions which operates on acceptance rejection technique are:
o Poisson Distribution
o Non-stationary Poisson process
o Gamma Distribution.

Summary

• Here, we understood about the Acceptance Rejection Technique.

• Credits, Jerry Banks book.

193

click here to download ppt

• Overview
 • Here, we will discuss about “Other topics in Randomness”.

Special Properties

• Special Properties are variate generation techniques based on a particular feature of

probability distributions.

Direct Transformation

• Direct transformation is used for the normal and lognormal distributions.

• The cdf is given by:

• Box and Muller made this method and is easy to program in scientific programming
languages.

Convolution Method

• The probability distribution of a sum of two or more independent random variables

is called convolution method.
• This method uses two random variables.
• Sum them and generate a new random variable.
• This technique can be applied on Erlang and Binomial variates.

More Special Properties

• The other relationship among distributions include relation between gamma and
beta distribution.
• We can generate beta variates, using two gamma variates with acceptance-
rejection technique.

Summary

• Here, We understood about the Special Properties, Direct transformation,

convolution method and other special properties.
• Credits: Jerry banks book.

194
click here to download ppt

Overview

• Here, we will discuss about “Data Collection”.

Data Collection

• The relation between construction of the model and the collection of needed input
data is constant.
• The required data elements changes with complexity of the model.
• Time taking nature of data collection.
• Which type of data is to be collected is dictated by the objectives of study.

Summary

• Here, We understood about the Data Collection.

• Credits: Jerry banks book.

195

click here to download ppt

Overview

• Here, we will understand About “Identifying the Distribution with Data”.

Identifying the Distribution with Data

• Data distributions are graphical methods.

• These methods are used for selecting input distribution families.
• The specific distribution is often specified by estimating its parameters.

Summary

• We understood about Data Distribution and how data parameters are selected.
• Credits, Jerry Banks book.

196

click here to download ppt

Overview
 • Here, we will understand about the “Histograms”.

Histograms

• Histograms is a powerful technique for identifying how a distribution is shaped.

• Histogram is also known as frequency distribution.
• There are some methods for constructing a histogram.

1. Divide the data ranges into intervals of equal width.

2. Label the horizontal axis to follow the nominated intervals.

3. Find the frequency of occurrences inside each intervals.

4. Label the vertical axis to plot total occurrences for every interval.

5. Plot the frequencies on the vertical axis.

• The number of class intervals is decided by the number of observations.

• Amount of scatter in the data.

Histograms

Summary

• Here, we understood about the “Histograms”.

• Credits: jerry banks book.

197
click here to download

Overview

• Here, we will understand about the “Selecting the Family of Distributions”.

Selecting the Family of Distributions

• The selection of a family of distributions is based on the context for what it is being
investigated.
• The shape of the histogram.
• The Distributions which are frequently encountered and easy to analyze includes
exponential, normal and poison distribution.
• Whereas difficult to analyze includes the beta, gamma and Weibull distributions.
• Out of many, Here are some examples of distributions which were created on the
physical basis.
o
o Binomial
o Negative Binomial
o Poisson
o Normal
o Lognormal
• An input Model is an estimate of reality.
• There is no “true” distribution for any stochastic input process.

Summary

• Here, we understood about the Family of distributions, its types and input model.
• Credits: Jerry banks book.

198

click here to download ppt

Overview

• Here, we will understand about the “Quantile-Quantile Plot”.

Quantile-Quantile Plot

• A Quantile-Quantile is a graphical method for comparison of two probability

distributions by plotting their qualities against each other.
• If Fγ=PX≤γ=q for 0<q<1
• Here x is random variable with cdf F .
 • Q-quantile of X is value of .
• We write γ=F-1q when F has an inverse.
• If xi,ⅈ=1,2,3⋅⋯n and yj,j=1,2,3⋅⋯n
• Where y1≤y2≤⋯≤yn
• Let j denote the ranking of order
• j=1 for smallest
• j=n for largest
• Then

• yj ≈ F-1j-12n

Quantile-Quantile Plot

Summary
 • Here, we understood about the “Quantile-Quantile Plot”.
• Credits: jerry banks book.

199

click here to download ppt

Overview

• Here, we will learn about the parameter estimation.

Parameter Estimation

• After selecting the family of distribution, the next step is to estimate the parameters
of the distribution.
• Many useful distribution estimators are described here.
• Software packages integrated into simulation languages are available.

Summary

• We have learnt about the parameter distribution.

• Credits: Jerry Banks book.

200

Click here to download ppt

Overview

• Here, we will learn about the preliminary statistics: sample mean and sample
variance.

Preliminary Statistics

• To estimate the parameters of a hypothesized distribution.

• When discrete and continuous raw data is available then the following equation is
used:

• x = sample mean.
• x1,x2,…,xn = observation in a sample of size n.
• The sample variance s2 is defined as:

• When discrete or continuous raw data are available following modified equations
are used:

• Where x is the sample mean.

• And the sample variance is :

• Here k is the number of distinct values of X.

• Fj is the observed frequency of the value X j of X.
• When data is discrete or continuous and have been placed in class intervals then
the sample mean and the sample variance are approximated from the following
equations:

• For the sample variance the following equation is used.

 • fj observed frequency of jth class interval.
• mj midpoint of jth interval.
• c = number of intervals.

Summary

• We have learnt about the preliminary statistics: sample mean and sample variance.
• Credits: Jerry Banks book.