tutorial Statistical reliability analysis on Rayleigh probability distributions

Size: px
Start display at page:

Download "tutorial Statistical reliability analysis on Rayleigh probability distributions"

Transcription

1 tutorial Statistical reliability analysis on Rayleigh probability distributions These techniques for determining a six-sigma capability index for a Rayleigh distribution use sample mean and deviation. By Russell J. Hoppenstein Statistical analysis is a useful tool for obtaining the reliability information of a device or process based on a limited number of samples. Sample sets are analyzed referenced to a known distribution function. Once a distribution function is assumed, the sample mean and variance are used to determine reliability information. Most processes follow the general trend of a Gaussian distribution. Gaussian statistics are most often used to characterize performance. The Gaussian random variable is generated by the infinite sum of a uniform random variable, per the Central Limit Theorem. The Gaussian probability distribution function (pdf) is Figure 1. Gaussian pdf. shown in Figure 1. It is bounded by plus and minus infinity and is symmetrical about its mean. The distribution is essentially defined by two parameters: its mean and standard deviation, sigma. Not all events can be characterized by a Gaussian distribution. Other types of distributions surface that behave in unique ways for which applying a Gaussian type of analysis does not fit the data set. For those types of distributions, the typical reliability equations must be altered to accurately relate the true characteristics of the process. Rayleigh pdf One common pdf is called the Rayleigh distribution, which characterizes processes that are determined by two independent, orthogonal, Gaussian random variables. The Rayleigh random variable is created by the following: R X + Y The variables X and Y are independent, zero mean Gaussian variables. The variable R is a Rayleigh random variable. The Rayleigh variable is essentially the distance from the origin to a point defined by two orthogonal parameters. The Rayleigh pdf is a twodimensional probability function, which can be difficult to analyze; however, the function can be simplified if the coordinate axes are transformed into cylindrical coordinates. As such, the pdf of the random variable R is symmetrical with respect to angle theta and can be expressed as: r r fr( r) exp a a The Rayleigh pdf is illustrated in Figure. Within this form, the function is more easily analyzed. The variable a is a constant used to determine the proper shape of the distribution. The distribution varies from zero to infinity. The mean and standard deviation of the distribution are not independent; they are both related to the variable a by: µ R a The variable a is derived from the standard deviation of the independent Gaussian functions that define it, which is: a σ X σ Y π π σ R a The Rayleigh random variable is used to define a process that varies as a Gaussian distribution within independent directions of X and Y. In the physical world this can be thought of as the position of an arrow on a target shot by an archer. The archer will aim for the bull s eye, but inevitable random errors in horizontal and vertical motion will cause the arrow to randomly hit the target. The position of the arrow with respect to the bull s eye is a form of a Rayleigh distribution. The Rayleigh distribution also characterizes more pertinent measurements within the communications field. Examples include measurements for input return loss, modulation sideband rejection, carrier suppression, and RF fading. These measurements consist of two independent and orthogonal parameters, expressed as either the amplitude and phase, or the real and imaginary parts. The magnitude of the vector from the starting point to the given point is defined as the Rayleigh random variable. continued on page October 000

2 Reliability The general purpose of applying statistical analysis is to determine reliability information. Because the Gaussian distribution is most common, the equations and terminology for reliability are generally based on those distributions; however, those equations do not apply to different types of random variables. First, the Gaussian reliability information and terminology are defined. The reliability information is expressed as a probability of finding a measurement value within certain limits. For example, the probability of finding a value within +/- one Gaussian sigma of the mean is 68.3% or stated mathematically as: P[ µ σ ] X P[ µ + σ] 68. 3% Furthermore, the error probability is calculated from the remaining percentage: 31.7% in the example above. Recall that the Gaussian distribution is symmetrical about its mean and each side or tail moves asymptotically toward zero; therefore, the error region within each tail contributes equally to the error probability. The terminology of achieving sixsigma reliability has become fairly renowned in the industry, but what does it mean? The six-sigma standard is initially confusing because the defined error probability is actually associated with a 4.5σ process. The error rate is determined by the probability of finding a point within one of the tails that is greater than 4.5σ away from the mean. The six-sigma nomenclature is introduced because the mean of the process is allowed to shift by 1.5σ from nominal as shown in Figure 3. Figure. Rayleigh pdf as a Function of Radius, r. Regardless of any mean shift, the defect rate derived from a 4.5σ process is kept constant. Table 1 shows probabilities for given deviations around the mean and the appropriate defect rate for a Gaussian distribution. Notice that no finite window encompasses every 76 October 000

3 Figure 3. Shifted mean Gaussian distribution. point on the distribution (i.e. probability is never 100%). The probability of a value falling outside 4.5σ per each side is or parts per million (ppm). This probability is small, but not zero; from here forth, six-sigma quality will be equated with an error probability fewer than 3.4 failures in a million tries. We can use the probability information of the Gaussian process as a function of its standard deviation to determine an equivalent expression for the Rayleigh distribution as a function of its standard deviation. The Rayleigh cumulative distribution function (cdf) is equated to the desired Gaussian probability: r 1 exp P + [ X µ nσ a ] G 1 Pe[ nσ G] where n represents any real positive number. Rearranging the equation and substituting the Rayleigh standard deviation in place of variable a yields the following: This equation provides a real scaling factor to equate a nσ Gaussian probability to an equal kσ Rayleigh probabilir σ R - ln Pe[ n G σ ] π October 000

4 Figure 4. Shifted mean Rayleigh distribution. ty. Table provides the scaling factors for commonly used probabilities. To achieve the equivalent error probability associated with a 4.5 Gaussian sigma for a Rayleigh distribution, the sample standard deviation must be scaled by Equivalently, the expression 4.5σ G 7.64σ R could be used. Capability index Reliability information is often expressed as a capability index. The capability index relates how well a sample data set meets a six-sigma probability for success. For a Gaussian distribution, the peak capability index, C pk, is given by: G min[ µ LSL, USL µ ] 3σ the mean could shift up to 1.5σ and still achieve the defect rate associated with a 4.5σ process. The scaled value of the Rayleigh standard deviation can be substituted into the above equation to generate a revised equation: R 15. min( USL µ R) 766. R µ R Shifted mean Rayleigh distribution Recall that the Rayleigh random variable is created from two independent, zero mean Gaussian variables. What happens if the contributing variables are not zero mean functions? In this case, the Rayleigh distribution becomes more complicated. With zero mean contributing variables, the Rayleigh distribution is symmetrical in variable theta within cylindrical coordinates. If one or both of the contributing variables has a shifted mean and then the Rayleigh distribution loses its symmetry, the distribution function becomes skewed when expressed as a function of r. Proper analysis would require the use of two-dimensional probability density functions, which negate the simple scaling factors used above; however, applying certain assumptions, the capability index can be approximated. For a skewed-mean Rayleigh distribution, the function is still bound by zero and infinity. The skewing effect essentially spreads out the distribution. This causes the mean and the standard deviation to increase. That, in turn, would push out the point for which a six-sigma probability is reached, but that point is unknown. Using the data from the sample set, some initial approximations can be made. Using the sample standard deviation, the equivalent non-skewed mean where LSL and USL represent lowerand upper-spec limits respectively. The C pk calculation explicitly uses the minimum distance in the numerator to provide the worst-case scenario. For convenience, the C pk equation can be scaled such that the denominator is equal to 4.5σ: Figure 5: Input return loss histogram and shifted mean distribution. G 15. min[ µ LSL, USL µ ] 45. σ It is evident that a 4.5σ probability of success occurs if the distance from the mean to the limit is equal to or greater than the distance of mean to the 4.5σ point; therefore, a C pk of value 1.5 or greater represents a process that meets a 4.5σ probability of success. Note, for a Gaussian distribution, a C pk value of or greater signifies that for Rayleigh distributions. Recall that for Rayleigh distributions, the sampled data are only valid for real positive numbers; so the mean should always be below the upper specification limit. Obtaining a C pk value of 1.5 using the above equation with a Rayleigh distributed sample set represents the same probability of success as that of a Gaussian distributed sample set meeting a six sigma defect rate. can be computed. This is the standard Rayleigh distribution with an equivalent variance as that of the sample data. Then, the non-skewed distribution is shifted such that its mean is equal to that of the sample mean. The result is a distribution that is more conservative than the actual data. The distance from the 4.5σ point to the mean can be approximated by the distance of the 4.5σ point of a standard Rayleigh distribution of given standard 80 October 000

5 ±σ Probability P e (per side) % 15.9% % 0.135% % % % % Table 1. Gaussian probabilities. deviation to that of its associated mean. That distance is then compared to the distance from the upper limit to the actual sample mean. The capability index is then found to be: 15. ( USL µ S 766. σ S µ 15. ( USL µ σ S ( USL µ σ S n P e [nσ G ] k % % 5.549% ppm ppm Table. Rayleigh scaling factors. where the subscript S is used to represent the parameters from a sample set. Recall that, for a standard Rayleigh distribution, the standard deviation and mean are not independent functions, so the denominator of the above equation can be simplified to a scaled quantity of sigma. The above equation will always provide a pessimistic estimate in which the error decreases as the means of the contributing variables approach zero. Experimental results The above techniques can be illustrated with an example and verified with experimental data. Simulated data were used to generate a skewedmean Rayleigh distribution. The distribution data were generated by two independent Gaussian variables with means that deviated from zero. Figure 4 illustrates three curves related to the data. The first is a histogram of the 5,000 data points. The second is a continuous Rayleigh distribution curve that has an equivalent standard deviation as the data set. The third curve represented the previous distribution that is shifted so that the mean matches the data set. Observe that the tail of the shifted mean distribution is higher than that of the data set; so error probabilities associated with the data set will be lower than that of the shifted mean distribution. Next, input return loss data from a 1900 MHz linear power amplifier was analyzed using the standard Gaussian equations and the modified Rayleigh equations. The measurements are originally recorded in decibels so the initial Gaussian equations are applied to those parameters. To apply the Rayleigh analysis, the data must first be converted to the linear scale. Table 3 shows the statistical data and the calculated capability index for the measurement. Note, the resultant C pk, using Gaussian techniques on the raw data, is less than unity. As such, the process would be incorrectly deemed less than six-sigma quality. On the other hand, using the more relevant Rayleigh analysis yields a C pk greater than 1.5, which characterizes a process below the six-sigma defect rate. To further enhance the notion that the data follows a Rayleigh distribution, the histogram of more than,000 data points and the equivalent shifted mean distribution are shown in Figure 5. Note that this sample set has a small mean shift which is imperceptible on the graph. Nevertheless, the distribution follows the general trend of a Rayleigh distribution. Conclusion Statistical analysis can be beneficial if the techniques and equations are used correctly. The modified capability Mean Std. Dev. Limit C pk Gaussian analysis db 3.7 db 8.0 db 0.80 Rayleigh analysis Table 3. Gaussian and Rayleigh C pk comparisons. index for Rayleigh distributed variables offers an additional tool for analyzing non-gaussian processes. Because the nature of the contributing variables cannot be known, the C pk for the shifted mean distribution should be used. It is guaranteed to provide reliability performance at a minimum, achieving six-sigma quality. Most of the parameters to which the Rayleigh analysis would apply are expressed in decibels. As expected, the transformation of the pdfs from the linear scale to the decibel scale is non-linear. Generating equivalent expressions for mean, standard deviation, and C pk for a Rayleigh distribution in log scale is not possible; therefore, the sample data must always be transformed to linear scale for the analysis to be valid. References [1] Williams, Richard H., Electrical Engineering Probability. St. Paul, MN: West Publishing Company, [] Papoulis, Athanasios, Probability, Random Variables, and Stochastic Processes. New York : McGraw Hill, About the author Russell J. Hoppenstein is the lead electrical engineer for Motorola in Ft. Worth, Texas. Hoppenstein acquired his undergraduate degree in electrical engineering at the University of Texas at Austin in 199 and a MSEE at the University of Texas at Arlington in He has worked in the development and design of cellular base stations at Motorola in Ft. Worth for eight years. Through the development and release of cellular products he has had extensive exposure to reliability and quality management. Currently, he is working in the Linear Power Amplifier group on 3G CDMA amplifiers. He can be reached at ; qrh00@ .mot.com. 8 October 000

1.0 Continuous Distributions. 5.0 Shapes of Distributions. 6.0 The Normal Curve. 7.0 Discrete Distributions. 8.0 Tolerances. 11.

1.0 Continuous Distributions. 5.0 Shapes of Distributions. 6.0 The Normal Curve. 7.0 Discrete Distributions. 8.0 Tolerances. 11. Chapter 4 Statistics 45 CHAPTER 4 BASIC QUALITY CONCEPTS 1.0 Continuous Distributions.0 Measures of Central Tendency 3.0 Measures of Spread or Dispersion 4.0 Histograms and Frequency Distributions 5.0

More information

Numerical Methods Lecture 7 - Statistics, Probability and Reliability

Numerical Methods Lecture 7 - Statistics, Probability and Reliability Topics Numerical Methods Lecture 7 - Statistics, Probability and Reliability A summary of statistical analysis A summary of probability methods A summary of reliability analysis concepts Statistical Analysis

More information

Statistical Concepts. Distributions of Data

Statistical Concepts. Distributions of Data Module : Review of Basic Statistical Concepts. Understanding Probability Distributions, Parameters and Statistics A variable that can take on any value in a range is called a continuous variable. Example:

More information

Ø Set of mutually exclusive categories. Ø Classify or categorize subject. Ø No meaningful order to categorization.

Ø Set of mutually exclusive categories. Ø Classify or categorize subject. Ø No meaningful order to categorization. Statistical Tools in Evaluation HPS 41 Dr. Joe G. Schmalfeldt Types of Scores Continuous Scores scores with a potentially infinite number of values. Discrete Scores scores limited to a specific number

More information

Advanced 3 G and 4 G Wireless Communication Prof. Aditya K Jagannathan Department of Electrical Engineering Indian Institute of Technology, Kanpur

Advanced 3 G and 4 G Wireless Communication Prof. Aditya K Jagannathan Department of Electrical Engineering Indian Institute of Technology, Kanpur Advanced 3 G and 4 G Wireless Communication Prof. Aditya K Jagannathan Department of Electrical Engineering Indian Institute of Technology, Kanpur Lecture - 19 Multi-User CDMA Uplink and Asynchronous CDMA

More information

Ø Set of mutually exclusive categories. Ø Classify or categorize subject. Ø No meaningful order to categorization.

Ø Set of mutually exclusive categories. Ø Classify or categorize subject. Ø No meaningful order to categorization. Statistical Tools in Evaluation HPS 41 Fall 213 Dr. Joe G. Schmalfeldt Types of Scores Continuous Scores scores with a potentially infinite number of values. Discrete Scores scores limited to a specific

More information

(a) The density histogram above right represents a particular sample of n = 40 practice shots. Answer each of the following. Show all work.

(a) The density histogram above right represents a particular sample of n = 40 practice shots. Answer each of the following. Show all work. . Target Practice. An archer is practicing hitting the bull s-eye of the target shown below left. For any point on the target, define the continuous random variable D = (signed) radial distance to the

More information

ECE Homework Set 2

ECE Homework Set 2 1 Solve these problems after Lecture #4: Homework Set 2 1. Two dice are tossed; let X be the sum of the numbers appearing. a. Graph the CDF, FX(x), and the pdf, fx(x). b. Use the CDF to find: Pr(7 X 9).

More information

Reference: Chapter 7 of Devore (8e)

Reference: Chapter 7 of Devore (8e) Reference: Chapter 7 of Devore (8e) CONFIDENCE INTERVAL ESTIMATORS Maghsoodloo An interval estimator of a population parameter is of the form L < < u at a confidence Pr (or a confidence coefficient) of

More information

λ = pn µt = µ(dt)(n) Phylomath Lecture 4

λ = pn µt = µ(dt)(n) Phylomath Lecture 4 Phylomath Lecture 4 Brigid O Donnell (17 February 2004) A return to sojourn times We began by returning to the idea of sojourn times, or the time period until the next disruption event occurs for a given

More information

As an example, consider the Bond Strength data in Table 2.1, atop page 26 of y1 y 1j/ n , S 1 (y1j y 1) 0.

As an example, consider the Bond Strength data in Table 2.1, atop page 26 of y1 y 1j/ n , S 1 (y1j y 1) 0. INSY 7300 6 F01 Reference: Chapter of Montgomery s 8 th Edition Point Estimation As an example, consider the Bond Strength data in Table.1, atop page 6 of By S. Maghsoodloo Montgomery s 8 th edition, on

More information

Wooldridge, Introductory Econometrics, 4th ed. Appendix C: Fundamentals of mathematical statistics

Wooldridge, Introductory Econometrics, 4th ed. Appendix C: Fundamentals of mathematical statistics Wooldridge, Introductory Econometrics, 4th ed. Appendix C: Fundamentals of mathematical statistics A short review of the principles of mathematical statistics (or, what you should have learned in EC 151).

More information

AUTOMATED TEMPLATE MATCHING METHOD FOR NMIS AT THE Y-12 NATIONAL SECURITY COMPLEX

AUTOMATED TEMPLATE MATCHING METHOD FOR NMIS AT THE Y-12 NATIONAL SECURITY COMPLEX AUTOMATED TEMPLATE MATCHING METHOD FOR NMIS AT THE Y-1 NATIONAL SECURITY COMPLEX J. A. Mullens, J. K. Mattingly, L. G. Chiang, R. B. Oberer, J. T. Mihalczo ABSTRACT This paper describes a template matching

More information

Notes on Noncoherent Integration Gain

Notes on Noncoherent Integration Gain Notes on Noncoherent Integration Gain Mark A. Richards July 2014 1 Noncoherent Integration Gain Consider threshold detection of a radar target in additive complex Gaussian noise. To achieve a certain probability

More information

Random Processes. By: Nick Kingsbury

Random Processes. By: Nick Kingsbury Random Processes By: Nick Kingsbury Random Processes By: Nick Kingsbury Online: < http://cnx.org/content/col10204/1.3/ > C O N N E X I O N S Rice University, Houston, Texas This selection and arrangement

More information

STATISTICS OF MULTIPLE EXTRANEOUS SIGNALS ON A COMPACT RANGE

STATISTICS OF MULTIPLE EXTRANEOUS SIGNALS ON A COMPACT RANGE STATISTICS OF MULTIPLE EXTRANEOUS SIGNALS ON A COMPACT RANGE by John R. Jones and Esko A. Jaska Microwave and Antenna Technology Development Laboratory Georgia Tech Research Institute Georgia Institute

More information

Advanced 3G and 4G Wireless Communication Prof. Aditya K. Jagannatham Department of Electrical Engineering Indian Institute of Technology, Kanpur

Advanced 3G and 4G Wireless Communication Prof. Aditya K. Jagannatham Department of Electrical Engineering Indian Institute of Technology, Kanpur Advanced 3G and 4G Wireless Communication Prof. Aditya K. Jagannatham Department of Electrical Engineering Indian Institute of Technology, Kanpur Lecture - 12 Doppler Spectrum and Jakes Model Welcome to

More information

Probability Methods in Civil Engineering Prof. Rajib Maity Department of Civil Engineering Indian Institute of Technology, Kharagpur

Probability Methods in Civil Engineering Prof. Rajib Maity Department of Civil Engineering Indian Institute of Technology, Kharagpur Probability Methods in Civil Engineering Prof. Rajib Maity Department of Civil Engineering Indian Institute of Technology, Kharagpur Lecture No. # 12 Probability Distribution of Continuous RVs (Contd.)

More information

EE 330 Lecture 3. Basic Concepts. Feature Sizes, Manufacturing Costs, and Yield

EE 330 Lecture 3. Basic Concepts. Feature Sizes, Manufacturing Costs, and Yield EE 330 Lecture 3 Basic Concepts Feature Sizes, Manufacturing Costs, and Yield Review from Last Time Analog Flow VLSI Design Flow Summary System Description Circuit Design (Schematic) SPICE Simulation Simulation

More information

Chapter 4: An Introduction to Probability and Statistics

Chapter 4: An Introduction to Probability and Statistics Chapter 4: An Introduction to Probability and Statistics 4. Probability The simplest kinds of probabilities to understand are reflected in everyday ideas like these: (i) if you toss a coin, the probability

More information

Part 4 (of 4): Chance

Part 4 (of 4): Chance To provide an example and referring to Figure 3.6, we can see how it is possible to obtain 80 possibilities corresponding to his second line. If a box is occupied by 3 particles out of an available 4,

More information

Probability Methods in Civil Engineering Prof. Dr. Rajib Maity Department of Civil Engineering Indian Institute of Technology Kharagpur

Probability Methods in Civil Engineering Prof. Dr. Rajib Maity Department of Civil Engineering Indian Institute of Technology Kharagpur Probability Methods in Civil Engineering Prof. Dr. Rajib Maity Department of Civil Engineering Indian Institute of Technology Kharagpur Lecture No. #13 Probability Distribution of Continuous RVs (Contd

More information

Discrete Simulation of Power Law Noise

Discrete Simulation of Power Law Noise Discrete Simulation of Power Law Noise Neil Ashby 1,2 1 University of Colorado, Boulder, CO 80309-0390 USA 2 National Institute of Standards and Technology, Boulder, CO 80305 USA ashby@boulder.nist.gov

More information

Chapter 1 Statistical Reasoning Why statistics? Section 1.1 Basics of Probability Theory

Chapter 1 Statistical Reasoning Why statistics? Section 1.1 Basics of Probability Theory Chapter 1 Statistical Reasoning Why statistics? Uncertainty of nature (weather, earth movement, etc. ) Uncertainty in observation/sampling/measurement Variability of human operation/error imperfection

More information

Chapter 2 Process Variability. Overview. 2.1 Sources and Types of Variations

Chapter 2 Process Variability. Overview. 2.1 Sources and Types of Variations Chapter 2 Process Variability Overview Parameter variability has always been an issue in integrated circuits. However, comparing with the size of devices, it is relatively increasing with technology evolution,

More information

Gaussian Random Fields

Gaussian Random Fields Gaussian Random Fields March 22, 2007 Random Fields A N dimensional random field is a set of random variables Y (x), x R N, which has a collection of distribution functions F (Y (x ) y,..., Y (x n ) y

More information

Collision Avoidance Lexicon

Collision Avoidance Lexicon Collision Avoidance Lexicon 2017 An understanding of collision avoidance terminology requires an understanding of position uncertainty terminology. This lexicon therefore includes the terminology of both

More information

The normal distribution

The normal distribution The normal distribution Patrick Breheny September 29 Patrick Breheny Biostatistical Methods I (BIOS 5710) 1/28 A common histogram shape The normal curve Standardization Location-scale families A histograms

More information

CHAPTER 1: Functions

CHAPTER 1: Functions CHAPTER 1: Functions 1.1: Functions 1.2: Graphs of Functions 1.3: Basic Graphs and Symmetry 1.4: Transformations 1.5: Piecewise-Defined Functions; Limits and Continuity in Calculus 1.6: Combining Functions

More information

Objective Experiments Glossary of Statistical Terms

Objective Experiments Glossary of Statistical Terms Objective Experiments Glossary of Statistical Terms This glossary is intended to provide friendly definitions for terms used commonly in engineering and science. It is not intended to be absolutely precise.

More information

Polytechnic Institute of NYU MA 2212 MIDTERM Feb 12, 2009

Polytechnic Institute of NYU MA 2212 MIDTERM Feb 12, 2009 Polytechnic Institute of NYU MA 2212 MIDTERM Feb 12, 2009 Print Name: Signature: Section: ID #: Directions: You have 55 minutes to answer the following questions. You must show all your work as neatly

More information

Example 2.1. Draw the points with polar coordinates: (i) (3, π) (ii) (2, π/4) (iii) (6, 2π/4) We illustrate all on the following graph:

Example 2.1. Draw the points with polar coordinates: (i) (3, π) (ii) (2, π/4) (iii) (6, 2π/4) We illustrate all on the following graph: Section 10.3: Polar Coordinates The polar coordinate system is another way to coordinatize the Cartesian plane. It is particularly useful when examining regions which are circular. 1. Cartesian Coordinates

More information

Human-Oriented Robotics. Probability Refresher. Kai Arras Social Robotics Lab, University of Freiburg Winter term 2014/2015

Human-Oriented Robotics. Probability Refresher. Kai Arras Social Robotics Lab, University of Freiburg Winter term 2014/2015 Probability Refresher Kai Arras, University of Freiburg Winter term 2014/2015 Probability Refresher Introduction to Probability Random variables Joint distribution Marginalization Conditional probability

More information

Revision notes for Pure 1(9709/12)

Revision notes for Pure 1(9709/12) Revision notes for Pure 1(9709/12) By WaqasSuleman A-Level Teacher Beaconhouse School System Contents 1. Sequence and Series 2. Functions & Quadratics 3. Binomial theorem 4. Coordinate Geometry 5. Trigonometry

More information

Recall the Basics of Hypothesis Testing

Recall the Basics of Hypothesis Testing Recall the Basics of Hypothesis Testing The level of significance α, (size of test) is defined as the probability of X falling in w (rejecting H 0 ) when H 0 is true: P(X w H 0 ) = α. H 0 TRUE H 1 TRUE

More information

2/2/2015 GEOGRAPHY 204: STATISTICAL PROBLEM SOLVING IN GEOGRAPHY MEASURES OF CENTRAL TENDENCY CHAPTER 3: DESCRIPTIVE STATISTICS AND GRAPHICS

2/2/2015 GEOGRAPHY 204: STATISTICAL PROBLEM SOLVING IN GEOGRAPHY MEASURES OF CENTRAL TENDENCY CHAPTER 3: DESCRIPTIVE STATISTICS AND GRAPHICS Spring 2015: Lembo GEOGRAPHY 204: STATISTICAL PROBLEM SOLVING IN GEOGRAPHY CHAPTER 3: DESCRIPTIVE STATISTICS AND GRAPHICS Descriptive statistics concise and easily understood summary of data set characteristics

More information

Modeling Uncertainty in the Earth Sciences Jef Caers Stanford University

Modeling Uncertainty in the Earth Sciences Jef Caers Stanford University Probability theory and statistical analysis: a review Modeling Uncertainty in the Earth Sciences Jef Caers Stanford University Concepts assumed known Histograms, mean, median, spread, quantiles Probability,

More information

Part 3: Parametric Models

Part 3: Parametric Models Part 3: Parametric Models Matthew Sperrin and Juhyun Park August 19, 2008 1 Introduction There are three main objectives to this section: 1. To introduce the concepts of probability and random variables.

More information

DUE to its practical importance in communications, the

DUE to its practical importance in communications, the IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS II: EXPRESS BRIEFS, VOL. 52, NO. 3, MARCH 2005 149 An Analytical Formulation of Phase Noise of Signals With Gaussian-Distributed Jitter Reza Navid, Student Member,

More information

Confidence Intervals. Confidence interval for sample mean. Confidence interval for sample mean. Confidence interval for sample mean

Confidence Intervals. Confidence interval for sample mean. Confidence interval for sample mean. Confidence interval for sample mean Confidence Intervals Confidence interval for sample mean The CLT tells us: as the sample size n increases, the sample mean is approximately Normal with mean and standard deviation Thus, we have a standard

More information

4 Resampling Methods: The Bootstrap

4 Resampling Methods: The Bootstrap 4 Resampling Methods: The Bootstrap Situation: Let x 1, x 2,..., x n be a SRS of size n taken from a distribution that is unknown. Let θ be a parameter of interest associated with this distribution and

More information

Circuit Theory Prof. S.C. Dutta Roy Department of Electrical Engineering Indian Institute of Technology, Delhi

Circuit Theory Prof. S.C. Dutta Roy Department of Electrical Engineering Indian Institute of Technology, Delhi Circuit Theory Prof. S.C. Dutta Roy Department of Electrical Engineering Indian Institute of Technology, Delhi Lecture - 43 RC and RL Driving Point Synthesis People will also have to be told I will tell,

More information

FOR beamforming and emitter localization applications in

FOR beamforming and emitter localization applications in IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 17, NO. 11, NOVEMBER 1999 1829 Angle and Time of Arrival Statistics for Circular and Elliptical Scattering Models Richard B. Ertel and Jeffrey H.

More information

An Introduction to Parameter Estimation

An Introduction to Parameter Estimation Introduction Introduction to Econometrics An Introduction to Parameter Estimation This document combines several important econometric foundations and corresponds to other documents such as the Introduction

More information

10.1 The Formal Model

10.1 The Formal Model 67577 Intro. to Machine Learning Fall semester, 2008/9 Lecture 10: The Formal (PAC) Learning Model Lecturer: Amnon Shashua Scribe: Amnon Shashua 1 We have see so far algorithms that explicitly estimate

More information

An interval estimator of a parameter θ is of the form θl < θ < θu at a

An interval estimator of a parameter θ is of the form θl < θ < θu at a Chapter 7 of Devore CONFIDENCE INTERVAL ESTIMATORS An interval estimator of a parameter θ is of the form θl < θ < θu at a confidence pr (or a confidence coefficient) of 1 α. When θl =, < θ < θu is called

More information

Performance Comparison of K-Means and Expectation Maximization with Gaussian Mixture Models for Clustering EE6540 Final Project

Performance Comparison of K-Means and Expectation Maximization with Gaussian Mixture Models for Clustering EE6540 Final Project Performance Comparison of K-Means and Expectation Maximization with Gaussian Mixture Models for Clustering EE6540 Final Project Devin Cornell & Sushruth Sastry May 2015 1 Abstract In this article, we explore

More information

Relating Graph to Matlab

Relating Graph to Matlab There are two related course documents on the web Probability and Statistics Review -should be read by people without statistics background and it is helpful as a review for those with prior statistics

More information

Chapter 4. Displaying and Summarizing. Quantitative Data

Chapter 4. Displaying and Summarizing. Quantitative Data STAT 141 Introduction to Statistics Chapter 4 Displaying and Summarizing Quantitative Data Bin Zou (bzou@ualberta.ca) STAT 141 University of Alberta Winter 2015 1 / 31 4.1 Histograms 1 We divide the range

More information

MATHEMATICAL TOOLS FOR DIGITAL TRANSMISSION ANALYSIS

MATHEMATICAL TOOLS FOR DIGITAL TRANSMISSION ANALYSIS ch03.qxd 1/9/03 09:14 AM Page 35 CHAPTER 3 MATHEMATICAL TOOLS FOR DIGITAL TRANSMISSION ANALYSIS 3.1 INTRODUCTION The study of digital wireless transmission is in large measure the study of (a) the conversion

More information

Continuous Random Variables. What continuous random variables are and how to use them. I can give a definition of a continuous random variable.

Continuous Random Variables. What continuous random variables are and how to use them. I can give a definition of a continuous random variable. Continuous Random Variables Today we are learning... What continuous random variables are and how to use them. I will know if I have been successful if... I can give a definition of a continuous random

More information

Time Series 3. Robert Almgren. Sept. 28, 2009

Time Series 3. Robert Almgren. Sept. 28, 2009 Time Series 3 Robert Almgren Sept. 28, 2009 Last time we discussed two main categories of linear models, and their combination. Here w t denotes a white noise: a stationary process with E w t ) = 0, E

More information

Model Calibration under Uncertainty: Matching Distribution Information

Model Calibration under Uncertainty: Matching Distribution Information Model Calibration under Uncertainty: Matching Distribution Information Laura P. Swiler, Brian M. Adams, and Michael S. Eldred September 11, 008 AIAA Multidisciplinary Analysis and Optimization Conference

More information

Probability Distributions for Continuous Variables. Probability Distributions for Continuous Variables

Probability Distributions for Continuous Variables. Probability Distributions for Continuous Variables Probability Distributions for Continuous Variables Probability Distributions for Continuous Variables Let X = lake depth at a randomly chosen point on lake surface If we draw the histogram so that the

More information

Comparative Distributions of Hazard Modeling Analysis

Comparative Distributions of Hazard Modeling Analysis Comparative s of Hazard Modeling Analysis Rana Abdul Wajid Professor and Director Center for Statistics Lahore School of Economics Lahore E-mail: drrana@lse.edu.pk M. Shuaib Khan Department of Statistics

More information

Sampling Distribution Models. Chapter 17

Sampling Distribution Models. Chapter 17 Sampling Distribution Models Chapter 17 Objectives: 1. Sampling Distribution Model 2. Sampling Variability (sampling error) 3. Sampling Distribution Model for a Proportion 4. Central Limit Theorem 5. Sampling

More information

SDS 321: Introduction to Probability and Statistics

SDS 321: Introduction to Probability and Statistics SDS 321: Introduction to Probability and Statistics Lecture 14: Continuous random variables Purnamrita Sarkar Department of Statistics and Data Science The University of Texas at Austin www.cs.cmu.edu/

More information

Probability Distribution

Probability Distribution Economic Risk and Decision Analysis for Oil and Gas Industry CE81.98 School of Engineering and Technology Asian Institute of Technology January Semester Presented by Dr. Thitisak Boonpramote Department

More information

Load-Strength Interference

Load-Strength Interference Load-Strength Interference Loads vary, strengths vary, and reliability usually declines for mechanical systems, electronic systems, and electrical systems. The cause of failures is a load-strength interference

More information

ECE 5615/4615 Computer Project

ECE 5615/4615 Computer Project Set #1p Due Friday March 17, 017 ECE 5615/4615 Computer Project The details of this first computer project are described below. This being a form of take-home exam means that each person is to do his/her

More information

in the company. Hence, we need to collect a sample which is representative of the entire population. In order for the sample to faithfully represent t

in the company. Hence, we need to collect a sample which is representative of the entire population. In order for the sample to faithfully represent t 10.001: Data Visualization and Elementary Statistical Analysis R. Sureshkumar January 15, 1997 Statistics deals with the collection and the analysis of data in the presence of variability. Variability

More information

IMPERIAL COLLEGE OF SCIENCE, TECHNOLOGY AND MEDICINE UNIVERSITY OF LONDON DEPARTMENT OF ELECTRICAL AND ELECTRONIC ENGINEERING EXAMINATIONS 2010

IMPERIAL COLLEGE OF SCIENCE, TECHNOLOGY AND MEDICINE UNIVERSITY OF LONDON DEPARTMENT OF ELECTRICAL AND ELECTRONIC ENGINEERING EXAMINATIONS 2010 Paper Number(s): E1.1 IMPERIAL COLLEGE OF SCIENCE, TECHNOLOGY AND MEDICINE UNIVERSITY OF LONDON DEPARTMENT OF ELECTRICAL AND ELECTRONIC ENGINEERING EXAMINATIONS 2010 EEE/ISE PART I: MEng, BEng and ACGI

More information

Modules 1-2 are background; they are the same for regression analysis and time series.

Modules 1-2 are background; they are the same for regression analysis and time series. Regression Analysis, Module 1: Regression models (The attached PDF file has better formatting.) Required reading: Chapter 1, pages 3 13 (until appendix 1.1). Updated: May 23, 2005 Modules 1-2 are background;

More information

Guidelines for comparing boxplots

Guidelines for comparing boxplots Comparing Data Sets Project IMP I Name When using boxplots to compare two or more batches of data, it is usually best to compare individual features in a methodical way. You may find the following guidelines

More information

2. If the values for f(x) can be made as close as we like to L by choosing arbitrarily large. lim

2. If the values for f(x) can be made as close as we like to L by choosing arbitrarily large. lim Limits at Infinity and Horizontal Asymptotes As we prepare to practice graphing functions, we should consider one last piece of information about a function that will be helpful in drawing its graph the

More information

Discrete Distributions

Discrete Distributions Discrete Distributions STA 281 Fall 2011 1 Introduction Previously we defined a random variable to be an experiment with numerical outcomes. Often different random variables are related in that they have

More information

Continuous Random Variables

Continuous Random Variables MATH 38 Continuous Random Variables Dr. Neal, WKU Throughout, let Ω be a sample space with a defined probability measure P. Definition. A continuous random variable is a real-valued function X defined

More information

0, otherwise, (a) Find the value of c that makes this a valid pdf. (b) Find P (Y < 5) and P (Y 5). (c) Find the mean death time.

0, otherwise, (a) Find the value of c that makes this a valid pdf. (b) Find P (Y < 5) and P (Y 5). (c) Find the mean death time. 1. In a toxicology experiment, Y denotes the death time (in minutes) for a single rat treated with a toxin. The probability density function (pdf) for Y is given by cye y/4, y > 0 (a) Find the value of

More information

Statistical Quality Control - Stat 3081

Statistical Quality Control - Stat 3081 Statistical Quality Control - Stat 3081 Awol S. Department of Statistics College of Computing & Informatics Haramaya University Dire Dawa, Ethiopia March 2015 Introduction Industrial Statistics and Quality

More information

Week 1 Quantitative Analysis of Financial Markets Distributions A

Week 1 Quantitative Analysis of Financial Markets Distributions A Week 1 Quantitative Analysis of Financial Markets Distributions A Christopher Ting http://www.mysmu.edu/faculty/christophert/ Christopher Ting : christopherting@smu.edu.sg : 6828 0364 : LKCSB 5036 October

More information

Error Analysis in Experimental Physical Science Mini-Version

Error Analysis in Experimental Physical Science Mini-Version Error Analysis in Experimental Physical Science Mini-Version by David Harrison and Jason Harlow Last updated July 13, 2012 by Jason Harlow. Original version written by David M. Harrison, Department of

More information

Lecture 2: Review of Probability

Lecture 2: Review of Probability Lecture 2: Review of Probability Zheng Tian Contents 1 Random Variables and Probability Distributions 2 1.1 Defining probabilities and random variables..................... 2 1.2 Probability distributions................................

More information

Chapter 5. Chapter 5 sections

Chapter 5. Chapter 5 sections 1 / 43 sections Discrete univariate distributions: 5.2 Bernoulli and Binomial distributions Just skim 5.3 Hypergeometric distributions 5.4 Poisson distributions Just skim 5.5 Negative Binomial distributions

More information

F X (x) = P [X x] = x f X (t)dt. 42 Lebesgue-a.e, to be exact 43 More specifically, if g = f Lebesgue-a.e., then g is also a pdf for X.

F X (x) = P [X x] = x f X (t)dt. 42 Lebesgue-a.e, to be exact 43 More specifically, if g = f Lebesgue-a.e., then g is also a pdf for X. 10.2 Properties of PDF and CDF for Continuous Random Variables 10.18. The pdf f X is determined only almost everywhere 42. That is, given a pdf f for a random variable X, if we construct a function g by

More information

Horizontal asymptotes

Horizontal asymptotes Roberto s Notes on Differential Calculus Chapter : Limits and continuity Section 5 Limits at infinity and Horizontal asymptotes What you need to know already: The concept, notation and terminology of its.

More information

Statistical Quality Control - Stat 3081

Statistical Quality Control - Stat 3081 Statistical Quality Control - Stat 3081 Awol S. Department of Statistics College of Computing & Informatics Haramaya University Dire Dawa, Ethiopia March 2015 Introduction Industrial Statistics and Quality

More information

LIKELIHOOD RECEIVER FOR FH-MFSK MOBILE RADIO*

LIKELIHOOD RECEIVER FOR FH-MFSK MOBILE RADIO* LIKELIHOOD RECEIVER FOR FH-MFSK MOBILE RADIO* Item Type text; Proceedings Authors Viswanathan, R.; S.C. Gupta Publisher International Foundation for Telemetering Journal International Telemetering Conference

More information

Introduction to Probability

Introduction to Probability LECTURE NOTES Course 6.041-6.431 M.I.T. FALL 2000 Introduction to Probability Dimitri P. Bertsekas and John N. Tsitsiklis Professors of Electrical Engineering and Computer Science Massachusetts Institute

More information

401 Review. 6. Power analysis for one/two-sample hypothesis tests and for correlation analysis.

401 Review. 6. Power analysis for one/two-sample hypothesis tests and for correlation analysis. 401 Review Major topics of the course 1. Univariate analysis 2. Bivariate analysis 3. Simple linear regression 4. Linear algebra 5. Multiple regression analysis Major analysis methods 1. Graphical analysis

More information

The Growth of Functions. A Practical Introduction with as Little Theory as possible

The Growth of Functions. A Practical Introduction with as Little Theory as possible The Growth of Functions A Practical Introduction with as Little Theory as possible Complexity of Algorithms (1) Before we talk about the growth of functions and the concept of order, let s discuss why

More information

Homework 4 Solutions, 2/2/7

Homework 4 Solutions, 2/2/7 Homework 4 Solutions, 2/2/7 Question Given that the number a is such that the following limit L exists, determine a and L: x 3 a L x 3 x 2 7x + 2. We notice that the denominator x 2 7x + 2 factorizes as

More information

Some Practice Questions for Test 2

Some Practice Questions for Test 2 ENGI 441 Probability and Statistics Faculty of Engineering and Applied Science Some Practice Questions for Test 1. The probability mass function for X = the number of major defects in a randomly selected

More information

Investigation of Possible Biases in Tau Neutrino Mass Limits

Investigation of Possible Biases in Tau Neutrino Mass Limits Investigation of Possible Biases in Tau Neutrino Mass Limits Kyle Armour Departments of Physics and Mathematics, University of California, San Diego, La Jolla, CA 92093 (Dated: August 8, 2003) We study

More information

Chapter 6. Estimation of Confidence Intervals for Nodal Maximum Power Consumption per Customer

Chapter 6. Estimation of Confidence Intervals for Nodal Maximum Power Consumption per Customer Chapter 6 Estimation of Confidence Intervals for Nodal Maximum Power Consumption per Customer The aim of this chapter is to calculate confidence intervals for the maximum power consumption per customer

More information

New Interpretation of Principal Components Analysis

New Interpretation of Principal Components Analysis Zeszyty Naukowe WWSI, No 16, Vol 11, 2017, pp 43-65 New Interpretation of Principal Components Analysis Zenon Gniazdowski * Warsaw School of Computer Science Abstract A new look on the principal component

More information

Chapter 6. Random Processes

Chapter 6. Random Processes Chapter 6 Random Processes Random Process A random process is a time-varying function that assigns the outcome of a random experiment to each time instant: X(t). For a fixed (sample path): a random process

More information

The Not-Formula Book for C2 Everything you need to know for Core 2 that won t be in the formula book Examination Board: AQA

The Not-Formula Book for C2 Everything you need to know for Core 2 that won t be in the formula book Examination Board: AQA Not The Not-Formula Book for C Everything you need to know for Core that won t be in the formula book Examination Board: AQA Brief This document is intended as an aid for revision. Although it includes

More information

Joint Probability Distributions and Random Samples (Devore Chapter Five)

Joint Probability Distributions and Random Samples (Devore Chapter Five) Joint Probability Distributions and Random Samples (Devore Chapter Five) 1016-345-01: Probability and Statistics for Engineers Spring 2013 Contents 1 Joint Probability Distributions 2 1.1 Two Discrete

More information

Fading Statistical description of the wireless channel

Fading Statistical description of the wireless channel Channel Modelling ETIM10 Lecture no: 3 Fading Statistical description of the wireless channel Fredrik Tufvesson Department of Electrical and Information Technology Lund University, Sweden Fredrik.Tufvesson@eit.lth.se

More information

CHAPTER 6 SOME CONTINUOUS PROBABILITY DISTRIBUTIONS. 6.2 Normal Distribution. 6.1 Continuous Uniform Distribution

CHAPTER 6 SOME CONTINUOUS PROBABILITY DISTRIBUTIONS. 6.2 Normal Distribution. 6.1 Continuous Uniform Distribution CHAPTER 6 SOME CONTINUOUS PROBABILITY DISTRIBUTIONS Recall that a continuous random variable X is a random variable that takes all values in an interval or a set of intervals. The distribution of a continuous

More information

IAM 530 ELEMENTS OF PROBABILITY AND STATISTICS LECTURE 3-RANDOM VARIABLES

IAM 530 ELEMENTS OF PROBABILITY AND STATISTICS LECTURE 3-RANDOM VARIABLES IAM 530 ELEMENTS OF PROBABILITY AND STATISTICS LECTURE 3-RANDOM VARIABLES VARIABLE Studying the behavior of random variables, and more importantly functions of random variables is essential for both the

More information

The Normal Distribution. The Gaussian Curve. Advantages of using Z-score. Importance of normal or Gaussian distribution (ND)

The Normal Distribution. The Gaussian Curve. Advantages of using Z-score. Importance of normal or Gaussian distribution (ND) Importance of normal or Gaussian distribution (ND) The Normal It is the most used distribution Most method are based on the assumption of ND Sum of many independent, random contributions variables (grain

More information

Tutorial 4. Fast Fourier Transforms Phase factors runs around on the unit circle contains Wave forms

Tutorial 4. Fast Fourier Transforms Phase factors runs around on the unit circle contains Wave forms Tutorial 4. Fast Fourier Transforms Phase factors There are functions that produce roots-of-one as a function of time (t) or place (x). A good example is a Bloch function φ(x) = exp(i kx) or the phase

More information

Since x + we get x² + 2x = 4, or simplifying it, x² = 4. Therefore, x² + = 4 2 = 2. Ans. (C)

Since x + we get x² + 2x = 4, or simplifying it, x² = 4. Therefore, x² + = 4 2 = 2. Ans. (C) SAT II - Math Level 2 Test #01 Solution 1. x + = 2, then x² + = Since x + = 2, by squaring both side of the equation, (A) - (B) 0 (C) 2 (D) 4 (E) -2 we get x² + 2x 1 + 1 = 4, or simplifying it, x² + 2

More information

Statistical Methods: Introduction, Applications, Histograms, Ch

Statistical Methods: Introduction, Applications, Histograms, Ch Outlines Statistical Methods: Introduction, Applications, Histograms, Characteristics November 4, 2004 Outlines Part I: Statistical Methods: Introduction and Applications Part II: Statistical Methods:

More information

Chapter Learning Objectives. Probability Distributions and Probability Density Functions. Continuous Random Variables

Chapter Learning Objectives. Probability Distributions and Probability Density Functions. Continuous Random Variables Chapter 4: Continuous Random Variables and Probability s 4-1 Continuous Random Variables 4-2 Probability s and Probability Density Functions 4-3 Cumulative Functions 4-4 Mean and Variance of a Continuous

More information

Statistics 100A Homework 5 Solutions

Statistics 100A Homework 5 Solutions Chapter 5 Statistics 1A Homework 5 Solutions Ryan Rosario 1. Let X be a random variable with probability density function a What is the value of c? fx { c1 x 1 < x < 1 otherwise We know that for fx to

More information

Frequency Dependent Aspects of Op-amps

Frequency Dependent Aspects of Op-amps Frequency Dependent Aspects of Op-amps Frequency dependent feedback circuits The arguments that lead to expressions describing the circuit gain of inverting and non-inverting amplifier circuits with resistive

More information

Self-healing tile sets: rates of regrowth in damaged assemblies

Self-healing tile sets: rates of regrowth in damaged assemblies University of Arkansas, Fayetteville ScholarWorks@UARK Mathematical Sciences Undergraduate Honors Theses Mathematical Sciences 5-2015 Self-healing tile sets: rates of regrowth in damaged assemblies Kevin

More information

REAL-TIME TIME-FREQUENCY BASED BLIND SOURCE SEPARATION. Scott Rickard, Radu Balan, Justinian Rosca. Siemens Corporate Research Princeton, NJ 08540

REAL-TIME TIME-FREQUENCY BASED BLIND SOURCE SEPARATION. Scott Rickard, Radu Balan, Justinian Rosca. Siemens Corporate Research Princeton, NJ 08540 REAL-TIME TIME-FREQUENCY BASED BLIND SOURCE SEPARATION Scott Rickard, Radu Balan, Justinian Rosca Siemens Corporate Research Princeton, NJ 84 fscott.rickard,radu.balan,justinian.roscag@scr.siemens.com

More information