Probability Distributions

Slides:



Advertisements
Similar presentations
Pattern Recognition and Machine Learning
Advertisements

Modeling of Data. Basic Bayes theorem Bayes theorem relates the conditional probabilities of two events A, and B: A might be a hypothesis and B might.
Introduction to Non Parametric Statistics Kernel Density Estimation.
JMB Chapter 6 Part 1 v4 EGR 252 Spring 2012 Slide 1 Continuous Probability Distributions Many continuous probability distributions, including: Uniform.
Kin 304 Regression Linear Regression Least Sum of Squares
Lecture Notes for E Alpaydın 2010 Introduction to Machine Learning 2e © The MIT Press (V1.0) ETHEM ALPAYDIN © The MIT Press, 2010
Hydrologic Statistics
Ch11 Curve Fitting Dr. Deshi Ye
Resampling techniques Why resampling? Jacknife Cross-validation Bootstrap Examples of application of bootstrap.
Chapter 10 Simple Regression.
A gentle introduction to Gaussian distribution. Review Random variable Coin flip experiment X = 0X = 1 X: Random variable.
Descriptive statistics Experiment  Data  Sample Statistics Experiment  Data  Sample Statistics Sample mean Sample mean Sample variance Sample variance.
Continuous Random Variables Chap. 12. COMP 5340/6340 Continuous Random Variables2 Preamble Continuous probability distribution are not related to specific.
CSE 221: Probabilistic Analysis of Computer Systems Topics covered: Continuous random variables Uniform and Normal distribution (Sec. 3.1, )
Statistical analysis and modeling of neural data Lecture 5 Bijan Pesaran 19 Sept, 2007.
Probability & Statistics for Engineers & Scientists, by Walpole, Myers, Myers & Ye ~ Chapter 11 Notes Class notes for ISE 201 San Jose State University.
The moment generating function of random variable X is given by Moment generating function.
Lecture II-2: Probability Review
Normal and Sampling Distributions A normal distribution is uniquely determined by its mean, , and variance,  2 The random variable Z = (X-  /  is.
Hydrologic Statistics
Binary Variables (1) Coin flipping: heads=1, tails=0 Bernoulli Distribution.
Moment Generating Functions 1/33. Contents Review of Continuous Distribution Functions 2/33.
Physics 114: Lecture 15 Probability Tests & Linear Fitting Dale E. Gary NJIT Physics Department.
Introduction and Motivation Approaches for DE: Known model → parametric approach: p(x;θ) (Gaussian, Laplace,…) Unknown model → nonparametric approach Assumes.
Moment Generating Functions
Statistics & Flood Frequency Chapter 3 Dr. Philip B. Bedient Rice University 2006.
Sampling Distributions & Standard Error Lesson 7.
Copyright © 2010 Pearson Addison-Wesley. All rights reserved. Chapter 6 Some Continuous Probability Distributions.
Regression Regression relationship = trend + scatter
Chapter 20 Statistical Considerations Lecture Slides The McGraw-Hill Companies © 2012.
Chapter 31Introduction to Statistical Quality Control, 7th Edition by Douglas C. Montgomery. Copyright (c) 2012 John Wiley & Sons, Inc.
Session V: The Normal Distribution Continuous Distributions (Zar, Chapter 6)
Unit 4 Review. Starter Write the characteristics of the binomial setting. What is the difference between the binomial setting and the geometric setting?
Statistics 350 Lecture 2. Today Last Day: Section Today: Section 1.6 Homework #1: Chapter 1 Problems (page 33-38): 2, 5, 6, 7, 22, 26, 33, 34,
Selecting Input Probability Distributions. 2 Introduction Part of modeling—what input probability distributions to use as input to simulation for: –Interarrival.
Statistics and probability Dr. Khaled Ismael Almghari Phone No:
Chapter 15: Exploratory data analysis: graphical summaries CIS 3033.
Normalizing Transformations and fitting a marginal distribution
Modeling and Simulation CS 313
The simple linear regression model and parameter estimation
Concepts in Probability, Statistics and Stochastic Modeling
Physics 114: Lecture 13 Probability Tests & Linear Fitting
Regression Analysis AGEC 784.
Why Stochastic Hydrology ?
Probability Theory and Parameter Estimation I
Basic Estimation Techniques
Modeling and Simulation CS 313
Random Variables and their Properties
Dept. Computer Science & Engineering, Shanghai Jiao Tong University
Models to Represent the Relationships Between Variables (Regression)
Statistical Hydrology and Flood Frequency
Nonparametric Density Estimation
BIVARIATE REGRESSION AND CORRELATION
Bias and Variance of the Estimator
Simple Linear Regression - Introduction
Multiple Random Variables
Basic Estimation Techniques
CHAPTER 29: Multiple Regression*
Alafia river: Autocorrelation Autocorrelation of standardized flow.
Regression Models - Introduction
Hydrologic Statistics
Undergraduated Econometrics
Econometrics I Professor William Greene Stern School of Business
Statistics & Flood Frequency Chapter 3
Chapter 6 Some Continuous Probability Distributions.
6.3 Sampling Distributions
Continuous Probability Distributions
Moments of Random Variables
Regression Models - Introduction
Presentation transcript:

Probability Distributions Learn the common parametric probability distributions that are used to represent the randomness of physical phenomena Learn the physical mechanisms that underlie common probability distributions Apply physical knowledge about the mechanisms that lead to randomness in a process to select an appropriate probability distribution to fit Distribution selection is more than selecting the best fit empirical curve Reading Kottegoda and Rosso Chapter 4.

Kottegoda and Rosso page 201

Kottegoda and Rosso page 228

From http://www.weibull.com/

Carl Friedrich Gauß, immortalized

Normal Distribution Standardized Cumulative Use Normal tables or normcdf, norminv

Linear combinations of normally distributed variables Regenerative Central Limit Theorem for n→ The normal distribution is the natural distribution to characterize any process where the randomness is due to the additive effect of many independent inputs, without domination by one specific input.

Log Normal Distribution X is log normally distributed if Y=ln(X) is normally distributed Derived distribution Parameters y, y, pertain to the ln of the variable.

Lognormal and Normal distribution relationships used for simulation Moments Coefficient of Variation Estimation of parameters from moments

Log normal distribution and central limit theorem Parameters , , pertain to the ln of the variable, the Y’s. The log normal distribution is the natural distribution to characterize any process where the randomness is due to the multiplicative effect of many independent inputs, without domination by one specific input. (If one Xi can be 0 it makes the product 0, dominating the resulting distribution)

Nonparametric Density Estimation Learn the kernel method for nonparametric density estimation Learn the difference between fixed kernel window width and variable kernel window width estimates Learn some of the approaches to kernel window width estimation Reference Least squares cross validation Reading Silverman Chapter 1, Chapter 2 (2.1-2.6, 2.9), Chapter 3 (3.1-3.4)

Parametric vs Nonparametric Assume data from a known distribution (e.g. Gamma, Normal) and estimate parameters Nonparametric Assume that the data has a probability density function but not of a specific known form Let data speak for themselves Exploratory data analysis

Which method conveys the information best to you ? Cumulative Density Probability Density Equation Probability Plot

San Juan River Annual Streamflow 1906-1983 Histogram, bin width 150 Flow KAF Number

San Juan River Annual Streamflow 1906-1983 Density Flow KAF

Kernel Density Estimator

Kernel Density Estimate (KDE) Place “kernels” at each data point Sum up the kernels Width of kernel determines level of smoothing Determining how to choose the width of the kernel is an important topic! Narrow kernel Medium kernel Wide kernel

KDE window width sensitivity

Variable Kernel

San Juan River Annual Streamflow 1906-1983, Variable Kernel Density Estimate Flow KAF

Some Common Kernels

Density estimate with different kernels with the same bandwidth

Least Squares Cross Validation Density Mo(h) Flow KAF Window width h