Lecture 2 Probability and Measurement Error, Part 1

1 / 68

# Lecture 2 Probability and Measurement Error, Part 1 - PowerPoint PPT Presentation

Lecture 2 Probability and Measurement Error, Part 1. Syllabus.

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.

## PowerPoint Slideshow about 'Lecture 2 Probability and Measurement Error, Part 1' - gaura

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

### Lecture 2 Probability and Measurement Error, Part 1

Syllabus

Lecture 01 Describing Inverse ProblemsLecture 02 Probability and Measurement Error, Part 1Lecture 03 Probability and Measurement Error, Part 2 Lecture 04 The L2 Norm and Simple Least SquaresLecture 05 A Priori Information and Weighted Least SquaredLecture 06 Resolution and Generalized Inverses

Lecture 07 Backus-Gilbert Inverse and the Trade Off of Resolution and VarianceLecture 08 The Principle of Maximum LikelihoodLecture 09 Inexact TheoriesLecture 10 Nonuniqueness and Localized AveragesLecture 11 Vector Spaces and Singular Value Decomposition

Lecture 12 Equality and Inequality ConstraintsLecture 13 L1 , L∞ Norm Problems and Linear ProgrammingLecture 14 Nonlinear Problems: Grid and Monte Carlo Searches Lecture 15 Nonlinear Problems: Newton’s Method Lecture 16 Nonlinear Problems: Simulated Annealing and Bootstrap Confidence Intervals Lecture 17 Factor AnalysisLecture 18 Varimax Factors, Empircal Orthogonal FunctionsLecture 19 Backus-Gilbert Theory for Continuous Problems; Radon’s ProblemLecture 20 Linear Operators and Their AdjointsLecture 21 Fréchet DerivativesLecture 22 Exemplary Inverse Problems, incl. Filter DesignLecture 23 Exemplary Inverse Problems, incl. Earthquake LocationLecture 24 Exemplary Inverse Problems, incl. Vibrational Problems

Purpose of the Lecture

review random variables and their probability density functions

introduce correlation and the multivariate Gaussian distribution

relate error propagation to functions of random variables

random variable, d

no fixed value until it is realized

d=?

d=?

d=1.04

d=0.98

indeterminate

indeterminate

random variables have systematics

tendency to takes on some values more often than others

Nrealization of data

index, i

10

di

5

0

(A)

(B)

(C)

in general probability is the integral

probability that d is between d1 and d2

the probability that d has some value is 100% or unity

probability that d is between its minimum and maximum bounds, dmin and dmax

Summarizing a probability density function

typical value

“center of the p.d.f.”

amount of scatter around the typical value

“width of the p.d.f.”

Several possibilities for a typical value

p(d)

d

dML

point beneath the peak or “maximum likelihood point” or “mode”

p(d)

50%

50%

d

dmedian

point dividing area in half or “median”

p(d)

d

<d>

balancing point or “mean” or “expectation”

can all be different

• dML≠ dmedian ≠ <d>

formula for

“mean” or “expected value”

<d>

step 1: usual formula for mean

<d>

d

data

step 2: replace data with its histogram

Ns

<d>

ds

s

histogram

step 3: replace histogram with probability distribution.

Ns

<d>

s

N

p

P(ds)

s

ds

probability distribution

P(ds)

<d>

s

<d>

This function grows away from the typical value:

• q(d) = (d-<d>)2

so the function q(d)p(d) is

small if most of the area is near <d> ,that is, a narrowp(d)

• large if most of the area is far from <d> , that is, a wide p(d)
• so quantify width as the area under q(d)p(d)

(A)

(B)

(C)

(D)

(E)

(F)

variance

mean

width is actually square root of variance, that is, σ

estimating mean and variance from data

usual formula for square of “sample standard deviation”

usual formula for “sample mean”

MabLab scripts for mean and variance

from tabulated p.d.f. p

dbar = Dd*sum(d.*p);

q = (d-dbar).^2;

sigma2 = Dd*sum(q.*p);

sigma = sqrt(sigma2);

from realizations of data

dbar = mean(dr);

sigma = std(dr);

sigma2 = sigma^2;

uniform p.d.f.

p(d)

box-shaped function

• 1/(dmax-dmin)

d

dmin

dmax

probability is the same everywhere in the range of possible values

Gaussian (or “Normal”) p.d.f.

bell-shaped function

d

Large probability near the mean, d. Variance is σ2.

uncorrelated random variables

no pattern of between values of one variable and values of another

when d1 is higher than its mean

d2 is higher or lower than its mean

with equal probability

<d2 >

0

10

p

d2

0

0.2

<d1 >

0.0

10

d1

no tendency for d2 to be either high or low when d1 is high

in uncorrelated case

jointp.d.f.is just the product of individualp.d.f.’s

<d2 >

10

0

p

d2

0.25

0

<d1 >

10

0.00

2σ1

d1

<d2 >

10

0

p

d2

0.25

0

<d1 >

10

0.00

2σ1

d1

tendency for d2to be high when d1 is high

<d2 >

10

0

p

d2

0.25

0

θ

2σ1

<d1 >

2σ2

10

0.00

2σ1

d1

<d2>

d2

<d1>

2σ1

d1

(A)

(B)

(C)

<d2>

<d2>

<d2>

d2

d2

d2

<d1>

<d1>

<d1>

d1

d1

d1

formula for covariance

+ positive correlation high d1 high d2

• - negative correlation high d1 low d2
joint p.d.f.mean is a vectorcovariance is a symmetric matrix

diagonal elements: variances

off-diagonal elements: covariances

estimating covariance from a table D of data

Dki: realization k of data-type i

in MatLab, C=cov(D)

univariatep.d.f. formed from joint p.d.f.

p(d) → p(di)

behavior of di irrespective of the other ds

integrate over everything but di

d2

p(d1,d2)

p(d1)

integrate over d2

d1

d1

integrate over d1

p(d2)

d2

mean

covariance matrix

functions of random variables

data with measurement error

inferences with uncertainty

data analysis process

functions of random variables

given p(d)

with m=f(d)

what is p(m)?

univariatep.d.f.use the chair rule

rule for transforming a univariatep.d.f.

multivariate p.d.f.
• Jacobian
• determinant
• determinant of matrix with elements ∂di/∂mj
multivariate p.d.f.

Jacobian determinant

rule for transforming a multivariate p.d.f.

simple example

data with measurement error

inferences with uncertainty

data analysis process

one datum, d

uniform p.d.f.

0<d<1

one model parameter, m

m = 2d

m=2d and d=½m

d=0 → m=0

d=1 → m=2

dd/dm = ½

p(d ) =1

• p(m ) = ½

p(d)

p(m)

1

1

0

0

d

m

1

1

2

2

0

0

another simple example

data with measurement error

inferences with uncertainty

data analysis process

one datum, d

uniform p.d.f.

0<d<1

one model parameter, m

m = d2

m=d2and d=m½

d=0 → m=0

d=1 → m=1

dd/dm = ½m-½

p(d ) =1

• p(m ) = ½ m-½

A)

2.0

p(d)

1.0

0.0

0.2

0.4

0.6

0.8

1.0

d

B)

2.0

p(m)

1.0

0.0

0.2

0.4

0.6

0.8

1.0

m

multivariate example

data with measurement error

inferences with uncertainty

data analysis process

2 data, d1 , d2

uniform p.d.f.

0<d<1

for both

• 2 model parameters, m1andm2

m1 = d1+d2

m2 = d2-d2

(A)

m2=d1-d1

-1

-1

1

d2

1

m1=d1+d1

d1

m=Mdwith

and|det(M)|=2

• d=M-1m so∂d/∂ m = M-1 and J= |det(M-1)|= ½
• p(d) = 1
• p(m) = ½

(C)

(A)

m2=d1-d1

-1

-1

1

-1

p(d1)

1

0

d2

1

1

m1=d1+d1

d1

d1

(B)

(D)

-1

-1

Note that the shape in m is different than the shape in d

1

-1

0.5

p(m1)

m2

0

1

1

2

2

m1

m1

(C)

(A)

m2=d1-d1

-1

-1

1

-1

p(d1)

1

0

d2

1

1

m1=d1+d1

d1

d1

(B)

(D)

-1

-1

• The shape is a square with sides of length √2. The amplitude is ½. So the area is ½⨉√2⨉√2 = 1

1

-1

0.5

p(m1)

m2

0

1

1

2

2

m1

m1

(C)

(A)

m2=d1-d1

-1

-1

1

-1

p(d1)

1

0

d2

1

1

m1=d1+d1

d1

d1

(B)

(D)

-1

-1

1

-1

0.5

p(m1)

m2

0

1

1

2

2

m1

m1

moral

p(m) can behavior quite differently from p(d)