Регрессия Excel

Содержание

Слайд 2

08/12/2023

Agenda

Introduction
TGA Example
NLR Basics
Multicollinearity
Prediction
Testing
Bayesian Estimation
Conclusions

08/12/2023 Agenda Introduction TGA Example NLR Basics Multicollinearity Prediction Testing Bayesian Estimation Conclusions

Слайд 3

08/12/2023

1. Introduction

08/12/2023 1. Introduction

Слайд 4

08/12/2023

Linear and Non-linear Regressions

Close relatives?

2

08/12/2023 Linear and Non-linear Regressions Close relatives? 2

Слайд 5

08/12/2023

2. Thermo Gravimetric Analysis Example

Let’s see it!

08/12/2023 2. Thermo Gravimetric Analysis Example Let’s see it!

Слайд 6

08/12/2023

TGA Experiment and Data

TGA Experiment

TGA Data

08/12/2023 TGA Experiment and Data TGA Experiment TGA Data

Слайд 7

08/12/2023

TGA Example Variables

Small size problem!

08/12/2023 TGA Example Variables Small size problem!

Слайд 8

08/12/2023

Plasticizer Evaporation Model

Diffusion is not relevant!

08/12/2023 Plasticizer Evaporation Model Diffusion is not relevant!

Слайд 9

08/12/2023

Fitter Worksheet for TGA Example

08/12/2023 Fitter Worksheet for TGA Example

Слайд 10

08/12/2023

Service Life Prediction by TGA Data

08/12/2023 Service Life Prediction by TGA Data

Слайд 11

08/12/2023

3. NLR Basics

08/12/2023 3. NLR Basics

Слайд 12

08/12/2023

Data and Errors

Weight is an effective instrument!

08/12/2023 Data and Errors Weight is an effective instrument!

Слайд 13

08/12/2023

Model f(x,a)

Presentation at worksheet

08/12/2023 Model f(x,a) Presentation at worksheet

Слайд 14

08/12/2023

Data & Model Prepared for Fitter

Apply Fitter!

08/12/2023 Data & Model Prepared for Fitter Apply Fitter!

Слайд 15

08/12/2023

Objective Function Q(a)

Parameter estimates

Weighted variance estimate

Objective function Q is a sum of squares and

08/12/2023 Objective Function Q(a) Parameter estimates Weighted variance estimate Objective function Q
may be more…

Слайд 16

08/12/2023

Very Important Matrix A

Matrix A is the cause of troubles..

08/12/2023 Very Important Matrix A Matrix A is the cause of troubles..

Слайд 17

08/12/2023

Quality of Estimation

Matrix A is the measure of quality!

08/12/2023 Quality of Estimation Matrix A is the measure of quality!

Слайд 18

08/12/2023

Search by Gradient Method

Matrix A is the
key to search!

08/12/2023 Search by Gradient Method Matrix A is the key to search!

Слайд 19

08/12/2023

4. Multicollinearity

08/12/2023 4. Multicollinearity

Слайд 20

08/12/2023

Multicollinearity: View

Multicollinearity is degradation of matrix A

Objective function Q(a)

1

N(A) =

2

4

5

6

7

08/12/2023 Multicollinearity: View Multicollinearity is degradation of matrix A Objective function Q(a)

Слайд 21

08/12/2023

Multicollinearity: Source

08/12/2023 Multicollinearity: Source

Слайд 22

08/12/2023

Data & Model Preprocessing

((a + b) + c) + d ≠ a

08/12/2023 Data & Model Preprocessing ((a + b) + c) + d
+ (b + (c + d)) as 1+10 –20 = 1

Слайд 23

08/12/2023

Example: The Arrhenius Law

08/12/2023 Example: The Arrhenius Law

Слайд 24

08/12/2023

Derivative Calculation and Precision

08/12/2023 Derivative Calculation and Precision

Слайд 25

08/12/2023

5. Prediction

08/12/2023 5. Prediction

Слайд 26

08/12/2023

Reliable Prediction

Forecast should include uncertainties!

08/12/2023 Reliable Prediction Forecast should include uncertainties!

Слайд 27

08/12/2023

Nonlinearity and Simulation

Non-linear models call for special methods of reliable prediction!

08/12/2023 Nonlinearity and Simulation Non-linear models call for special methods of reliable prediction!

Слайд 28

08/12/2023

Prediction: Example

Accelerated aging tests

Upper confidence limits

Model of rubber aging

08/12/2023 Prediction: Example Accelerated aging tests Upper confidence limits Model of rubber aging

Слайд 29

08/12/2023

6. Testing

08/12/2023 6. Testing

Слайд 30

08/12/2023

Hypotheses Testing

Test statistics ξ is compared with critical value t (α)

Test

08/12/2023 Hypotheses Testing Test statistics ξ is compared with critical value t
don’t prove a model! It just shows that the hypothesis is accepted or rejected!

Слайд 31

08/12/2023

Lack-of-Fit and Variances Tests

These hypotheses are based on variances and they can’t

08/12/2023 Lack-of-Fit and Variances Tests These hypotheses are based on variances and
be tested without replicas!

Lack-of-Fit is a wily test!

Слайд 32

08/12/2023

Outlier and Series Tests

These hypotheses are based on residuals and they can

08/12/2023 Outlier and Series Tests These hypotheses are based on residuals and
be tested without replicas

Series test is very sensitive!

Слайд 33

08/12/2023

7. Bayesian Estimation

08/12/2023 7. Bayesian Estimation

Слайд 34

08/12/2023

Bayesian Estimation

How to eat away an elephant? Slice by slice!

08/12/2023 Bayesian Estimation How to eat away an elephant? Slice by slice!

Слайд 35

08/12/2023

Posterior and Prior Information. Type I

The same error in each portion of data!

08/12/2023 Posterior and Prior Information. Type I The same error in each portion of data!

Слайд 36

08/12/2023

Posterior and Prior Information. Type II

Different errors in
each portion of data!

08/12/2023 Posterior and Prior Information. Type II Different errors in each portion of data!
Имя файла: Регрессия-Excel-.pptx
Количество просмотров: 146
Количество скачиваний: 1