IX. Transient Model Nonlinear Regression and Statistical Analysis.

Slides:



Advertisements
Similar presentations
IX. Transient Model Sensitivity Analysis. Sensitivity Analysis for the Initial Model One-percent sensitivities: Can be explained using principle of superposition.
Advertisements

11-1 Empirical Models Many problems in engineering and science involve exploring the relationships between two or more variables. Regression analysis.
Forecasting Using the Simple Linear Regression Model and Correlation
Inference for Regression
VII. Observation-Parameter Statistics
Simple Linear Regression. Start by exploring the data Construct a scatterplot  Does a linear relationship between variables exist?  Is the relationship.
Regression Analysis Once a linear relationship is defined, the independent variable can be used to forecast the dependent variable. Y ^ = bo + bX bo is.
Descriptive statistics Experiment  Data  Sample Statistics Experiment  Data  Sample Statistics Sample mean Sample mean Sample variance Sample variance.
Chapter Topics Types of Regression Models
The Calibration Process
RLR. Purpose of Regression Fit data to model Known model based on physics P* = exp[A - B/(T+C)] Antoine eq. Assumed correlation y = a + b*x1+c*x2 Use.
Simple Linear Regression Analysis
1 Doing Statistics for Business Doing Statistics for Business Data, Inference, and Decision Making Marilyn K. Pelosi Theresa M. Sandifer Chapter 11 Regression.
Quantify prediction uncertainty (Book, p ) Prediction standard deviations (Book, p. 180): A measure of prediction uncertainty Calculated by translating.
Introduction to Linear Regression and Correlation Analysis
Regression Analysis Regression analysis is a statistical technique that is very useful for exploring the relationships between two or more variables (one.
Inference for regression - Simple linear regression
Calibration Guidelines 1. Start simple, add complexity carefully 2. Use a broad range of information 3. Be well-posed & be comprehensive 4. Include diverse.
Chapter 14 – Correlation and Simple Regression Math 22 Introductory Statistics.
6.1 What is Statistics? Definition: Statistics – science of collecting, analyzing, and interpreting data in such a way that the conclusions can be objectively.
Hydrologic Modeling: Verification, Validation, Calibration, and Sensitivity Analysis Fritz R. Fiedler, P.E., Ph.D.
IV. Sensitivity Analysis for Initial Model 1. Sensitivities and how are they calculated 2. Fit-independent sensitivity-analysis statistics 3. Scaled sensitivities.
Uncertainty Analysis and Model “Validation” or Confidence Building.
Identify Parameters Important to Predictions using PPR & Identify Existing Observation Locations Important to Predictions using OPR.
III. Ground-Water Management Problem Used for the Exercises.
STAT E100 Section Week 3 - Regression. Review  Descriptive Statistics versus Hypothesis Testing  Outliers  Sample vs. Population  Residual Plots.
Analytical vs. Numerical Minimization Each experimental data point, l, has an error, ε l, associated with it ‣ Difference between the experimentally measured.
Using Ground-Water Model Predictions and the ppr and opr Statistics to Guide Data Collection.
VI. Evaluate Model Fit Basic questions that modelers must address are: How well does the model fit the data? Do changes to a model, such as reparameterization,
V. Nonlinear Regression Objective-Function Surfaces Thus far, we have: Parameterized the forward model Obtained head and flow observations and their weights.
2014. Engineers often: Regress data  Analysis  Fit to theory  Data reduction Use the regression of others  Antoine Equation  DIPPR We need to be.
Go to Table of Content Single Variable Regression Farrokh Alemi, Ph.D. Kashif Haqqi M.D.
VIII: Methods for Evaluating Model Predictions 1. Define predictive quantity and calculate sensitivities and standard deviations (Ex8.1a) 2. Assess data.
Lecture 8 Simple Linear Regression (cont.). Section Objectives: Statistical model for linear regression Data for simple linear regression Estimation.
Inference for Regression Simple Linear Regression IPS Chapter 10.1 © 2009 W.H. Freeman and Company.
Chapter 4 Linear Regression 1. Introduction Managerial decisions are often based on the relationship between two or more variables. For example, after.
Multiple Regression Petter Mostad Review: Simple linear regression We define a model where are independent (normally distributed) with equal.
1 Multiple Regression A single numerical response variable, Y. Multiple numerical explanatory variables, X 1, X 2,…, X k.
Calibration Guidelines 1. Start simple, add complexity carefully 2. Use a broad range of information 3. Be well-posed & be comprehensive 4. Include diverse.
Regression Analysis Part C Confidence Intervals and Hypothesis Testing
Simple Linear Regression. Data available : (X,Y) Goal : To predict the response Y. (i.e. to obtain the fitted response function f(X)) Least Squares Fitting.
Review Lecture 51 Tue, Dec 13, Chapter 1 Sections 1.1 – 1.4. Sections 1.1 – 1.4. Be familiar with the language and principles of hypothesis testing.
Scatter Diagrams scatter plot scatter diagram A scatter plot is a graph that may be used to represent the relationship between two variables. Also referred.
9. Testing Model Linearity Modified Beale’s Measure (p ) Total model nonlinearity (p. 144) Intrinsic model nonlinearity (p. 145) For total and.
IX. Transient Forward Modeling. Ground-Water Management Issues Recall the ground-water management issues for the simple flow system considered in the.
Regression Analysis1. 2 INTRODUCTION TO EMPIRICAL MODELS LEAST SQUARES ESTIMATION OF THE PARAMETERS PROPERTIES OF THE LEAST SQUARES ESTIMATORS AND ESTIMATION.
1 Multiple Regression and Correlation KNN Ch. 6 CC Ch 3.
Using the Model to Evaluate Observation Locations and Parameter Information in the Context of Predictions.
AP Statistics Section 15 A. The Regression Model When a scatterplot shows a linear relationship between a quantitative explanatory variable x and a quantitative.
REGRESSION MODELS OF BEST FIT Assess the fit of a function model for bivariate (2 variables) data by plotting and analyzing residuals.
Statistics and probability Dr. Khaled Ismael Almghari Phone No:
AUTOMATED PARAMETER ESTIMATION Model Parameterization, Inverse Modeling, PEST.
Chapter 4: Basic Estimation Techniques
Correlation and Linear Regression
Chapter 4 Basic Estimation Techniques
AP Statistics Chapter 14 Section 1.
Basic Estimation Techniques
ENM 310 Design of Experiments and Regression Analysis
B&A ; and REGRESSION - ANCOVA B&A ; and
The Calibration Process
S519: Evaluation of Information Systems
Basic Estimation Techniques
Statistical Methods For Engineers
CHAPTER 29: Multiple Regression*
Linear Regression.
EQ: How well does the line fit the data?
Descriptive vs. Inferential
Correlation and Regression
11C Line of Best Fit By Eye, 11D Linear Regression
Business Statistics - QBM117
Presentation transcript:

IX. Transient Model Nonlinear Regression and Statistical Analysis

Nonlinear Regression When all K and S parameters are log-transformed, the regression for the transient problem will converge, and optimal estimates of the nine model parameters will be obtained. EXERCISE 9.7: Estimate parameters for the transient system by nonlinear regression.

Evaluate Model Fit Now, we will perform the same analysis of the regression results for the transient problem that was performed for the steady-state problem. EXERCISE 9.8: Evaluate measures of model fit Statistical measures of overall model fit, S, s 2, and s, are shown in Figure 9.13, p. 246.

Evaluate Model Fit EXERCISE 9.9: Use Graphs for Analyzing Model Fit and Evaluate Related Statistics EXERCISE 9.9a: Evaluate graphs of weighted residuals and weighted and unweighted simulated and observed values. See Figure 9.14, p. 247 of Hill and Tiedeman and statistic R in Figure Which graphs are most useful to understanding model fit? Is R helpful?

Weighed Residuals vs. Simulated Values Figure 9.14a of Hill and Tiedeman (page 247)

Weighted Observed Values vs. Weighted Simulated Values Figure 9.14b of Hill and Tiedeman (page 247)

Evaluate Model Fit EXERCISE 9.9b. Evaluate graphs of weighted residuals against independent variables and the runs statistic. The runs statistic is given in Figure 9.16, p EXERCISE 9.9c: Assess independence and normality of the weighted residuals. The normal probability graph and the R N 2 statistic are shown in Figure 9.17, p. 250.

Normal Probability Graph Figure 9.17 of Hill and Tiedeman (page 250)

Evaluate Parameter Estimates EXERCISE 9.10: Evaluate Estimated Parameters EXERCISE 9.10a. Composite scaled sensitivities. EXERCISE 9.10b: Parameter estimates and confidence intervals. EXERCISE 9.10c: Reasonable parameter ranges. EXERCISE 9.10d: Parameter correlation coefficients.

Composite Scaled Sens. Figure 9.18 of Hill and Tiedeman Final Composite Scaled Sensitivities (page 251) Figure 9.11 of Hill and Tiedeman Initial Composite Scaled Sensitivities (page 243)

Confidence Intervals Figure 9.19 of Hill and Tiedeman: Confidence Intervals for Transient Regression (page 252) Figure 7.7 of Hill and Tiedeman: Confidence Intervals for Steady State Regression (page 153)

Final Parameter Correlation Coefficients Q_1&2SS_1HK_1K_RBVK_CBSS_2HK_2RCH_1RCH_2 Q_1& SS_ HK_ K_RB VK_CB SS_2 symmetric HK_ RCH_ RCH_ Table 9.7 of Hill and Tiedeman (page 253)

Model Linearity EXERCISE 9.11: Test for linearity. See Figure 9.20, p The modified Beale’s measure is 84. The model is effectively linear if this measure is less than 0.04, and the model is nonlinear if this measure is greater than 0.44.

IX. Transient Predictions

Update: Ground-Water Management Issues Results from the recalibrated model can now be used to update the advective transport predictions. Many of landfill developer’s concerns have been addressed: Model has been calibrated with head and flow data collected under same stress conditions that will exist during operation of the landfill, and under which the advective transport will be predicted. Uncertainty of most flow model parameters has been reduced, compared to their uncertainty in steady-state model. Advective travel will be analyzed under steady-state pumping conditions, because these are the conditions under which the landfill will operate.

Predicting Advective Transport Figure 9.21 of Hill and Tiedeman (page 255) Exercise 9.12a: Plot predicted path

Predicting Advective Transport Landfill

Figure 9.22 of Hill and Tiedeman (page 256) Parameters Important to Advective Paths EXERCISE 9.12b: Evaluate the model’s ability to simulate predictions using composite and prediction scaled sensitivities, and parameter correlation coefficients.

Q_1&2SS_1HK_1K_RBVK_CBSS_2HK_2RCH_1RCH_2 Q_1& SS_ HK_ K_RB VK_CB SS_2 symmetric HK_ RCH_ RCH_ Q_1&2SS_1HK_1K_RBVK_CBSS_2HK_2RCH_1RCH_2 Q_1& SS_ HK_ K_RB VK_CB SS_2 symmetric HK_ RCH_ RCH_ Table 9.7 of Hill and Tiedeman: without predictions Table 9.8 of Hill and Tiedeman: with predictions

Prediction Uncertainty: Linear Simultaneous Confidence Intervals Fig 8.15b, p. 210 From calibration with transient data From calibration with steady-state data Fig 9.23a, p. 258 EXERCISE 9.12c: Evaluate prediction uncertainty using inferential statistics.

Fig 8.15d, p. 210Fig 9.23d, p. 258 From calibration with transient data From calibration with steady-state data Prediction Uncertainty: Nonlinear Simultaneous Confidence Intervals

Finally: Should the landfill be approved? Why or why not?