Copyright © 2015 Actelion Pharmaceuticals Ltd CONFIDENCE INTERVALS FOR FUNCTIONS Anne Kümmel, Actelion Pharmaceuticals Ltd. 2015 – 07 – 16 BaselR.

Slides:



Advertisements
Similar presentations
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.
Advertisements

Design of Experiments Lecture I
1 Regression as Moment Structure. 2 Regression Equation Y =  X + v Observable Variables Y z = X Moment matrix  YY  YX  =  YX  XX Moment structure.
Objectives 10.1 Simple linear regression
Uncertainty and confidence intervals Statistical estimation methods, Finse Friday , 12.45–14.05 Andreas Lindén.
458 Quantifying Uncertainty using Classical Methods (Likelihood Profile, Bootstrapping) Fish 458, Lecture 12.
1 Regression Models & Loss Reserve Variability Prakash Narayan Ph.D., ACAS 2001 Casualty Loss Reserve Seminar.
Hypothesis Testing I 2/8/12 More on bootstrapping Random chance
Small Area Prediction under Alternative Model Specifications By Wayne A. Fuller and Andreea L. Erciulescu Department of Statistics, Iowa State University.
Correlation Mechanics. Covariance The variance shared by two variables When X and Y move in the same direction (i.e. their deviations from the mean are.
Nonlinear Feedback Loops Adding uncertainty to a dynamic model of oil prices William Strauss FutureMetrics, LLC Presented at the Palisade.
What role should probabilistic sensitivity analysis play in SMC decision making? Andrew Briggs, DPhil University of Oxford.
Uncertainty Representation. Gaussian Distribution variance Standard deviation.
GRA 6020 Multivariate Statistics The regression model OLS Regression Ulf H. Olsson Professor of Statistics.
Confidence Interval Estimation in System Dynamics Models
Lecture 8 The Principle of Maximum Likelihood. Syllabus Lecture 01Describing Inverse Problems Lecture 02Probability and Measurement Error, Part 1 Lecture.
Statistical inference form observational data Parameter estimation: Method of moments Use the data you have to calculate first and second moment To fit.
GRA 6020 Multivariate Statistics The regression model OLS Regression Ulf H. Olsson Professor of Statistics.
Lecture 10 Comparison and Evaluation of Alternative System Designs.
Bootstrap Estimation of the Predictive Distributions of Reserves Using Paid and Incurred Claims Huijuan Liu Cass Business School Lloyd’s of London 11/07/2007.
Mapping Chemical Contaminants in Oceanic Sediments Around Point Loma’s Treated Wastewater Outfall Kerry Ritter Ken Schiff N. Scott Urquhart Dawn Olson.
Stat 301 – Day 37 Bootstrapping, cont (5.5). Last Time - Bootstrapping A simulation tool for exploring the sampling distribution of a statistic, using.
Copyright (c) 2004 Brooks/Cole, a division of Thomson Learning, Inc. Chapter 7 Statistical Intervals Based on a Single Sample.
Quiz 6 Confidence intervals z Distribution t Distribution.
Standard error of estimate & Confidence interval.
Difference Two Groups 1. Content Experimental Research Methods: Prospective Randomization, Manipulation Control Research designs Validity Construct Internal.
1 MULTI VARIATE VARIABLE n-th OBJECT m-th VARIABLE.
The Triangle of Statistical Inference: Likelihoood
SIMULATION USING CRYSTAL BALL. WHAT CRYSTAL BALL DOES? Crystal ball extends the forecasting capabilities of spreadsheet model and provide the information.
Correlation and Prediction Error The amount of prediction error is associated with the strength of the correlation between X and Y.
2014. Engineers often: Regress data  Analysis  Fit to theory  Data reduction Use the regression of others  Antoine Equation  DIPPR We need to be.
Inference for Regression Simple Linear Regression IPS Chapter 10.1 © 2009 W.H. Freeman and Company.
Why Model? Make predictions or forecasts where we don’t have data.
Regression Analysis Part C Confidence Intervals and Hypothesis Testing
Limits to Statistical Theory Bootstrap analysis ESM April 2006.
Introduction to Statistical Inference Jianan Hui 10/22/2014.
Engineers often: Regress data to a model  Used for assessing theory  Used for predicting  Empirical or theoretical model Use the regression of others.
Using Microsoft Excel to Conduct Regression Analysis.
1 Probability and Statistics Confidence Intervals.
Sampling variability & the effect of spread of population.
Matrix Models for Population Management & Conservation March 2014 Lecture 10 Uncertainty, Process Variance, and Retrospective Perturbation Analysis.
Module 25: Confidence Intervals and Hypothesis Tests for Variances for One Sample This module discusses confidence intervals and hypothesis tests.
Anders Nielsen Technical University of Denmark, DTU-Aqua Mark Maunder Inter-American Tropical Tuna Commission An Introduction.
Quantifying Uncertainty
Chapter 3: Uncertainty "variation arises in data generated by a model" "how to transform knowledge of this variation into statements about the uncertainty.
Combining Deterministic and Stochastic Population Projections Salvatore BERTINO University “La Sapienza” of Rome Eugenio SONNINO University “La Sapienza”
An Introduction to AD Model Builder PFRP
Simulation setup Model parameters for simulations were tuned using repeated measurement data from multiple in-house completed studies and baseline data.
CEE 6410 Water Resources Systems Analysis
Decomposition of Sum of Squares
Probability Theory and Parameter Estimation I
Statistical Quality Control, 7th Edition by Douglas C. Montgomery.
Facilitating pharmacometric workflow with the metrumrg package for R
Orlistat for Fat Absorption
Linear models OUTLINE: Predictions
Statistics in Applied Science and Technology
Monte Carlo Simulation Managing uncertainty in complex environments.
Nonlinear Regression KNNL – Chapter 13.
CI for μ When σ is Unknown
Quantifying uncertainty using the bootstrap
HS 167 Test Prep Wednesday 5/23/07; 9:45 – 12:00
Modelling data and curve fitting
Volume 66, Issue 4, Pages (May 2010)
Analytics – Statistical Approaches
Chengyuan Yin School of mathematics
GENERALIZATION OF RESULTS OF A SAMPLE OVER POPULATION
Volume 66, Issue 4, Pages (May 2010)
Decomposition of Sum of Squares
Introductory Statistics
Probabilistic Surrogate Models
Presentation transcript:

Copyright © 2015 Actelion Pharmaceuticals Ltd CONFIDENCE INTERVALS FOR FUNCTIONS Anne Kümmel, Actelion Pharmaceuticals Ltd – 07 – 16 BaselR

© 2015 Actelion Pharmaceuticals Ltd  Scenario: –Dose response is sigmoidal –All parameters have a variability of 0.2 s.d. (log-normal)  At which dose range will I observe a response of 50% of the maximal? OFTEN, THE RESULT OF A SIMULATION, NOT THE PARAMETER VALUE ITSELF IS THE KEY INTEREST MOTIVATION 14 Jul

© 2015 Actelion Pharmaceuticals Ltd  Parameter estimation is a key tool for pharmacometric analysis of clinical data  Estimation software usually provides estimation error for the parameter estimates  How uncertain are the model predictions?  Implementation of a R framework to calculate confidence intervals for model functions –Different CI calculation methods –Single function for parameter estimation and confidence interval calculation –Any user-specified, closed-form models SUMMARY 3 14 Jul 2015

© 2015 Actelion Pharmaceuticals Ltd …. IS ONLY THE FIRST STEP UNCERTAINTY OF PARAMETER ESTIMATES 4 14 Jul 2015 ParameterEstimateRSE (%) E Emax EC  (hill coeffcient) Estimation error usually provided by estimation software Often based on the Fisher Information matrix to calculate the parameter variance- covariance matrix 

© 2015 Actelion Pharmaceuticals Ltd DIFFERENT METHODS FOR CI CALCULATION 5 14 Jul 2015

© 2015 Actelion Pharmaceuticals Ltd IMPLEMENTATION IN A SINGLE R FRAMEWORK 6 14 Jul 2015 Input parameters for parameter estimation and confidence calculation:  Error model ( errmod )  Estimation method ( estmethod )  Initial parameter estimates ( init )  Confidence interval (CI) calculation method ( CImethod )  Vector of independent variable values for which to calculate CI ( xsupport )

© 2015 Actelion Pharmaceuticals Ltd parameter.estimation() 1) Parameter estimation2) Confidence calculation3) Visualization R FRAMEWORK SETUP 7 14 Jul 2015 parameter.estimation.nls() parameter.estimation.nlm() CI.boot() CI.delta() CI.sim() CI.MC() plot.data.CI() calc.jacobian() calc.VarFun() est.nls() est.nlm()

© 2015 Actelion Pharmaceuticals Ltd EXAMPLE: ESTIMATION OF CI FOR DOSE RESPONSE CURVE 8 14 Jul 2015

© 2015 Actelion Pharmaceuticals Ltd COMPARISON OF DIFFERENT METHODS 9 14 Jul 2015 MethodCalculation time (s) Delta method0.32 Simulation (k=1000)0.92 MC simulation-estimation (k=1000) Bootstrap (k=1000)224.53

© 2015 Actelion Pharmaceuticals Ltd SINGLE-INTERFACE FRAMEWORK FOR DATA FITTING, CONFIDENCE AND PREDICTION INTERVAL CALCULATION SUMMARY Jul 2015 Possible extensions: ODE model, function for prospective analysis, link to PFIM model library Alternatives which you are using?

© 2015 Actelion Pharmaceuticals Ltd THANK YOU Jul 2015