Psychometrics, Dynamics, and Functional Data Analysis Jim Ramsay Jim Ramsay McGill University “The views “The views.

Slides:



Advertisements
Similar presentations
First Order Systems: Dynamic Systems ISAT 300 Spring 1999.
Advertisements

1 Machine Learning: Lecture 1 Overview of Machine Learning (Based on Chapter 1 of Mitchell T.., Machine Learning, 1997)
Technique of nondimensionalization Aim: –To remove physical dimensions –To reduce the number of parameters –To balance or distinguish different terms in.
Polynomial Curve Fitting BITS C464/BITS F464 Navneet Goyal Department of Computer Science, BITS-Pilani, Pilani Campus, India.
Experiments and Variables
Experimental Design, Response Surface Analysis, and Optimization
Process Design (Specification)
Princeton University Using the computer to select the right variables Rationale: Lake Carnegie, Princeton, NJ Straight Line Distance Actual transition.
Cost Accounting Dr. Baldwin University of Arkansas – Fort Smith Fall 2010.
6 - 1 © 2012 Person Education, Inc.. All rights reserved. Chapter 6 Applications of the Derivative.
Selected from presentations by Jim Ramsay, McGill University, Hongliang Fei, and Brian Quanz Basis Basics.
An Introduction to Functional Data Analysis Jim Ramsay McGill University.
Jim Ramsay McGill University Basis Basics. Overview  What are basis functions?  What properties should they have?  How are they usually constructed?
Ekstrom Math 115b Mathematics for Business Decisions, part II Project 1: Marketing Computer Drives Math 115b.
x – independent variable (input)
1 Fifth Lecture Dynamic Characteristics of Measurement System (Reference: Chapter 5, Mechanical Measurements, 5th Edition, Bechwith, Marangoni, and Lienhard,
From Data to Differential Equations Jim Ramsay McGill University With inspirations from Paul Speckman and Chong Gu.
Development of Empirical Models From Process Data
Human Growth: From data to functions. Challenges to measuring growth We need repeated and regular access to subjects for up to 20 years. We need repeated.
Project 1: Marketing An Introduction. Goal Your goal is three-fold: To find the price of a particular product that will maximize profits. Determine the.
The Game of Algebra or The Other Side of Arithmetic The Game of Algebra or The Other Side of Arithmetic © 2007 Herbert I. Gross by Herbert I. Gross & Richard.
Breakeven Analysis for Profit Planning
Classification and Prediction: Regression Analysis
From Data to Differential Equations Jim Ramsay McGill University.
Human Growth: From data to functions. Challenges to measuring growth We need repeated and regular access to subjects for up to 20 years. We need repeated.
Managerial Economics Managerial Economics = economic theory + mathematical eco + statistical analysis.
Chapter 3 Introduction to the Derivative Sections 3. 5, 3. 6, 4
LSS Black Belt Training Forecasting. Forecasting Models Forecasting Techniques Qualitative Models Delphi Method Jury of Executive Opinion Sales Force.
Asymptotic Techniques
Gaussian process modelling
Ch. 6 Single Variable Control
Copyright © 2008 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin 12 Financial and Cost- Volume-Profit Models.
© 2008 Pearson Addison-Wesley. All rights reserved Chapter 1 Section 13-6 Regression and Correlation.
Dynamic Presentation of Key Concepts Module 5 – Part 1 Fundamentals of Operational Amplifiers Filename: DPKC_Mod05_Part01.ppt.
Chapter © 2015 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole or.
Slide 1 Lesson 76 Graphing with Rates Chapter 14 Lesson 76 RR.7Understand that multiplication by rates and ratios can be used to transform an input into.
Chapter 3 mathematical Modeling of Dynamic Systems
Various topics Petter Mostad Overview Epidemiology Study types / data types Econometrics Time series data More about sampling –Estimation.
2.5 Implicit Differentiation
Chapter 2 Describing Motion: Kinematics in One Dimension.
Major objective of this course is: Design and analysis of modern algorithms Different variants Accuracy Efficiency Comparing efficiencies Motivation thinking.
Progress in identification of damping: Energy-based method with incomplete and noisy data Marco Prandina University of Liverpool.
Signals and Systems Dr. Mohamed Bingabr University of Central Oklahoma
Chaos in a Pendulum Section 4.6 To introduce chaos concepts, use the damped, driven pendulum. This is a prototype of a nonlinear oscillator which can.
Fundamentals of Electric Circuits Chapter 16 Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
1 CMSC 671 Fall 2001 Class #25-26 – Tuesday, November 27 / Thursday, November 29.
SYSTEMS Identification Ali Karimpour Assistant Professor Ferdowsi University of Mashhad Reference: “System Identification Theory For The User” Lennart.
 For years, corporations have used computer-based simulations with employee-training programs, augmenting traditional on-the-job training with virtual.
Feedback Control Systems (FCS) Dr. Imtiaz Hussain URL :
Chapter 2 Describing Motion: Kinematics in One Dimension © 2014 Pearson Education, Inc.
ME 431 System Dynamics Dept of Mechanical Engineering.
IST_Seminar II CHAPTER 12 Instructional Methods. Objectives: Students will: Explain the role of all teachers in the development of critical thinking skills.
Options and generalisations. Outline Dimensionality Many inputs and/or many outputs GP structure Mean and variance functions Prior information Multi-output,
Understanding the difference between an engineer and a scientist There are many similarities and differences.
Introduction: Thinking Like an Economist 1 CHAPTER 2 Production and Cost Analysis II Economic efficiency consists of making things that are worth more.
1 Development of Empirical Models From Process Data In some situations it is not feasible to develop a theoretical (physically-based model) due to: 1.
Science and Engineering Practices K–2 Condensed Practices3–5 Condensed Practices6–8 Condensed Practices9–12 Condensed Practices Developing and Using Models.
Chapter 3 Describing Motion: Kinematics in One Dimension.
Modelling & Simulation of Semiconductor Devices Lecture 1 & 2 Introduction to Modelling & Simulation.
Skill acquisition
A few illustrations on the Basic Concepts of Nonlinear Control
MESB374 System Modeling and Analysis
Applications of the Derivative
Applications of the Derivative
Digital Control Systems (DCS)
Objective of This Course
Digital Control Systems (DCS)
Analytics for the IoT: A Deep Dive into Algorithms
Human Growth: From data to functions
Coordinate Transformation in 3D Final Project Presentation
Presentation transcript:

Psychometrics, Dynamics, and Functional Data Analysis Jim Ramsay Jim Ramsay McGill University “The views “The views

Testing as Input/Output Analysis A test score is actually a derivative with respect to time. A test score is actually a derivative with respect to time. Consequently a differential equation model for testing data seems natural. Consequently a differential equation model for testing data seems natural. Dynamic testing data will be more and more important. Dynamic testing data will be more and more important. We have some new tools for working with dynamic data. We have some new tools for working with dynamic data. So let’s consider how to use time as a covariate. So let’s consider how to use time as a covariate.

Learning to Play Golf We buy some clubs. We play a few games, each being an 18 item test. It’s harder than it looks. We buy some clubs. We play a few games, each being an 18 item test. It’s harder than it looks. We take a lesson. We play a few games. Our score improves to a better level. We take a lesson. We play a few games. Our score improves to a better level. We take another lesson, play some games, and things improve again. We take another lesson, play some games, and things improve again. Key question: How quickly is a lesson reflected in an improvement in score? Key question: How quickly is a lesson reflected in an improvement in score?

Brainergy Energy is defined as “the capacity to do work.” Energy is defined as “the capacity to do work.” Kinetic energy E = Mv 2 /2, and involves mass, distance, and time. Kinetic energy E = Mv 2 /2, and involves mass, distance, and time. We are interested in “the capacity to solve problems.” We are interested in “the capacity to solve problems.” Problems involve difficulties (=mass), number of problems (=distance) and time. Problems involve difficulties (=mass), number of problems (=distance) and time. Let’s call mental energy brainergy. Let’s call mental energy brainergy.

Brain Power What counts is problem solving per unit time. What counts is problem solving per unit time. Power = energy expended per unit time. Power = energy expended per unit time. Brain power = maximum difficulty of problem solvable per unit time, or Brain power = maximum difficulty of problem solvable per unit time, or number of lighter problems solved per unit time. number of lighter problems solved per unit time. That is, brain power = d brainergy/dt. That is, brain power = d brainergy/dt.

Brain Power and Time Scales We need the concept of brain power when we consider intelligence on two time scales: 1. Long term: How much knowledge is available over large time intervals, like a school year 2. Short term: How much new knowledge can be acquired over a short time interval, like a single class.

Tests Measure Brain Power Mental tests and psychological scales are one of the greatest technological achievements of the 20 th century. Mental tests and psychological scales are one of the greatest technological achievements of the 20 th century. Tests work so well because they are time- limited. Tests work so well because they are time- limited. Test scores reflect brain power rather than brainergy. Test scores reflect brain power rather than brainergy.

Inputs to Brain Power Information about the structure of the problems. Information about the structure of the problems. A set of tools to solve them. A set of tools to solve them. Training in the use of these tools. Training in the use of these tools. All these require time. All these require time. Inputs to acquisition of brain power are functions of time. Inputs to acquisition of brain power are functions of time.

A Differential Equation in Time Links one or more time-derivatives, dx/dt, d 2 x/dt 2,…, to the function x(t) itself. Links one or more time-derivatives, dx/dt, d 2 x/dt 2,…, to the function x(t) itself. Is a model for system dynamics: change over time. Is a model for system dynamics: change over time. Can also include one or more input or covariate functions. Can also include one or more input or covariate functions. x(t) is a long-term description. x(t) is a long-term description. dx/dt is a short-term description. dx/dt is a short-term description.

A Simple Example E(t) is brainergy, dE/dt is brain power. E(t) is brainergy, dE/dt is brain power. f(t) is an input function of time, such as education. f(t) is an input function of time, such as education. α and β are constants, β > 0. α and β are constants, β > 0.

Most differential equations don’t have explicit solutions, but this one does. Most differential equations don’t have explicit solutions, but this one does. Let E 0 be brainergy at time t = 0, and which will often be 0. Let E 0 be brainergy at time t = 0, and which will often be 0. Let’s see what happens when α=1, β varies, and f(t) is a step function. Let’s see what happens when α=1, β varies, and f(t) is a step function.

The slope of E(t) when f(t) goes positive is β. The slope of E(t) when f(t) goes positive is β. β controls how fast the system responds to the input f(t). β controls how fast the system responds to the input f(t). If the system is a problem solver, then β indicates how quickly the person learns to solve a problem. If the system is a problem solver, then β indicates how quickly the person learns to solve a problem. After about 4/β time units, full capacity is reached, and the system is ready for more input. After about 4/β time units, full capacity is reached, and the system is ready for more input.

Fitting Differential Equations We have noisy discrete-time data, and want to use them to estimate a differential equation. We have noisy discrete-time data, and want to use them to estimate a differential equation. We want a solution E(t) to the equation to fit the data as well as possible. We want a solution E(t) to the equation to fit the data as well as possible. We need lots of flexibility in choosing a differential equation, and we can’t assume that there is an explicit solution to the equation. We need lots of flexibility in choosing a differential equation, and we can’t assume that there is an explicit solution to the equation.

Functional Data Analysis A collection of methods for analyzing curves or functions as data A collection of methods for analyzing curves or functions as data A common theme is using derivatives in various ways A common theme is using derivatives in various ways See Ramsay and Silverman (1997) Functional Data Analysis. Springer. See Ramsay and Silverman (1997) Functional Data Analysis. Springer. And Ramsay and Silverman (2002) Applied Functional Data Analysis. Springer. And Ramsay and Silverman (2002) Applied Functional Data Analysis. Springer.

Two Functional Data Analysis Techniques L-spline Smoothing: given noisy data and a differential equation, find a function E(t) that will smooth the data and at the same time be nearly a solution to the differential equation. L-spline Smoothing: given noisy data and a differential equation, find a function E(t) that will smooth the data and at the same time be nearly a solution to the differential equation. Principal Differential Analysis: given a function E(t), estimate a linear differential equation for which E(t) is a solution. Principal Differential Analysis: given a function E(t), estimate a linear differential equation for which E(t) is a solution.

Estimating a DIFE from noisy data We’ve recently combined these two methods into a technique for estimating a differential equation from noisy data. We’ve recently combined these two methods into a technique for estimating a differential equation from noisy data. In our simple example, this amounts to estimating parameters α and β. In our simple example, this amounts to estimating parameters α and β. But much more complex DIFE’s can be estimated as well, including linear or nonlinear, and single or multiple variable systems. But much more complex DIFE’s can be estimated as well, including linear or nonlinear, and single or multiple variable systems.

An Oil Refinery Here are some data from an oil refinery in Corpus Christi. Here are some data from an oil refinery in Corpus Christi. The input f(t) (reflux flow) is negatively coupled to the output E(t) (tray 47 level). The input f(t) (reflux flow) is negatively coupled to the output E(t) (tray 47 level). The smooth curve is a solution to the differential equation that best represents this relationship. The smooth curve is a solution to the differential equation that best represents this relationship.

Perhaps this oil refinery is not too smart!

Many situations will call for multiple outputs: Performance with a putter, a driver, and an iron, for example. Or in algebra and geometry. Many situations will call for multiple outputs: Performance with a putter, a driver, and an iron, for example. Or in algebra and geometry. And many situations will involve multiple inputs: Regular classes, tutoring sessions, labs and etc. And many situations will involve multiple inputs: Regular classes, tutoring sessions, labs and etc. The technology used in these illustrations can handle these situations, at least for linear differential equations. Nonlinear equations don’t pose any problem in principle. The technology used in these illustrations can handle these situations, at least for linear differential equations. Nonlinear equations don’t pose any problem in principle.

Some Simulated Data Imagine that the data are golf scores over successive games, and that the input is a set of three equally-spaced lessons from a golf pro. Imagine that the data are golf scores over successive games, and that the input is a set of three equally-spaced lessons from a golf pro. The following slides show three golfers. Which is a future Tiger Woods? The following slides show three golfers. Which is a future Tiger Woods?

These lessons are nicely timed.

This person needs to find another sport!

This person should get lessons more often!

Is this Model Good Enough? Specifying β to be constant is too simple. Allowing for fatigue, boredom, and other things requires a function β(t). Specifying β to be constant is too simple. Allowing for fatigue, boredom, and other things requires a function β(t). A first order equation can’t allow for sudden transient effects like insight. We may need a differential equation involving higher derivatives. A first order equation can’t allow for sudden transient effects like insight. We may need a differential equation involving higher derivatives. We may need nonlinear equations as well. We may need nonlinear equations as well.

A Nonlinear Differential Equation The summed output from these two equations will exhibit both the rapid learning and long-term retention required of human learners. See H. R. Wilson (1999) Spikes, Decisions and Actions, Oxford, for many more examples of differential equation models in neuroscience.

Control Theory Engineers who work with input/output systems have developed ways of designing feedback loops to optimize outputs. Engineers who work with input/output systems have developed ways of designing feedback loops to optimize outputs. We’re working with a team of chemical engineers at Queen’s University. We’re working with a team of chemical engineers at Queen’s University.

Where Would the Data Come From? Can we design customized learning situations, like golf, and track how a learner makes progress as a function of time and inputs? Can we design customized learning situations, like golf, and track how a learner makes progress as a function of time and inputs? Perhaps video and computer games are nearly what we need. Perhaps video and computer games are nearly what we need. We already know that people will pay big money to have these experiences. We already know that people will pay big money to have these experiences. Would corporations with deep pockets pay for this kind of testing? Would corporations with deep pockets pay for this kind of testing?

Conclusions Dynamic testing would generate performance data over time that depend on one or more functional covariates. Dynamic testing would generate performance data over time that depend on one or more functional covariates. New tools are available for these data that fit them with a differential equation. New tools are available for these data that fit them with a differential equation. Dynamic psychometrics looks promising! Dynamic psychometrics looks promising!