Time Series Analysis, Part I. A time series is A time series is a sequence of measurements. Usually we deal with equi-spaced measurements. What distinguishes.

Slides:



Advertisements
Similar presentations
FINANCIAL TIME-SERIES ECONOMETRICS SUN LIJIAN Feb 23,2001.
Advertisements

Time Series Analysis -- An Introduction -- AMS 586 Week 2: 2/4,6/2014.
Time Series Analysis Topics in Machine Learning Fall 2011 School of Electrical Engineering and Computer Science.
Dates for term tests Friday, February 07 Friday, March 07
Signal Processing in the Discrete Time Domain Microprocessor Applications (MEE4033) Sogang University Department of Mechanical Engineering.
Nonstationary Time Series Data and Cointegration
ELG5377 Adaptive Signal Processing
R. Werner Solar Terrestrial Influences Institute - BAS Time Series Analysis by means of inference statistical methods.
1 Alberto Montanari University of Bologna Simulation of synthetic series through stochastic processes.
EE-2027 SaS 06-07, L11 1/12 Lecture 11: Fourier Transform Properties and Examples 3. Basis functions (3 lectures): Concept of basis function. Fourier series.
Time series of the day. Stat Sept 2008 D. R. Brillinger Simple descriptive techniques Trend X t =  +  t +  t Filtering y t =  r=-q s a r.
Modeling Cycles By ARMA
3F4 Power and Energy Spectral Density Dr. I. J. Wassell.
Laplace Transforms Important analytical method for solving linear ordinary differential equations. - Application to nonlinear ODEs? Must linearize first.
3 mo treasury yield borrowing costs Dow industrials NY Times 18 Sept 2008 front page.
In the preceding chapter we used the Laplace transform to obtain transfer function models representing linear, time-invariant physical systems described.
ARIMA Using Stata. Time Series Analysis Stochastic Data Generating Process –Stable and Stationary Process Autoregressive Process: AR(p) Moving Average.
Laplace Transforms 1. Standard notation in dynamics and control (shorthand notation) 2. Converts mathematics to algebraic operations 3. Advantageous for.
Lecture 8 Topics Fourier Transforms –As the limit of Fourier Series –Spectra –Convergence of Fourier Transforms –Fourier Transform: Synthesis equation.
ARIMA Forecasting Lecture 7 and 8 - March 14-16, 2011
Lecture 9: Fourier Transform Properties and Examples
Modern methods The classical approach: MethodProsCons Time series regression Easy to implement Fairly easy to interpret Covariates may be added (normalization)
Chapter 3 1 Laplace Transforms 1. Standard notation in dynamics and control (shorthand notation) 2. Converts mathematics to algebraic operations 3. Advantageous.
Computational Geophysics and Data Analysis
Random Process The concept of random variable was defined previously as mapping from the Sample Space S to the real line as shown below.
STAT 497 LECTURE NOTES 2.
1 Part 5 Response of Linear Systems 6.Linear Filtering of a Random Signals 7.Power Spectrum Analysis 8.Linear Estimation and Prediction Filters 9.Mean-Square.
Linear Stationary Processes. ARMA models. This lecture introduces the basic linear models for stationary processes. Considering only stationary processes.
Laplace Transforms 1. Standard notation in dynamics and control (shorthand notation) 2. Converts mathematics to algebraic operations 3. Advantageous for.
Laplace Transforms 1. Standard notation in dynamics and control (shorthand notation) 2. Converts mathematics to algebraic operations 3. Advantageous for.
MGS3100_01.ppt/Aug 25, 2015/Page 1 Georgia State University - Confidential MGS 3100 Business Analysis Introduction - Why Business Analysis Aug 25 and 26,
Data analyses 2008 Lecture Last Lecture Basic statistics Testing Linear regression parameters Skill.
Time Series Basics (2) Fin250f: Lecture 3.2 Fall 2005 Reading: Taylor, chapter , 3.9(skip 3.6.1)
Week 21 Stochastic Process - Introduction Stochastic processes are processes that proceed randomly in time. Rather than consider fixed random variables.
Lecturer CS&E Department,
Signals and Systems 1 Lecture 3 Dr. Ali. A. Jalali August 23, 2002.
Time Series Basics Fin250f: Lecture 8.1 Spring 2010 Reading: Brooks, chapter
ENEE631 Digital Image Processing (Spring'04) Basics on 2-D Random Signal Spring ’04 Instructor: Min Wu ECE Department, Univ. of Maryland, College Park.
Statistics 349.3(02) Analysis of Time Series. Course Information 1.Instructor: W. H. Laverty 235 McLean Hall Tel:
Stochastic models - time series. Random process. an infinite collection of consistent distributions probabilities exist Random function. a family of random.
Linear Filters. denote a bivariate time series with zero mean. Let.
Discrete-time Random Signals
Lecture 12: Parametric Signal Modeling XILIANG LUO 2014/11 1.
Review and Summary Box-Jenkins models Stationary Time series AR(p), MA(q), ARMA(p,q)
Components of Time Series Su, Chapter 2, section II.
Correlogram - ACF. Modeling for Forecast Forecast Data The Base Model Linear Trend Logistic Growth Others Models Look for a best approximation of the.
Lecture 2: The Laplace Transform Laplace transform definition Laplace transform properties Relation between time and Laplace domains Initial and Final.
Introduction to stochastic processes
Ordinary Differential Equations (ODEs). Objectives of Topic  Solve Ordinary Differential Equations (ODEs).  Appreciate the importance of numerical methods.
1 Autocorrelation in Time Series data KNN Ch. 12 (pp )
Stochastic Process - Introduction
SIGNALS PROCESSING AND ANALYSIS
Laplace and Z transforms
Everything You Ever Wanted to Know About Filters*
Statistics 153 Review - Sept 30, 2008
Random Process The concept of random variable was defined previously as mapping from the Sample Space S to the real line as shown below.
Stochastic models - time series.
Research Methods in Acoustics Lecture 9: Laplace Transform and z-Transform Jonas Braasch.
Modeling in the Time Domain
Stochastic models - time series.
Fourier Series: Examples
EE-314 Signals and Linear Systems
1 Z Transform Dr.P.Prakasam Professor/ECE 9/18/2018SS/Dr.PP/ZT.
State Space Method.
Analysis of Control Systems in State Space
The Spectral Representation of Stationary Time Series
Use power series to solve the differential equation. y ' = 7xy
Time Series introduction in R - Iñaki Puigdollers
BOX JENKINS (ARIMA) METHODOLOGY
Presentation transcript:

Time Series Analysis, Part I

A time series is A time series is a sequence of measurements. Usually we deal with equi-spaced measurements. What distinguishes a time series from a sequence of random numbers? A dependence between values at time t and time t+k. Time Series Analysis is concerned with techniques for analyzing this dependence.

Definitions Stationary Stochastic

Time Series Examples Features to note –how processed are they? –are there periodicities? –are they stationary? –how predictable are they? –what process generated it?

Time Series Image Features to note –how processed are they? –are there periodicities? –are they stationary? –how predictable are they? –what process generated it?

Time Series Image Features to note –how processed are they? –are there periodicities? –are they stationary? –how predictable are they? –what process generated it?

Time Series Image Features to note –how processed are they? –are there periodicities? –are they stationary? –how predictable are they? –what process generated it?

Definitions Stationary Stochastic Deterministic

Time Series Analysis Identification (of a model) –Diagram of black box concept –In space sciences, identification of the black box is non-trivial Prediction or Forecasting (using a model) –Less concerned with getting the right “model” –More concerned with getting the right prediction. See diagram.

Identification Example Given the model dx/dt = -x/tau + f(t), what is tau?

Identification Example Given the model dx/dt = -x/tau + f(t), what is tau?

Is there a different way?

Is there a better way? How could you determine tau from this graph? , , , , , , x f/10

An even better way Matrix method

End intuitive, begin formal Laplace Transform (L) = 1-sided Fourier Transofrm, (FT) Transfer function Ordinary differential equation (ODE) Impulse response function (IRF)

Types of filters Linear filter Autoregressive filter Moving average filter

Techniques for Dependence Analysis Autocovariance and the Autocorrelation Function (ACF)

Sketch ACF for these functions x(t) = 1 (t = 1, 2, …, 10) x(t) = t (t = 1, 2, …, 10) x(t) = sin(2  t/10) (t = 1, 2, …, 10)

Matrix forms

Estimation

What is better ACF or FT?

Relationship between ACF and FT

Linear Stationary Models