Components of Time Series Su, Chapter 2, section II.

Slides:



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

Cointegration and Error Correction Models
Autocorrelation Functions and ARIMA Modelling
Time Series Analysis -- An Introduction -- AMS 586 Week 2: 2/4,6/2014.
Filtering the data. Detrending Economic time series are a superposition of various phenomena If there exists a « business cycle », we need to insulate.
Nonstationary Time Series Data and Cointegration
DSCI 5340: Predictive Modeling and Business Forecasting Spring 2013 – Dr. Nick Evangelopoulos Exam 1 review: Quizzes 1-6.
Long run models in economics Professor Bill Mitchell Director, Centre of Full Employment and Equity School of Economics University of Newcastle Australia.
Economics 20 - Prof. Anderson1 Stationary Stochastic Process A stochastic process is stationary if for every collection of time indices 1 ≤ t 1 < …< t.
Unit Roots & Forecasting
STAT 497 APPLIED TIME SERIES ANALYSIS
Stationary Stochastic Process
Stationary process NONSTATIONARY PROCESSES 1 In the last sequence, the process shown at the top was shown to be stationary. The expected value and variance.
Ka-fu Wong © 2003 Chap Dr. Ka-fu Wong ECON1003 Analysis of Economic Data.
Forecasting Purpose is to forecast, not to explain the historical pattern Models for forecasting may not make sense as a description for ”physical” behaviour.
Moving Averages Ft(1) is average of last m observations
Economics Prof. Buckles1 Time Series Data y t =  0 +  1 x t  k x tk + u t 1. Basic Analysis.
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.
1 Econ 240 C Lecture White noise inputoutput 1/(1 – z) White noise input output Random walkSynthesis 1/(1 – bz) White noise input output.
Analyzing and Forecasting Time Series Data
1 Power 2 Econ 240C. 2 Lab 1 Retrospective Exercise: –GDP_CAN = a +b*GDP_CAN(-1) + e –GDP_FRA = a +b*GDP_FRA(-1) + e.
Data Sources The most sophisticated forecasting model will fail if it is applied to unreliable data Data should be reliable and accurate Data should be.
ARIMA Forecasting Lecture 7 and 8 - March 14-16, 2011
Macroeconomic Facts Chapter 3. 2 Introduction Two kinds of regularities in economic data: -Relationships between the growth components in different variables.
Economics 20 - Prof. Anderson
Modern methods The classical approach: MethodProsCons Time series regression Easy to implement Fairly easy to interpret Covariates may be added (normalization)
Part II – TIME SERIES ANALYSIS C2 Simple Time Series Methods & Moving Averages © Angel A. Juan & Carles Serrat - UPC 2007/2008.
K. Ensor, STAT Spring 2005 The Basics: Outline What is a time series? What is a financial time series? What is the purpose of our analysis? Classification.
Copyright 2013 John Wiley & Sons, Inc. Chapter 8 Supplement Forecasting.
Modern methods The classical approach: MethodProsCons Time series regression Easy to implement Fairly easy to interpret Covariates may be added (normalization)
1Prof. Dr. Rainer Stachuletz Time Series Data y t =  0 +  1 x t  k x tk + u t 1. Basic Analysis.
R. Werner Solar Terrestrial Influences Institute - BAS Time Series Analysis by descriptive statistic.
© 2003 Prentice-Hall, Inc.Chap 12-1 Business Statistics: A First Course (3 rd Edition) Chapter 12 Time-Series Forecasting.
BOX JENKINS METHODOLOGY
© 2002 Prentice-Hall, Inc.Chap 13-1 Statistics for Managers using Microsoft Excel 3 rd Edition Chapter 13 Time Series Analysis.
Time Series “The Art of Forecasting”. What Is Forecasting? Process of predicting a future event Underlying basis of all business decisions –Production.
TIME SERIES by H.V.S. DE SILVA DEPARTMENT OF MATHEMATICS
STAT 497 LECTURE NOTES 2.
Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin Time Series Forecasting Chapter 16.
McGraw-Hill/Irwin Copyright © 2007 by The McGraw-Hill Companies, Inc. All rights reserved. Time Series Forecasting Chapter 13.
Time series Decomposition Farideh Dehkordi-Vakil.
Statistics and Modelling 3.1 Credits: 3 Internally Assessed.
Time-Series Forecasting Overview Moving Averages Exponential Smoothing Seasonality.
FAME Time Series Econometrics Daniel V. Gordon Department of Economics University of Calgary.
Lecture 6: Topic #1 Forecasting trend and seasonality.
Introduction to Time Series Analysis
Big Data at Home Depot KSU – Big Data Survey Course Steve Einbender Advanced Analytics Architect.
Week 21 Stochastic Process - Introduction Stochastic processes are processes that proceed randomly in time. Rather than consider fixed random variables.
© 1999 Prentice-Hall, Inc. Chap Chapter Topics Component Factors of the Time-Series Model Smoothing of Data Series  Moving Averages  Exponential.
Economics 173 Business Statistics Lecture 23 © Fall 2001, Professor J. Petry
COMPLETE BUSINESS STATISTICS
Previously Definition of a stationary process (A) Constant mean (B) Constant variance (C) Constant covariance White Noise Process:Example of Stationary.
The Box-Jenkins (ARIMA) Methodology
NONSTATIONARY PROCESSES 1 In the last sequence, the process shown at the top was shown to be stationary. The expected value and variance of X t were shown.
Time Series and Forecasting
Computational Finance II: Time Series K.Ensor. What is a time series? Anything observed sequentially (by time?) Returns, volatility, interest rates, exchange.
Irwin/McGraw-Hill © Andrew F. Siegel, 1997 and l Chapter 14 l Time Series: Understanding Changes over Time.
Economics 20 - Prof. Anderson1 Time Series Data y t =  0 +  1 x t  k x tk + u t 1. Basic Analysis.
Forecasting is the art and science of predicting future events.
Ch16: Time Series 24 Nov 2011 BUSI275 Dr. Sean Ho HW8 due tonight Please download: 22-TheFed.xls 22-TheFed.xls.
Times Series Forecasting and Index Numbers Chapter 16 Copyright © 2014 by The McGraw-Hill Companies, Inc. All rights reserved.McGraw-Hill/Irwin.
Chapter 20 Time Series Analysis and Forecasting. Introduction Any variable that is measured over time in sequential order is called a time series. We.
Chapter 20 Time Series Analysis and Forecasting. Introduction Any variable that is measured over time in sequential order is called a time series. We.
EC 827 Module 2 Forecasting a Single Variable from its own History.
Yandell – Econ 216 Chap 16-1 Chapter 16 Time-Series Forecasting.
Analysis of Financial Data Spring 2012 Lecture 4: Time Series Models - 1 Priyantha Wijayatunga Department of Statistics, Umeå University
Chapter 6: Autoregressive Integrated Moving Average (ARIMA) Models
Statistics for Managers using Microsoft Excel 3rd Edition
Chapter 8 Supplement Forecasting.
Presentation transcript:

Components of Time Series Su, Chapter 2, section II.

Four Primary Components of a Time Series: Secular Trend Seasonal Trend Cyclical Movements Irregular Components

Example: Secular Trend

Example: Seasonal Component

Example: Cyclical Component

Example: Random/Irregular

Mathematical Representations Additive: Y = T + S + C + I Multiplicative: Y = T x S x C x I

Observations: Traditional time series analysis is “atheoretic”. No economic theory guides us in writing down this decomposition. Typically, one of these components will dominate and this will affect the behavior of the series.

Secular Trends Often called “Time Trends” Visual representation is called “Time Path” or “Time Shape” A continuous set of integers is used to represent time in these models. Linear Time Trend Model Y t =  0 +  1 T t

How predictable is the secular trend in a series?

Removing Time Trends: Detrending Often, the trend component of a time series dominates, but the interesting part of the series is another component. Example: Civilian Employment

Civilian Employment Time Series The trend component is dominant. Why? Suppose we’re interested in changes in employment over the business cycle. –Problem: The cyclical component subordinate to the secular trend. –Solution: Detrend

Detrending Step 1: Estimate the Secular Trend using regression model Step 2: Subtract the estimated secular trend from the original series. Note: This is also the “Residual Approach” to analyzing cyclical data in C1.

Regression Output

Plot of Detrended Data

Seasonal Component Found in High Frequency data (Quarterly, monthly) Caused by natural or budget calendars –Retail Sales higher during holidays –Travel more frequent in summer –Weather Want to quantify or remove in forecasting How predictable is this component?

“Unadjusted Data”

Same series after adjustment

Seasonal Adjustment Methods Dummy Variables Ratio-to-moving-average X-11 - Asymmetric Moving Averages

A (Partial) Example: Income Tax Revenues

Moving Average Computation

Regression Method

Irregular or Random Components Special events that pull macro variables off their usual paths. Can be expected or unexpected.

Modeling Random / Irregular Components May stem from randomness or irregularities in human behavior. Keynes: “Animal Spirits” Some irregular events are not random, they are caused by specific factors If they are truly random, there is no way to predict them. If they are just irregular, they can be handled by dummy variables.

Modeling Time Series Goal is to distinguish between the deterministic (or predictable) and stochastic (or random) parts Y t =  t + u t  t is the deterministic component – secular trend, seasonal and cyclical movements u t is the stochastic component Y t = T t + C t + S t + u t

Assumptions: Random Component Typically make three assumptions about u t Mean zero: E(u t ) = 0 Constant variance/no covariance E(u t u t+i ) =  2 u if i=0 (Constant variance) E(u t u t+i ) = 0 if i  0 (Zero covariance) Normally distributed u t ~ N(0,  2 u )

Stationarity Refers to the idea that a time series should be stable over time – returns to an equilibrium level Stationarity is an important concept for forecasting because only stationary time series are predictable A stationary time series has a mean, variance and autocovariances that do not change over time Many economic time series are not stationary – alternative is “random walk” Tests for stationarity exist – later in semester

Transformations A nonstationary series must be transformed to make it stationary (First) Differencing:  Y t = Y t – Y t-1 Detrending

Stationary Time Path t YtYt Equilibrium Time Path Shock

Autocovariances Covariance between two observations Example: kth-order autocovariance is the covariance between observations of a time series k periods apart (or lagged k periods) Cov(Y t Y t-k ) If the autocovariances of a time series are stationary (do not change over time) then they can be used to forecast a series Autocovariances are a measure of predictability

Autocorrelations Closely related to autocovariances Just the correlation between any two observations of a time series If Cov(Y t Y t-k ) is the autocovariance, then cor(Y t Y t-k ) = Cov(Y t Y t-k )/var(Y t )