Part II – TIME SERIES ANALYSIS C5 ARIMA (Box-Jenkins) Models

Slides:



Advertisements
Similar presentations
Autocorrelation Functions and ARIMA Modelling
Advertisements

DSCI 5340: Predictive Modeling and Business Forecasting Spring 2013 – Dr. Nick Evangelopoulos Exam 1 review: Quizzes 1-6.
Part II – TIME SERIES ANALYSIS C3 Exponential Smoothing Methods © Angel A. Juan & Carles Serrat - UPC 2007/2008.
Part II – TIME SERIES ANALYSIS C4 Autocorrelation Analysis © Angel A. Juan & Carles Serrat - UPC 2007/2008.
Inference for Regression
Model Building For ARIMA time series
Unit Roots & Forecasting
Slide 1 DSCI 5340: Predictive Modeling and Business Forecasting Spring 2013 – Dr. Nick Evangelopoulos Lecture 7: Box-Jenkins Models – Part II (Ch. 9) Material.
Applied Business Forecasting and Planning
Time Series Building 1. Model Identification
R. Werner Solar Terrestrial Influences Institute - BAS Time Series Analysis by means of inference statistical methods.
Part II – TIME SERIES ANALYSIS C1 Introduction to TSA © Angel A. Juan & Carles Serrat - UPC 2007/2008 "If we could first know where we are, then whither.
How should these data be modelled?. Identification step: Look at the SAC and SPAC Looks like an AR(1)- process. (Spikes are clearly decreasing in SAC.
Business Forecasting Chapter 10 The Box–Jenkins Method of Forecasting.
Tutorial for solution of Assignment week 40 “Forecasting monthly values of Consumer Price Index Data set: Swedish Consumer Price Index” sparetime.
Non-Seasonal Box-Jenkins Models
13 Introduction toTime-Series Analysis. What is in this Chapter? This chapter discusses –the basic time-series models: autoregressive (AR) and moving.
BABS 502 Lecture 9 ARIMA Forecasting II March 23, 2009.
Time series analysis - lecture 2 A general forecasting principle Set up probability models for which we can derive analytical expressions for and estimate.
ARIMA-models for non-stationary time series
Modeling Cycles By ARMA
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.
BABS 502 Lecture 8 ARIMA Forecasting II March 16 and 21, 2011.
Prediction and model selection
ARIMA Forecasting Lecture 7 and 8 - March 14-16, 2011
Financial Econometrics
Non-Seasonal Box-Jenkins Models
Business Forecasting Chapter 5 Forecasting with Smoothing Techniques.
1 Simple Linear Regression 1. review of least squares procedure 2. inference for least squares lines.
BOX JENKINS METHODOLOGY
Box Jenkins or Arima Forecasting. H:\My Documents\classes\eco346\Lectures\chap ter 7\Autoregressive Models.docH:\My Documents\classes\eco346\Lectures\chap.
AR- MA- och ARMA-.
Time Series Forecasting (Part II)
STAT 497 LECTURE NOTES 2.
1 Least squares procedure Inference for least squares lines Simple Linear Regression.
Slide 1 DSCI 5340: Predictive Modeling and Business Forecasting Spring 2013 – Dr. Nick Evangelopoulos Lecture 8: Estimation & Diagnostic Checking in Box-Jenkins.
#1 EC 485: Time Series Analysis in a Nut Shell. #2 Data Preparation: 1)Plot data and examine for stationarity 2)Examine ACF for stationarity 3)If not.
Slide 1 DSCI 5340: Predictive Modeling and Business Forecasting Spring 2013 – Dr. Nick Evangelopoulos Exam 2 review: Quizzes 7-12* (*) Please note that.
Tutorial for solution of Assignment week 39 “A. Time series without seasonal variation Use the data in the file 'dollar.txt'. “
The Examination of Residuals. Examination of Residuals The fitting of models to data is done using an iterative approach. The first step is to fit a simple.
Lecture 7: Forecasting: Putting it ALL together. The full model The model with seasonality, quadratic trend, and ARMA components can be written: Ummmm,
Autoregressive Integrated Moving Average (ARIMA) Popularly known as the Box-Jenkins methodology.
John G. Zhang, Ph.D. Harper College
Autocorrelation, Box Jenkins or ARIMA Forecasting.
Big Data at Home Depot KSU – Big Data Survey Course Steve Einbender Advanced Analytics Architect.
STAT 497 LECTURE NOTE 9 DIAGNOSTIC CHECKS 1. After identifying and estimating a time series model, the goodness-of-fit of the model and validity of the.
Time Series Analysis and Forecasting. Introduction to Time Series Analysis A time-series is a set of observations on a quantitative variable collected.
MULTIVARIATE TIME SERIES & FORECASTING 1. 2 : autocovariance function of the individual time series.
Auto Regressive, Integrated, Moving Average Box-Jenkins models A stationary times series can be modelled on basis of the serial correlations in it. A non-stationary.
Time Series Analysis Lecture 11
Forecasting (prediction) limits Example Linear deterministic trend estimated by least-squares Note! The average of the numbers 1, 2, …, t is.
The Box-Jenkins (ARIMA) Methodology
Seasonal ARIMA FPP Chapter 8.
Ch16: Time Series 24 Nov 2011 BUSI275 Dr. Sean Ho HW8 due tonight Please download: 22-TheFed.xls 22-TheFed.xls.
Introduction to stochastic processes
Time Series Analysis PART II. Econometric Forecasting Forecasting is an important part of econometric analysis, for some people probably the most important.
Subodh Kant. Auto-Regressive Integrated Moving Average Also known as Box-Jenkins methodology A type of linear model Capable of representing stationary.
MODEL DIAGNOSTICS By Eni Sumarminingsih, Ssi, MM.
Forecasting. Model with indicator variables The choice of a forecasting technique depends on the components identified in the time series. The techniques.
Inference for Least Squares Lines
Lecture 8 ARIMA Forecasting II
Chapter 6: Autoregressive Integrated Moving Average (ARIMA) Models
TM 745 Forecasting for Business & Technology Dr. Frank Joseph Matejcik
Statistics 153 Review - Sept 30, 2008
Model Building For ARIMA time series
CHAPTER 16 ECONOMIC FORECASTING Damodar Gujarati
CH2 Time series.
BOX JENKINS (ARIMA) METHODOLOGY
Chap 7: Seasonal ARIMA Models
Presentation transcript:

Part II – TIME SERIES ANALYSIS C5 ARIMA (Box-Jenkins) Models © Angel A. Juan & Carles Serrat - UPC 2007/2008

2.5.1: Introduction to ARIMA models The Autoregressive Integrated Moving Average (ARIMA) models, or Box-Jenkins methodology, are a class of linear models that is capable of representing stationary as well as nonstationary time series. ARIMA models rely heavily on autocorrelation patterns in data  both ACF and PACF are used to select an initial model. The Box-Jenkins methodology uses an iterative approach: An initial model is selected, from a general class of ARIMA models, based on an examination of the TS and an examination of its autocorrelations for several time lags The chosen model is then checked against the historical data to see whether it accurately describes the series: the model fits well if the residuals are generally small, randomly distributed, and contain no useful information. If the specified model is not satisfactory, the process is repeated using a new model designed to improve on the original one. Once a satisfactory model is found, it can be used for forecasting. Recall that stationary processes vary about a fixed level, and nonstationary processes have no natural constant mean level. The ACF and PACF associated to the TS are matched with the theoretical autocorrelation pattern associated with a particular ARIMA model.

2.5.2: Autoregressive Models AR(p) A pth-order autoregressive model, or AR(p), takes the form: Autoregressive models are appropriate for stationary time series, and the coefficient Ф0 is related to the constant level of the series. Theoretical behavior of the ACF and PACF for AR(1) and AR(2) models: An AR(p) model is a regression model with lagged values of the dependent variable in the independent variable positions, hence the name autoregressive model. AR(1) AR(2) ACF  0 PACF = 0 for lag > 1 ACF  0 PACF = 0 for lag > 2

2.5.3: Moving Average Models MA(q) A qth-order moving average model, or MA(q), takes the form: MA models are appropriate for stationary time series. The weights ωi do not necessarily sum to 1 and may be positive or negative. Theoretical behavior of the ACF and PACF for MA(1) and MA(2) models: The term Moving Average is historical and should not be confused with the moving average smoothing procedures. An MA(q) model is a regression model with the dependent variable, Yt, depending on previous values of the errors rather than on the variable itself. MA(1) MA(2) ACF = 0 for lag > 1; PACF  0 ACF = 0 for lag > 2; PACF  0

2.5.4: ARMA(p,q) Models ACF PACF AR(p) MA(q) ARMA(p,q) A model with autoregressive terms can be combined with a model having moving average terms to get an ARMA(p,q) model: ARMA(p,q) models can describe a wide variety of behaviors for stationary time series. Theoretical behavior of the ACF and PACF for autoregressive-moving average processes: Note that: ARMA(p,0) = AR(p) ARMA(0,q) = MA(q) In practice, the values of p and q each rarely exceed 2. ACF PACF AR(p) Die out Cut off after the order p of the process MA(q) Cut off after the order q of the process ARMA(p,q) In this context… “Die out” means “tend to zero gradually” “Cut off” means “disappear” or “is zero”

2.5.5: Building an ARIMA model (1/2) The first step in model identification is to determine whether the series is stationary. It is useful to look at a plot of the series along with the sample ACF. If the series is not stationary, it can often be converted to a stationary series by differencing: the original series is replaced by a series of differences and an ARMA model is then specified for the differenced series (in effect, the analyst is modeling changes rather than levels) Models for nonstationary series are called Autoregressive Integrated Moving Average models, or ARIMA(p,d,q), where d indicates the amount of differencing. Once a stationary series has been obtained, the analyst must identify the form of the model to be used by comparing the sample ACF and PACF to the theoretical ACF and PACF for the various ARIMA models. Principle of parsimony: “all things being equal, simple models are preferred to complex models” Once a tentative model has been selected, the parameters for that model are estimated using least squares estimates. A nonstationary TS is indicated if the series appears to grow or decline over time and the sample ACF fail to die out rapidly. In some cases, it may be necessary to difference the differences before stationary data are obtained. Note that: ARIMA(p,0,q) = ARMA(p,q) By counting the number of significant sample autocorrelations and partial autocorrelations, the orders of the AR and MA parts can be determined. Advice: start with a model containing few rather than many parameters. The need for additional parameters will be evident from an examination of the residual ACF and PACF.

2.5.5: Building an ARIMA model (2/2) Before using the model for forecasting, it must be checked for adequacy. Basically, a model is adequate if the residuals cannot be used to improve the forecasts, i.e., The residuals should be random and normally distributed The individual residual autocorrelations should be small. Significant residual autocorrelations at low lags or seasonal lags suggest the model is inadequate After an adequate model has been found, forecasts can be made. Prediction intervals based on the forecasts can also be constructed. As more data become available, it is a good idea to monitor the forecast errors, since the model must need to be reevaluated if: The magnitudes of the most recent errors tend to be consistently larger than previous errors, or The recent forecast errors tend to be consistently positive or negative Seasonal ARIMA (SARIMA) models contain: Regular AR and MA terms that account for the correlation at low lags Seasonal AR and MA terms that account for the correlation at the seasonal lags Many of the same residual plots that are useful in regression analysis can be developed for the residuals from an ARIMA model (histogram, normal probability plot, time sequence plot, etc.) In general, the longer the forecast lead time, the larger the prediction interval (due to greater uncertainty) In addition, for nonstationary seasonal series, an additional seasonal difference is often required

2.5.6: ARIMA with Minitab – Ex. 1 (1/4) File: PORTFOLIO_INVESTMENT.MTW Stat > Time Series > … A consulting corporation wants to try the Box-Jenkins technique for forecasting the Transportation Index of the Dow Jones. The series show an upward trend. The first several autocorrelations are persistently large and trailed off to zero rather slowly  a trend exists and this time series is nonstationary (it does not vary about a fixed level) Idea: to difference the data to see if we could eliminate the trend and create a stationary series.

2.5.6: ARIMA with Minitab – Ex. 1 (2/4) First order differences. A plot of the differenced data appears to vary about a fixed level. Comparing the autocorrelations with their error limits, the only significant autocorrelation is at lag 1. Similarly, only the lag 1 partial autocorrelation is significant. The PACF appears to cut off after lag 1, indicating AR(1) behavior. The ACF appears to cut off after lag 1, indicating MA(1) behavior  we will try: ARIMA(1,1,0) and ARIMA(0,1,1) A constant term in each model will be included to allow for the fact that the series of differences appears to vary about a level greater than zero.

2.5.6: ARIMA with Minitab – Ex. 1 (3/4) The LBQ statistics are not significant as indicated by the large p-values for either model.

2.5.6: ARIMA with Minitab – Ex. 1 (4/4) Finally, there is no significant residual autocorrelation for the ARIMA(1,1,0) model. The results for the ARIMA(0,1,1) are similar. Therefore, either model is adequate and provide nearly the same one-step-ahead forecasts.

2.5.7: ARIMA with Minitab – Ex. 2 (1/3) File: READINGS.MTW Stat > Time Series > … A consulting corporation wants to try the Box-Jenkins technique for forecasting a process. The time series of readings appears to vary about a fixed level of around 80, and the autocorrelations die out rapidly toward zero  the time series seems to be stationary. The first sample ACF coefficient is significantly different form zero. The autocorrelation at lag 2 is close to significant and opposite in sign from the lag 1 autocorrelation. The remaining autocorrelations are small. This suggests either an AR(1) model or an MA(2) model. The first PACF coefficient is significantly different from zero, but none of the other partial autocorrelations approaches significance, This suggests an AR(1) or ARIMA(1,0,0)

2.5.7: ARIMA with Minitab – Ex. 2 (2/3) A constant term is included in both models to allow for the fact that the readings vary about a level other than zero. MA(2) = ARIMA(0,0,2) AR(1) = ARIMA(1,0,0) Both models appear to fit the data well. The estimated coefficients are significantly different from zero and the mean square (MS) errors are similar. Let’s take a look at the residuals ACF…

2.5.7: ARIMA with Minitab – Ex. 2 (3/3) Finally, there is no significant residual autocorrelation for the ARIMA(1,0,0) model. The results for the ARIMA(0,0,2) are similar. Therefore, either model is adequate and provide nearly the same three-step-ahead forecasts. Since the AR(1) model has two parameters (including the constant term) and the MA(2) model has three parameters, applying the principle of parsimony we would use the simpler AR(1) model to forecast future readings.