Chapter 4: Seasonal Series: Forecasting and Decomposition

Slides:



Advertisements
Similar presentations
Decomposition Method.
Advertisements

©The McGraw-Hill Companies, Inc. 2008McGraw-Hill/Irwin Lesson 12.
Part II – TIME SERIES ANALYSIS C3 Exponential Smoothing Methods © Angel A. Juan & Carles Serrat - UPC 2007/2008.
Exponential Smoothing Methods
Time Series Analysis Autocorrelation Naive & Simple Averaging
Ka-fu Wong © 2003 Chap Dr. Ka-fu Wong ECON1003 Analysis of Economic Data.
Moving Averages Ft(1) is average of last m observations
Analyzing and Forecasting Time Series Data
Chapter 5 Time Series Analysis
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.
Using a Centered Moving Average to Extract the Seasonal Component of a Time Series If we are forecasting with say, quarterly time series data, a 4-period.
MOVING AVERAGES AND EXPONENTIAL SMOOTHING
Chapter 11 Multiple Regression.
Time series Decomposition Additive Model Farideh Dehkordi-Vakil.
McGraw-Hill/Irwin Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved. Time Series and Forecasting Chapter 16.
Time Series and Forecasting
Slides 13b: Time-Series Models; Measuring Forecast Error
CHAPTER 18 Models for Time Series and Forecasting
1 1 Slide © 2009 South-Western, a part of Cengage Learning Chapter 6 Forecasting n Quantitative Approaches to Forecasting n Components of a Time Series.
Slides by John Loucks St. Edward’s University.
1 1 Slide © 2011 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
TIME SERIES by H.V.S. DE SILVA DEPARTMENT OF MATHEMATICS
1 1 Slide © 2012 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
Forecasting supply chain requirements
Copyright © 2015 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
© 2004 Prentice-Hall, Inc. Chapter 7 Demand Forecasting in a Supply Chain Supply Chain Management (2nd Edition) 7-1.
Chapter 17 Time Series Analysis and Forecasting ©.
McGraw-Hill/Irwin Copyright © 2007 by The McGraw-Hill Companies, Inc. All rights reserved. Time Series Forecasting Chapter 13.
Time-Series Forecasting Learning Objectives 1.Describe What Forecasting Is 2. Forecasting Methods 3.Explain Time Series & Components 4.Smooth a Data.
MBA.782.ForecastingCAJ Demand Management Qualitative Methods of Forecasting Quantitative Methods of Forecasting Causal Relationship Forecasting Focus.
Time Series Analysis and Forecasting
Inference for Regression Simple Linear Regression IPS Chapter 10.1 © 2009 W.H. Freeman and Company.
Time series Decomposition Farideh Dehkordi-Vakil.
Week 11 Introduction A time series is an ordered sequence of observations. The ordering of the observations is usually through time, but may also be taken.
1 1 Slide © 2011 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
© 1999 Prentice-Hall, Inc. Chap Chapter Topics Component Factors of the Time-Series Model Smoothing of Data Series  Moving Averages  Exponential.
COMPLETE BUSINESS STATISTICS
1 BABS 502 Moving Averages, Decomposition and Exponential Smoothing Revised March 14, 2010.
Time Series and Forecasting
Irwin/McGraw-Hill © Andrew F. Siegel, 1997 and l Chapter 14 l Time Series: Understanding Changes over Time.
Forecasting is the art and science of predicting future events.
MBF1413 | Quantitative Methods Prepared by Dr Khairul Anuar 8: Time Series Analysis & Forecasting – Part 1
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 15 Forecasting. Forecasting Methods n Forecasting methods can be classified as qualitative or quantitative. n Such methods are appropriate when.
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 3 Lect 6 Forecasting. Seasonality – Repetition at Fixed Intervals Seasonal variations –Regularly repeating movements in series values that can.
TIME SERIES MODELS. Definitions Forecast is a prediction of future events used for planning process. Time Series is the repeated observations of demand.
©The McGraw-Hill Companies, Inc. 2008McGraw-Hill/Irwin Time Series and Forecasting Chapter 16.
Yandell – Econ 216 Chap 16-1 Chapter 16 Time-Series Forecasting.
Financial Analysis, Planning and Forecasting Theory and Application
Short-Term Forecasting
TIME SERIES ANALYSIS.
Chapter Nineteen McGraw-Hill/Irwin
What is Correlation Analysis?
Demand Management and Forecasting
Chapter 6: Autoregressive Integrated Moving Average (ARIMA) Models
“The Art of Forecasting”
John Loucks St. Edward’s University . SLIDES . BY.
Cycles and Exponential Smoothing Models
MOVING AVERAGES AND EXPONENTIAL SMOOTHING
MBF1413 | Quantitative Methods Prepared by Dr Khairul Anuar
The Simple Linear Regression Model: Specification and Estimation
State-Space Models for Time Series
Chapter 8 Supplement Forecasting.
Chapter Nineteen McGraw-Hill/Irwin
1 1 Slide © 2013 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in whole.
Chap 4: Exponential Smoothing
Chap 7: Seasonal ARIMA Models
Exponential Smoothing
Presentation transcript:

Chapter 4: Seasonal Series: Forecasting and Decomposition 4.1 Components of a Time Series 4.2 Forecasting Purely Seasonal Series 4.3 Forecasting Using a Seasonal Decomposition 4.4 Pure Decomposition 4.5 The Census X-12 Decomposition 4.6 The Holt-Winters Seasonal Smoothing Methods 4.7 The Multiplicative Holt-Winters Method 4.8 Weekly Data 4.9 Prediction Intervals 4.10 Principles

4.1: Components of a Time Series A time series is said to have a seasonal component if it displays a recurrent pattern with a fixed and known duration [e.g. months of the year, days of the week]. A time series is said to have a cyclical component if it displays somewhat regular fluctuations about the trend but those fluctuations have a periodicity of variable and unknown duration [e.g. a business cycle].

4.1: Components of a Time Series Define T = Trend, S = Seasonal, E = Error Additive model: Multiplicative model: Mixed additive-multiplicative model:

4.1: Components of a Time Series

4.2: Forecasting Purely Seasonal Series Purely seasonal series usually occur only because the series has been decomposed into separate trend and seasonal components. Use an update from the previous seasonal value (e.g. July to July or Monday to Monday). If m denotes the number of seasons [m=12 for monthly, etc.], the forecasts function is: Essentially similar to SES, save that we update one sub-series m time periods; also we use a common smoothing parameter. Note that m starting values are required.

Figure 4.2: Monthly Temperatures in Boulder Source: Earth System Research Laboratory, Physical Sciences Division, National Oceanic and Atmospheric Administration (www.cdc.noaa.gov/Boulder/Boulder.mm.html). Data shown is from file Boulder.xlsx.

Figure 4.3: Plot of Seasonal Patterns for Boulder Data ©Cengage Learning 2013.

Table 4.2: Summary Results for Boulder Temperature Data Optimal γ=0.177 Also consider γ=0 [long-run average] and γ=1 [last year]

4.3: Forecasting Using a Seasonal Decomposition The basic steps in the process are: Generate estimates of the trend by averaging out the seasonal component. Estimate the seasonal component. Create a deseasonalized series. Forecast the trend and the seasonal pattern separately. Recombine the trend forecast with the seasonal component to produce a forecast for the original series.

4.3: Forecasting Using a Seasonal Decomposition Pure Decomposition - use all n data points to estimate the fitted value at each time t (two-sided decomposition) Forecasting Decomposition - use only the data up to and including time t to predict value at time t+1 (one-sided decomposition) Q: Under what circumstances is one approach preferable to the other?

4.3: Forecasting Using a Seasonal Decomposition Moving Average A (simple) moving average of order K, denoted by MA(K), is the average of K successive terms in a time series, taken over successive sets of K, so that the first average is , the second is and so on.

4.3: Forecasting Using a Seasonal Decomposition Centered Moving Average A centered moving average (CMA) of order K, denoted by CM(t|K), is defined by taking the average of successive pairs of simple moving averages. Thus, denote the first average of K terms by and the second one by then define the centered MA as When K is odd, it is usually not necessary to calculate a CMA. However, when K is even, K=2J say, the first average corresponds to an “average time” of t-J-0.5 and the second to t-J+0.5 so the centered MA corresponds to time t-J as desired.

Table 4.3: Forecasting Using a Seasonal Decomposition

Table 4.3: Forecasting Using a Seasonal Decomposition

4.4: Pure Decomposition Calculate the 4-term MA and then the centered MA. The first CMA term corresponds to period 3 and the last one to period (t-2); K = 4 so we “lose” two values at each end of the series. Divide [subtract] observations 3, …, (t-2) by [from] their corresponding CMA to obtain a detrended series. Calculate the average value (across years) of the detrended series for each quarter j (j = 1, 2, 3, 4) to produce the initial seasonal factors. Standardize the seasonal factors by computing their average and then setting the final seasonal factor equal to the initial value divided by [minus] the overall average. Estimate the error term by dividing the detrended series by the final seasonal factor [subtracting the final seasonal factor from the detrended series].

4.4: Pure Decomposition Table 4.4: Multiplicative Decomposition

4.5: The Census X-12 Decomposition Brief outline of steps: Identify anomalous observations and make any necessary adjustments. Use a (usually multiplicative) decomposition to estimate preliminary seasonal factors. Apply an initial set of seasonal adjustments using [more complex] moving averages. Extend the series using ARIMA so that moving averages can be calculated up to the end of the series. Apply a final round of detrending and then estimate the components.

4.6: The Holt-Winters Seasonal Smoothing Methods The forecast function includes both local trend and seasonal components. The additive form is: where is local estimate of level at time t. is local estimate of trend at time t. is the local estimate of the seasonal effect [from same ‘month’ last year].

4.6.1: The Additive Holt-Winters Method When a new observation is recorded, the complete set of updating equations takes the form: Observed error: 1-step ahead forecast: Updating relationships:

4.6.1: The Additive Holt-Winters Method The updating equations may be expressed in error correction form:

4.6.1: The Additive Holt-Winters Method A variety of special cases exists: Fixed seasonal pattern: γ = 0 (no seasonal updating) No seasonal pattern: γ = 0 and all initial S values are set equal to zero Fixed trend: β = 0 Zero trend: β = 0 and T0 = 0 All fixed components: α = β = γ = 0 [leading to the regression model in Section 9.1.1]

4.6: The Holt-Winters Seasonal Smoothing Methods Starting Values Can use the Method of Least Squares to estimate {, , } BUT Need m+2 starting values [=14 for monthly data]; problematic for short series, so various heuristic methods are used for the starting values. Q: Suppose you has three years of monthly data. How would you specify starting values for the level, trend and seasonal components?

Example 4.4 [Table 4.5]

Example 4.5: U.S. Auto Sales Original series Components

4.6: The Holt-Winters Seasonal Smoothing Methods Q: Which method would you use?

4.7: The Multiplicative Holt-Winters Method The forecast function includes the local trend and seasonal components as before, but the seasonal effect is now multiplicative: The error correction updating relationships are:

Plot suggests similar multiplicative seasonal factors. 4.8: Weekly Data Often working with one or two years of data, yet need 52 “seasonal values”. Proceed by estimating common (multiplicative seasonal values for similar products. Deseasonalize the series and apply SES to each series. Plot suggests similar multiplicative seasonal factors.

4.9: Prediction Intervals As before, we can define a prediction interval using the normal distribution as: Q: These intervals are very wide. Why?

4.10: Principles Examine carefully the structure of the seasonal pattern. When there are limited data from which to calculate seasonal components, consider using the seasonals from elated data series, such as the aggregate. Alternatively, average the estimates across similar products.

Take-Aways Seasonal series need to be decomposed into trend and seasonal components. The type of decomposition used depends upon whether we are interested in pure forecasting or smoothing. The seasonal component may have either an additive or a multiplicative interaction with the trend. Short series mean that special methods involving the assumption of common seasonal patterns must be used.