Scribd
Upload
79 views20 pages

Ch07 - Forecast

This document introduces time series forecasting and its key components. It discusses that forecasting predicts future values and is important for business planning. The main components of a time series that affect forecasting are trends, seasonality, cyclical patterns, and randomness. The document outlines the steps in time series forecasting which include identifying these components, selecting a forecasting technique, and preparing forecasts. It also describes stationary and moving average forecasting models that do not assume trends over time.

Uploaded by

Aya_Noah
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
79 views20 pages

Ch07 - Forecast

This document introduces time series forecasting and its key components. It discusses that forecasting predicts future values and is important for business planning. The main components of a time series that affect forecasting are trends, seasonality, cyclical patterns, and randomness. The document outlines the steps in time series forecasting which include identifying these components, selecting a forecasting technique, and preparing forecasts. It also describes stationary and moving average forecasting models that do not assume trends over time.

Uploaded by

Aya_Noah
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 20

9.

1 Introduction to Time Series Forecasting


Forecasting is the process of predicting the future. Forecasting is an integral part of almost all business enterprises. Examples
Manufacturing firms forecast demand for their product, to schedule manpower and raw material allocation. Service organizations forecast customer arrival patterns to maintain adequate customer service.

Introduction
More examples
Security analysts forecast revenues, profits, and debt ratios, to make investment recommendations. Firms consider economic forecasts of indicators (housing starts, changes in gross national profit) before deciding on capital investments.

Introduction
Good forecasts can lead to
Reduced inventory costs. Lower overall personnel costs. Increased customer satisfaction.

The forecasting process can be based on:


Educated guess. Expert opinions. Past history of data values, known as a time

series.

Components of a Time Series


Long-term trend
A time series may be stationary or exhibit trend over time. Long-term trend is typically modeled as a linear, quadratic or exponential function.

Seasonal variation
When a repetitive pattern is observed over some time horizon, the series is said to have seasonal behavior. Seasonal effects are usually associated with calendar or climatic changes. Seasonal variation is frequently tied to yearly cycles.
4

Components of a Time Series


Cyclical variation
An upturn or downturn not tied to seasonal variation. Usually results from changes in economic conditions.

Random effects

Time series value Linear trend and seasonality time series

Components of a Time Series

Linear trend time series A stationary time series

Future

Time

The goal of a time series forecast is to identify factors that can be predicted. This is a systematic approach involving the following steps.
Step 1: Hypothesize a form for the time series model. Step 2: Select a forecasting technique. Step 3: Prepare a forecast.
7

Steps in the Time Series Forecasting Process

Steps in the Time Series Forecasting Process


Step 1: Identify components included in the time series

Collect historical data. Graph the data vs. time. Hypothesize a form for the time series model. Verify this hypothesis statistically.
8

Steps in the Time Series Forecasting Process


Step 2: Select a Forecasting Technique
Select a forecasting technique from among several techniques available. The selection includes
Determination of input parameter values Performance evaluation on past data of each technique

Step 3: Prepare a Forecast using the selected technique

7.2 Stationary Forecasting Models


In a stationary model the mean value of the time series is assumed to be constant.
The values of t The general form of such a model is are assumed to be independent yt = 0 + t The values Where: of yt = the value of the time series at timet period t. 0 = the unchanged mean value of are time series. the assumed to t = a random error term at time period t. have a 10 mean of 0.

Stationary Forecasting Models Checking for trend


Use Linear Regression if t is normally distributed. Use a nonparametric test if t is not normally distributed. Autocorrelation measures the relationship between the values of the time series in different periods. Lag k autocorrelation measures the correlation between time series values which are k periods apart.
Autocorrelation between successive periods indicates a possible trend. Lag 7 autocorrelation indicates one week seasonality (daily data). Lag 12 autocorrelation indicates 12-month seasonality (monthly data).

Checking for seasonality component

Checking for Cyclical Components


11

Moving Average Methods


The last period technique The forecast for the next period is the last observed value. Ft+1 = yt
t t t+1 t+1 t+1 t+1 t t

12

Moving Average Methods


The moving average method
The forecast is the average of the last n observations yt + yt1 + ...+ ytn+1 of Ft+1 = the n t-2 t-2t-2 t-1 t-1 t t t t t t-2 t-1 t-1 t-1 t-2t-2 tt-1t+1 t+1 time series.
13

Moving Average Methods


The weighted moving average method
More recent values of the time series get larger weights than past values when performing the forecast.
F +1 = wt1yt + w2yt-1 +w3yt-2 + + wnyt-n+ 1 w1 w2 wn
wi = 1
14

Moving Average Methods


Forecasts for Future Time Periods The forecast for time period t+ 1 is the forecast for all future time periods:

Ft+k = yt for k = 12 3 ... , , ,


This forecast is revised only when new data becomes available.
15

YOHO BRAND YO - YOs Moving Average Methods Galaxy Industries is interested in forecasting weekly demand for its YoHo brand yo-yos. The yo-yo is a mature product. This year demand pattern is expected to repeat next year. To forecast next year demand, the past 52 weeks demand records were collected.

16

YOHO BRAND YO - YOs Moving Average Methods Three forecasting methods were suggested:
Last period technique - suggested by Amy Chang. Four-period moving average - suggested by Bob Gunther. Four-period weighted moving average - suggested by Carlos Gonzalez.

Management wants to determine:


If a stationary model can be used. What forecast will be obtained using each method?

17

YOHO BRAND YO YOs- Solution


Construct the time series plot
Wek D e a n ee em d n We k Dmad
1 1 2 2 3 3 4 4 5 5 6 6 7 7 8 8 9 9 10 10 11 11 12 12 13 13 4 15 15 4 2 36 36 2 3 48 48 3 2 72 72 2 2 80 80 2 3 95 95 3 4 38 38 4 4 31 31 4 4 46 46 4 3 54 54 3 5 29 29 5 2 41 41 2 2 62 62 2

Wek ee We k
14 4 1 15 5 1 16 6 1 17 7 1 18 8 1 19 9 1 20 0 2 21 1 2 22 2 2 23 3 2 24 4 2 25 5 2 26 6 2

Dean em d n Dmad
3 65 65 3 4 71 71 4 4 02 02 4 4 29 29 4 3 76 76 3 3 63 63 3 5 13 13 5 1 97 97 1 4 38 38 4 5 57 57 5 6 25 25 6 2 66 66 2 5 51 51 5

Wek ee We k
27 27 28 28 29 29 30 30 31 31 32 32 33 33 34 34 35 35 36 36 37 37 38 38 39 39

Dean em d n Dmad
35 1 1 35 38 8 8 38 33 6 6 33 41 4 4 41 34 6 6 34 25 2 2 25 25 6 6 25 37 8 8 37 39 1 1 39 21 7 7 21 42 7 7 42 29 3 3 29 28 8 8 28

Wek ee We k
4 0 40 4 1 41 4 2 42 4 3 43 4 4 44 4 5 45 4 6 46 4 7 47 4 8 48 4 9 49 5 0 50 5 1 51 5 2 52

Dean em d n Dmad
282 28 2 399 39 9 309 30 9 435 43 5 299 29 9 522 52 2 376 37 6 483 48 3 416 41 6 245 24 5 393 39 3 482 48 2 484 48 4

18

Neither seasonality nor cyclical effects can be observed

700 600 500 400 300 200 100 0


1 5 9 13 17 21 25 29 33 37 41 45 49

D em and

S eries 1

We eks
19

Is trend present?
Run linear regression to test yt= 0+ 1t+ t Excel results
C ef o f. Inec p tret We s ek S a d Er tn. r t tt -S a P au -v l e L w r9 % U p r9 % oe 5 pe 5
1

in the model

39 7 6 .2 0 39 .3 3

2 .7 4 6 1 .2 5 7 93 3 87 5 8 E-1 33 4 1 .4 0 1 6 1 0 6 8 0.71601 -1 9 1 .9 2 4 .3 5 6 0 1 0 .7 6 1 .4 9 9

45 9 2 .0 4 2 69 .1 6 9

This large P-value indicates that there is little evidence that trend exists

Conclusion: A stationary model is appropriate


20

You might also like