forecasting_moving average
DESCRIPTION
A statement about the future value of a variable of interest such as demand.TRANSCRIPT
![Page 1: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/1.jpg)
FORECASTINGJefferson Palma CIE, AAE
![Page 2: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/2.jpg)
LLEARNINGEARNING OOBJECTIVESBJECTIVES
�List the elements of a good forecast.�Outline the steps in the forecasting process.�Describe averaging techniques, trend and
seasonal techniques, and regression analysis, and solve typical problems.
�Explain three measures of forecast accuracy.�Compare two ways of evaluating and
controlling forecasts.�Assess the major factors and trade-offs to
consider when choosing a forecasting technique.
![Page 3: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/3.jpg)
FORECAST:� A statement about the future value of a variable of
interest such as demand.
� Forecasts affect decisions and activities throughout an organization� Accounting, finance� Accounting, finance� Human resources� Marketing� MIS� Operations� Product / service design
![Page 4: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/4.jpg)
UUSESSES OFOF FFORECASTSORECASTS
Accounting Cost/profit estimates
Finance Cash flow and funding
Human Resources Hiring/recruiting/training
Marketing Pricing, promotion, strategy
MIS IT/IS systems, services
Operations Schedules, MRP, workloads
Product/service design New products and services
![Page 5: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/5.jpg)
�Assumes causal systempast ==> future
�Forecasts rarely perfect because of randomness
�Forecasts more accurate for�Forecasts more accurate forgroups vs. individuals
�Forecast accuracy decreases as time horizon increases
I see that you willget an A this semester.
![Page 6: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/6.jpg)
EELEMENTSLEMENTS OFOF AA GGOODOOD FFORECASTORECAST
Timely
AccurateReliable
Written
![Page 7: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/7.jpg)
SSTEPSTEPS ININ THETHE FFORECASTINGORECASTING PPROCESSROCESS
“The forecast”
Step 1 Determine purpose of forecast
Step 2 Establish a time horizon
Step 3 Select a forecasting techniqueStep 4 Gather and analyze data
Step 5 Prepare the forecast
Step 6 Monitor the forecast
![Page 8: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/8.jpg)
TTYPESYPES OFOF FFORECASTSORECASTS
�Judgmental - uses subjective inputs
�Time series - uses historical data assuming the future will be like the past
�Associative models - uses explanatory variables to predict the futurevariables to predict the future
![Page 9: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/9.jpg)
JJUDGMENTALUDGMENTAL FFORECASTSORECASTS
�Executive opinions
�Sales force opinions
�Consumer surveys
�Outside opinion
�Delphi method
� Opinions of managers and staff
� Achieves a consensus forecast
![Page 10: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/10.jpg)
TTIMEIME SSERIESERIES FFORECASTSORECASTS
�Trend - long-term movement in data�Seasonality - short-term regular variations in
data�Cycle – wavelike variations of more than one
year’s durationyear’s duration� Irregular variations - caused by unusual
circumstances�Random variations - caused by chance
![Page 11: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/11.jpg)
FFORECASTORECAST VVARIATIONSARIATIONS
Trend
Irregularvariation
Figure 3.1
Seasonal variations
908988
Cycles
![Page 12: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/12.jpg)
NNAIVEAIVE FFORECASTSORECASTS
Uh, give me a minute.... We sold 250 wheels lastweek.... Now, next week we should sell....we should sell....
The forecast for any period equals the previous period’s actual value.
![Page 13: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/13.jpg)
NNAÏVEAÏVE FFORECASTSORECASTS
�Simple to use�Virtually no cost�Quick and easy to prepare�Data analysis is nonexistent�Easily understandable�Easily understandable�Cannot provide high accuracy�Can be a standard for accuracy
![Page 14: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/14.jpg)
UUSESSES FORFOR NNAÏVEAÏVE FFORECASTSORECASTS
�Stable time series data� F(t) = A(t-1)
�Seasonal variations� F(t) = A(t-n)
�Data with trends�Data with trends� F(t) = A(t-1) + (A(t-1) – A(t-2))
![Page 15: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/15.jpg)
TECHNIQUES FOR AVERAGING
�Moving average
�Weighted moving average
�Exponential smoothing
![Page 16: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/16.jpg)
MOVING AVERAGES
�Moving average – A technique that averages a number of recent actual values, updated as new values become available.
MA =Ai
i = 1∑n
�Weighted moving average – More recent values in a series are given more weight in computing the forecast.
MAn =n
i = 1
![Page 17: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/17.jpg)
SSIMPLEIMPLE MMOVINGOVING AAVERAGEVERAGE
39
41
43
45
47
ActualMA5
MAn =n
Aii = 1∑n
35
37
39
1 2 3 4 5 6 7 8 9 10 11 12
MA3
![Page 18: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/18.jpg)
EEXPONENTIALXPONENTIAL SSMOOTHINGMOOTHING
• Premise--The most recent observations might have the highest predictive value.
Ft = Ft-1 + α(At-1 - Ft-1)
have the highest predictive value.
� Therefore, we should give more weight to the more recent time periods when forecasting.
![Page 19: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/19.jpg)
EEXPONENTIALXPONENTIAL SSMOOTHINGMOOTHING
�Weighted averaging method based on previous forecast plus a percentage of the forecast error
Ft = Ft-1 + α(At-1 - Ft-1)
forecast plus a percentage of the forecast error�A-F is the error term, � is the % feedback
![Page 20: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/20.jpg)
Period Actual Alpha = 0.1 Error Alpha = 0.4 Error1 422 40 42 -2.00 42 -23 43 41.8 1.20 41.2 1.84 40 41.92 -1.92 41.92 -1.925 41 41.73 -0.73 41.15 -0.15
Exponential Exponential SmoothingSmoothing
5 41 41.73 -0.73 41.15 -0.156 39 41.66 -2.66 41.09 -2.097 46 41.39 4.61 40.25 5.758 44 41.85 2.15 42.55 1.459 45 42.07 2.93 43.13 1.87
10 38 42.36 -4.36 43.88 -5.8811 40 41.92 -1.92 41.53 -1.5312 41.73 40.92
![Page 21: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/21.jpg)
PPICKINGICKING AA SSMOOTHINGMOOTHING CCONSTANTONSTANT
40
45
50D
eman
d α = .1α = .4
Actual
35
40
1 2 3 4 5 6 7 8 9 10 11 12
Period
Dem
and
![Page 22: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/22.jpg)
TECHNIQUES FOR AVERAGING
� Linear Trend Equation
� Trend Adjusted Exponential Smoothing
![Page 23: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/23.jpg)
CCOMMONOMMON NNONLINEARONLINEAR TTRENDSRENDS
Parabolic
Figure 3.5
Exponential
Growth
![Page 24: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/24.jpg)
LLINEARINEAR TTRENDREND EEQUATIONQUATION
Ft = a + bt
0 1 2 3 4 5 t
Ft
�Ft = Forecast for period t� t = Specified number of time periods�a = Value of Ft at t = 0�b = Slope of the line
0 1 2 3 4 5 t
![Page 25: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/25.jpg)
CCALCULATINGALCULATING AA ANDAND BB
b =n (ty) - t y
n t2 - ( t)2∑∑∑
∑∑
a =y - b t
n∑∑
![Page 26: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/26.jpg)
LLINEARINEAR TTRENDREND EEQUATIONQUATION EEXAMPLEXAMPLE
t yWeek t2 Sales ty
1 1 150 1502 4 157 3143 9 162 4863 9 162 4864 16 166 6645 25 177 885
Σ t = 15 Σ t2 = 55 Σ y = 812 Σ ty = 2499(Σ t)2 = 225
![Page 27: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/27.jpg)
LLINEARINEAR TTRENDREND CCALCULATIONALCULATION
b =5 (2499) - 15(812)
5(55) - 225=
12495-12180
275 -225= 6.3
y = 143.5 + 6.3t
a =812 - 6.3(15)
5= 143.5
![Page 28: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/28.jpg)
AASSOCIATIVESSOCIATIVE FFORECASTINGORECASTING
�Predictor variables - used to predict values of variable interest
�Regression - technique for fitting a line to a set of points
Least squares line - minimizes sum of squared �Least squares line - minimizes sum of squared deviations around the line
![Page 29: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/29.jpg)
LLINEARINEAR MMODELODEL SSEEMSEEMS RREASONABLEEASONABLE
30
40
50
X Y7 152 106 134 15
14 25
Computedrelationship
A straight line is fitted to a set of sample points.
0
10
20
0 5 10 15 20 25
15 2716 2412 2014 2720 4415 347 17
![Page 30: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/30.jpg)
FFORECASTORECAST AACCURACYCCURACY
�Error - difference between actual value and predicted value
�Mean Absolute Deviation (MAD)
� Average absolute error� Average absolute error
�Mean Squared Error (MSE)
� Average of squared error
�Mean Absolute Percent Error (MAPE)
� Average absolute percent error
![Page 31: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/31.jpg)
MAD, MSE, MAD, MSE, ANDAND MAPEMAPE
MAD =Actual forecast−∑
n
MSE =Actual forecast)
2−∑ (
MSE =-1
∑
n
MAPE =Actual forecas
t−
n
/ Actual*100)∑(
![Page 32: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/32.jpg)
EEXAMPLEXAMPLE 1010
Period Actual Forecast (A-F) |A-F| (A-F)^2 (|A-F|/Actual) *1001 217 215 2 2 4 0.922 213 216 -3 3 9 1.413 216 215 1 1 1 0.464 210 214 -4 4 16 1.905 213 211 2 2 4 0.946 219 214 5 5 25 2.287 216 217 -1 1 1 0.468 212 216 -4 4 16 1.89
-2 22 76 10.26
MAD= 2.75MSE= 10.86
MAPE= 1.28
![Page 33: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/33.jpg)
CONTROLLING THE FORECAST
�Control chart
� A visual tool for monitoring forecast errors
� Used to detect non-randomness in errors
�Forecasting errors are in control if�Forecasting errors are in control if
� All errors are within the control limits
� No patterns, such as trends or cycles, are
present
![Page 34: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/34.jpg)
SSOURCESOURCES OFOF FFORECASTORECAST ERRORSERRORS
�Model may be inadequate� Irregular variations� Incorrect use of forecasting technique
![Page 35: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/35.jpg)
TTRACKINGRACKING SSIGNALIGNAL
Tracking signal = (Actual-forecast)∑
•Tracking signal
–Ratio of cumulative error to MAD
Tracking signal =MAD
Bias – Persistent tendency for forecasts to beGreater or less than actual values.
![Page 36: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/36.jpg)
CHOOSING A FORECASTING TECHNIQUE
�No single technique works in every situation�Two most important factors
� Cost� Accuracy
�Other factors include the availability of:�Other factors include the availability of:� Historical data� Computers� Time needed to gather and analyze the data� Forecast horizon
![Page 37: Forecasting_Moving Average](https://reader031.vdocuments.us/reader031/viewer/2022020722/577cc1171a28aba711923e5d/html5/thumbnails/37.jpg)
ENDEND