Home > Preview
The flashcards below were created by user
Cshowalter
on FreezingBlue Flashcards.

exponential smoothing
a weighted averaging method based on previous forecast plus a percentage of the forecast error. ( using less makes it more accurate)

the smaller the a (alpha) the more _______
stable

the larger the a (alpha) the more____
responsive

if your data has randomness and your data has trend these are good to use
double moving average and double exponential smoothing( trend adjusted exponential smoothing)

regression is good for measuring ______
trend

when you have randomness, trend, seasonality, and multi year data you would use
classical decomposition method or box Jenkins

seasonality
regularly repeating movements in series values that can be tried to recurring events

calculating seasonal indexes

if you have no seasonality the index number will be ____
1

cycles
are similar to seasonal variations but are of longer durations


Components of a time series
 trends tendency of a time series to move upwards or downwards over time
 cycle the effect on a time series caused by economic cycles
 seasonal short term regular variations in data caused by calendar related events
 irregular variations unpredictable and unexplainable fluctuations normal unusal circumstance

Involves all the _____ and control activites associated with the transformation process of both goods and service
planning

types of forecasting
 Judgmental uses subjective inputs
 Associative uses explanatory variables to predict furture
 Quantitive uses historical data assuming the future will be like the past

time series forecasts
forecasts the project patterns identified in recent time series observation

time series forecasts Navie forecasting
uses a single previous value of a time as the basis for a forecast

time series forecasting Techniques
 Moving avrg
 weighted moving avrg
 Exponential smoothing

Moving Average
 technique that averages a number of most recent actual values in generating forecast
 As new data becomes available, the forecast is updated by adding new values and dropping the oldest
 fewer data points more responsive
 more data points less responsive

Weighted Moving Average
The most recent values in a time series are given more weight in computing forecasts

Exponentiation Smoothing
 the most recent observations might have the highest predictive value
 we should give more weight to the more recent time periods

Trends Adjusted Exponential Smoothing
 Alpha and Beta are smoothing constants
 has the ability to respond to changes in trends
 has to components; smoothing error, trend factor

Model of seasonality
 Additive seasonality is expressed as quantity that gets added or subtracted
 Multiplicativeis expressed as a percentage of the average amount which is then used to multiply the value of a series

Seasonal relative
 the seasonal percentage used in the multiplicative seasonally adjusted forecasting model
 to deseasonalize datadone to get a clear picture of nonseasonal components of data series
 to incorporate seasonality in a forecast

Techniques for Cycles
cycles are similar to seasonal variation but are of longer durations

Regression a technique for fitting a line to a set of data points

ForecastedAccuracy and control
 forecaster want to minimize forecast errors
 forecast accurarcy should be an important forecasting technique selection criterion
 Error=Actual Foretasted

error
the difference between actual and predicted value

Operations Strategy
the better forecasts are , the more able companies will be to take advantage of future opportunities and reduce potential risks

two important aspects of forecasting

