BSAD 30 Test 3
mean absolute percentage error (MAPE)
average of absolute values of the percentage forecast errors
mean squared error (MSE)
average of sum of squared forecast errors; total squared error/number of periods
exponential trend line
increasing at increasing rate
negative forecast error
indicates forecasting method overestimated the actual value
positive forecast error
indicates forecasting method underestimated actual value
Calculate slack
LS - ES LF - EF
forecast error
actual value - forecast
cyclical pattern
exists if time series plot shows alternating sequence of points below and above trend line lasting more than one year
horizontal pattern
exists when data fluctuate around constant mean
slack
length of time an activity can be delayed without delaying the entire project
asymptotic trend line
line increasing at decreasing rate
mean absolute error (MAE)
mean of absolute values of forecast errors
mean error
not very useful; simple measure of forecast accuracy
seasonal patterns
recognized by seeing same repeating pattern of highs and lows over successive periods of time within one year ex. chocolate sales in Feb.
weighted moving averages
selecting different weight for the k most recent data values in the time series and computing a weighted average of values; sum must equal 1
trend pattern
time series may show gradual shifts or movements to relatively higher/lower values over longer period of time; linear or non linear
moving averages method
uses average of most recent k data values in time series; each observation receives same weight
exponential smothing
uses weighted average of past time series values as the forecast; special case of weighted moving averages method