Forecasted could goal identify the pattern of

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Goal: Identify the “pattern” of historical data Goal: and use this pattern to forecast the future and Simple or Composite Numbers Simple Seasonality Effects Types of Time Series Models Types Moving Average (MA) Moving Average last several data points Example: 3 period moving average model; Example: abbrev. MA(3) abbrev. Weighted Moving Average Apply different weights to past data pts. Ex., Weights of .5, .3, and .2; abbrev. Ex., Weighted MA(3) Weighted Our Data Our For Moving Averages… For Assessing Forecasting Accuracy Accuracy Best model is one with least error in Best prediction/forecast prediction/forecast Forecast error: Forecast The difference between an observed criterion The value and the predicted value (forecast) for that same point in time that 3 error types: MAD, Bias, RMSE Assessing Forecasting Accuracy Assessing MAD: Mean absolute deviation The average of the absolute values of the The absolute forecast errors of the Time Series forecast Should be as small as possible RMSE: Root mean squared error Equal to the square root of the mean squared Equal of the forecast errors of the TS of Should be as small as possible. Assessing Forecasting Accuracy Ac...
View Full Document

{[ snackBarMessage ]}

Ask a homework question - tutors are online