Predictive Data Mining Models by David L. Olson & Desheng Wu

Author:David L. Olson & Desheng Wu
Language: eng
Format: epub
Publisher: Springer Singapore, Singapore

whereSSE sum of squared errors

MSRsum of squared predicted values

TSSSSE + MSR

nnumber of observations

knumber of independent variables.

Adjusted R2 can be used to select more robust regression models. We might get almost as much predictive accuracy without needing to gather these variables. Here Adj R Square goes up from 0.644 to 0.766, indicating that adding the two additional independent variables paid their way in additional explanatory power.

Using the model to forecast requires knowing (or guessing at) future independent variable values. A very good feature for Time is that there is no additional error introduced in estimating future time values. That is not the case for NYSE or for Brent. In Table 4.4 we guess at slight increases for both NYSE and Brent values, and compare the simple time regression with the multiple regression model.Table 4.4Forecasts

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.