Example of a prediction for a general linear model
main topic     interpreting results     session command     see also 

As part of a test of solar thermal energy, you measured the total heat flux from homes. In the multiple regression example, you determined that the total heat flux (HeatFlux) can be predicted by the position of the focal points in the east, south, and north directions. Now, you want to use this model to predict HeatFlux for specific values of the predictors East, South, and North.

You do not need to re-analyze the model. The worksheet contains the stored model that is necessary for the prediction. This example uses a regression model. However, the principles are the same for an analysis of variance model.

1    Open the worksheet SOLARENERGY_MODEL.MTW.

2    Choose Stat > Regression > Regression > Predict.

3    In Response, choose HeatFlux.

4    In the second drop-down list, choose Enter individual values.

5    In the variables table, enter the setting for each variable as shown below.

6    Click OK.

Session Window Output

Interpreting the results

Minitab uses the equation obtained from the stored model to calculate that the predicted response value (fit) for the specified predictor values is 206.783.

Additionally, the confidence interval indicates that you can be 95% confident that the mean of the heat flux at these predictor values is between 198.247 and 215.320. The prediction interval indicates that you can be 95% confident that a single new observation will fall between 187.126 and 226.441.

The prediction interval is always wider than the corresponding confidence interval because of the added uncertainty involved in predicting a single response versus the mean response.

Keep in mind that this prediction is based on a model equation. You must be sure that your model is adequate before interpreting the predictions.