Example of a prediction for a stability study with a random batch factor
main topic     interpreting results     session command     see also 

In Example of a stability study with a random batch factor, you determine that the shelf life of a new medication is 53.18 months. The shelf life for this analysis is defined as the time at which you can no longer be 95% confident that 95% of the product is at least 90% of the intended strength. You want to predict the mean strength of the pills at 53.18 months.

You do not need to re-analyze the stability study model. The worksheet contains the model for the prediction.

1    Open the worksheet SHELFLIFERANDOM2.MTW.

2    Choose Stat > Regression > Stability Study > Predict.

3    In Response, choose Drug%.

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

The stored model is based on batches that you sampled randomly from a larger population. Therefore, you can use the fitted equation for a random batch to predict the mean response (Fit) for any batch from the population. The predicted mean strength for any batch at 53.18 is approximately 100.06 - 0.14 x 53.18 = 92.68.

The confidence interval (CI) indicates that you can be 95% confident that the mean strength is between approximately 91.84 and 93.52. The prediction interval indicates that you can be 95% confident that the strength of any single pill that you test from the population is between approximately 90.15 and 95.22. The prediction interval is always wider than the corresponding confidence interval because of the added uncertainty that is involved in predicting a single response versus the mean response.

This prediction is based on the marginal fitted equation. Minitab also displays the equation and predicted mean for the batch that you select in the dialog box.