Select the option that best describes how you want to enter the data to predict the mean response at a specified time. If the batch variable is included in the final model, you can also predict responses for specific batches.

Complete the following steps if you want to enter variable settings directly in the dialog.

- From Response, select a response variable to predict.
- In the second drop-down list, select Enter individual values.
- In the table, enter at least one value for each variable. For the time variable, match the data type from the model. For example, if the model is based on number of months, then enter number of months. If you have a batch variable, select a value from the drop-down list.

Complete the following steps if you want to enter columns of data.

- From Response, select a response variable to predict.
- In the second drop-down list, select Enter columns of values.
- In the table, enter one column for each variable. Each column must contain the values for one variable. All columns must have the same number of rows. The columns must match the data types from the model. For example, if the time variable is the number of months, then enter a numeric column with number of months. If the batch variable is the text values of "1" and "2," then enter a text column with the values "1" and "2". Minitab stores the results in the worksheet columns and does not display them. To display the results, check the items in the Results sub-dialog box.

In this worksheet, C1 – C4 contain the data that Minitab used to estimate the stability study model. Month_New contains the new observations for the month variable to predict. Batch_New contains the batch corresponding to each month.

C1 | C2 | C3 | C4 | C5 | C6 |
---|---|---|---|---|---|

RunOrder | Month | Batch | Drug% | Month_New | Batch_New |

1 | 0 | 4 | 99.270 | 12 | 2 |

2 | 0 | 3 | 100.026 | 18 | 4 |

3 | 0 | 5 | 100.403 | 24 | 1 |

4 | 0 | 2 | 100.108 | 30 | 3 |

5 | 0 | 1 | 99.903 | 36 | 5 |