The CUSUM Procedure |

Syntax |

The syntax for the INSET statement is as follows:

- INSET
*keyword-list*</*options*> ;

You can use any number of INSET statements in the CUSUM procedure. Each INSET statement produces a separate inset and must follow an XCHART statement. The inset appears on every panel (page) produced by the last XCHART statement preceding it.

Keywords specify the statistics to be displayed in an inset; options control the inset’s location and appearance. A complete description of the INSET statement syntax is given in the section Syntax of Chapter 13, The SHEWHART Procedure. The INSET statement options are identical in the CUSUM and SHEWHART procedures, but the available keywords are different. The options are listed in Table 13.87. The keywords available with the CUSUM procedure are listed in Table 6.8 to Table 6.11.

Keyword |
Description |
---|---|

ARL0 |
average run length for zero shift |

ARLDELTA |
average run length for shift of |

DATA= |
arbitrary values from |

N |
nominal subgroup size |

NMIN |
minimum subgroup size |

NMAX |
maximum subgroup size |

Keyword |
Description |
---|---|

DELTA |
shift to be detected as multiple of standard error |

H |
decision interval as a multiple of standard error |

HEADSTART |
headstart value as a multiple of standard error |

K |
reference value |

MU0 |
target mean |

SHIFT |
shift to be detected in data units |

STDDEV |
estimated or specified process standard deviation |

Keyword |
Description |
---|---|

ALPHA |
probability of Type 1 error |

BETA |
probability of Type 2 error |

H |
vertical distance between V-mask origin and upper (or lower) arm |

K |
slope of lower arm of V-mask |

SIGMAS |
probability of Type 1 error as probability that standard normally distributed variable exceeds a specified value in absolute value |

You can use the keywords in Table 6.11 only when producing ODS Graphics output. The labels for the statistics use Greek letters.

Keyword |
Description |
---|---|

UALPHA |
probability of Type 1 error |

UARLDELTA |
average run length for shift of |

UBETA |
probability of Type 2 error |

UDELTA |
shift to be detected as multiple of standard error |

UMU0 |
target mean |

USIGMA |
estimated or specified process standard deviation |

Copyright © SAS Institute, Inc. All Rights Reserved.