Use this to set various options related to calibrating survey weights.

## Display

Specifies the output to be displayed.

Summary |
Summary of the calibration. |

Totals |
Displays estimates of the totals. |

Monitoring |
Displays progress of the iterative process. |

## Convergence

Controls parameters for the iterative calculations used for the **Truncated linear** and **Logistic** methods.

Maximum number of iterations |
Maximum number of iterations used in iterative calculations. |

Tolerance |
Tolerance for convergence of iterative calculations. |

## Method

Specifies the method used for the calibration. **Linear** calculates estimates equivalent to the usual regression estimates. The g-weights are not restricted and may be negative. **Truncated linear** restricts the g-weights within the range specified by the **Bounds for g-weights** settings by replacing extreme values by these bounds. **Logistic** uses a logistic transformation to ensure that the weights remain within the specified bounds. Fitted values can be produced without any calibration by using the **Fitted values** setting. This setting avoids having to repeat the full calibration process when analysing additional Y variates.

## Bounds for g-weights

Specifies the bounds for the g-weights when using either the **Truncated linear** or **Logistic** methods. The lower bound can be specified using the **Lower** field and similarly the upper bound can be specified using the **Upper** field. The default settings are 0.1 for the lower bound and 10 for upper bound, thus allowing the adjusted weights to differ from the initial weights be a factor of 10 in each direction.

## Graphics

Specifies what type of plots can be produced with the analysis. Select **Weights plot** to produce a plot of the weights.

## See also

- Survey Calibration Weighting menu
- SVCALIBRATE procedure