Return to Stat Tools Home
# Title: Logistic Regression

## Evaluate this tool by clicking on the stars:

Score: $score, My vote: $mylast, Total votes: $sum |" }%## Add Comments and then click on "Add comment" button

#### Please note any errors found in the macros.

%COMMENT%
### StatToolsForm edit

This topic: CTSpedia > WebHome > StatToolsHome > StatToolsTopic003

Topic revision: 01 Jul 2012, PeterBacchetti

- 5 stars = highest rating
- 1 star = lowest
- SCORE = average of all votes

Score: $score, My vote: $mylast, Total votes: $sum |" }%

Type of Tool | SAS Macro |

Title | Logistc Regression |

Programmer/Email | PeterBacchetti |

Contributing Site | UCSF |

Description | This is a macro for performing logistic regression with a binary outcome, checking model assumptions, and producing output that is arranged in a way suitable for use in publications. Some features that add convenience include automatically creating quartile categories for specified predictors, rescaling numeric predictors to make odds ratios more interpretable, and showing counts for categorical predictors. |

Classification Graph Type | |

References | |

Datasets | |

Data | |

Software Program | SAS |

Software | SAS |

Macro Parameters | Parameters for Logistic Macro: DSName = name of SAS data set Outcome = name of outcome variable. This variable is assumed to be coded 0, 1 and the model predicts the probability of a 1 outcome. Predictors = list of predictor variables. This list should include class variables and variables to be 'quartiled.' Interaction variables may be included in which case the predictor list should be quoted, e.g. Predictors = %quote(Age Gender Age*Gender). Note: all variables appearing in interaction terms should also appear as main effects. Classvar = classification variables with reference level. Use %quote function. For example, Classvar=%quote(Gender(ref='Male')Race(ref='Asian')) Gender and Race could be numeric variables with formatted values of Male or Asian or they could be alphanumeric variables with string values of 'Male' and 'Asian'. Do not include variables for which quartiles are to be generated. Quartiles = list of numeric variables which should be entered into the model as class variables with four quartile levels. Variables in this list will not be included in the model as linear predictors. Quad = Y/N for creating and testing quadratic terms (default = Y) Out = name of output SAS data set containing logistic estimates, p-values, etc. Print = Y/N (default = N) Include = F,P,FP,or N F = frequencies of outcome variable are displayed for each class variable value, e.g., Sex Female 31/100 P = percentages are displayed with class variable values, e.g., Sex Female 31.0% FP = frequencies and percentages are displayed with class variable values, e.g., Sex Female 31/100 (31.0%). N = no frequencies or percentages are displayed with class variable values (default = FP) Obs = Y/N If Y, then the number of observations in each outcome group and the total are displayed. (default = Y) LL = Y/N If Y, then -2log(likelihood) is displayed. (default = N) HL = Y/N If Y, then the Hosmer-Lemeshow p-value is displayed along with the number of groups of estimates (default = N). |

R-Code - Attachment | |

R-Code | |

SAS-Code - Attachment | UCSF_Logistic Regression-Main Macro |

SAS-Code | |

Stata-Code - Attachment | |

Stata-Code | |

Other Code - Attachment | |

Other Code | |

Called Data Manipulation | |

Called Tool/Utility | |

Called Checking Macro | |

Called Other | |

Creation-Date | |

Revision-Date | |

Example Code | SAS Example Code |

Example Output | SAS Example Output |

SAS Examples | Macro to make hyperlinked output Logistic Regression Example with hyperlinked output |

SAS_Example_Code | |

SAS_Example_Output | Output file created by above example |

R Examples | |

R_Example_Code | |

R_Example_Output | |

Stats Examples | |

Special Features | Printing |

Special Features Attached | Ordinary printing macro |

Special Features Text | Special Features of Logistic macro: The macro will automatically test a quadratic term for numeric (non-class) predictors with more than two values. The macro will automatically output odds ratios that have been rescaled until the units > 1/10 of the IQR. |

Notes1-Legend | Section with calls to needed macros |

Notes1 | |

Notes2-Legend | Macro call with fully specified parameters |

Notes2 | |

Notes3-Legend | Section with definitions of global macro parameters |

Notes3 | |

Notes4-Legend | Section with definitions of local macro variables |

Notes4 | |

See Also | Research Topic: Logistic Regression |

Checklists | Logistic Regression Checklists |

Stat Tools Disclaimer | |

Discliamer | The views expressed within CTSpedia are those of the author and must not be taken to represent policy or guidance on the behalf of any organization or institution with which the author is affiliated. |

Permission | Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF ERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT OLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. |

Reference Image |

Topic revision: 01 Jul 2012, PeterBacchetti

Copyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.

Ideas, requests, problems regarding Foswiki? Send feedback

Ideas, requests, problems regarding Foswiki? Send feedback