SOLAS

SOLAS for Missing Data Analysis Version 4.0 (just released)  is developed with guidance from Prof. Donald B. Rubin, the inventor of mulitple imputation. It provides researchers with a range of imputation techniques in an easy to use, validated software application. SOLAS offers principled approaches to analyzing data with missing values, featuring 9 different imputation techniques and its own script language so that you can easily record your imputation choices.

Multiple Imputation techniques include:

• Mahalanobis Distance Matching Method* ….more
• Predictive Mean Matching Method* …more
• Predictive Model Based Multiple Imputation …more
• Propensity Score Based Multiple Imputation …more (Non Parametric Approach Based on Propensity Scores and the Approximate Bayesian Bootstrap)
• Propensity Score/Predictive Mean Matching/Mahalanobis Distance Combination Method* …more

Single Imputation techniques include:

• Hot Deck Imputation …more
• Predicted Mean Imputation (using Regression)…more (Ordinary Least Squares Method & Discriminant Method)
• Last Value Carried Forward (LVCF / LOCF)…more
• Group Means…more

Other Important Features in SOLAS:

• Unique Missing Data Pattern* …more
• New Pre Imputation Marginplots* …more
• Post Imputation Scatterplots* …more
• Script Language Facility …more
• Post Imputation Analyses …more
• Available in 32- & 64-bit application*

* Denotes New or Improved in SOLAS Version 4.0

The incorrect analysis of datasets with incomplete data can lead to biased analysis and incorrect inference. SOLAS, with its 9 different imputation techniques, unique missing data pattern feature and script language facility, is the missing data software most research statisticians and data analysts choose when working with incomplete data or missing values. Read more about SOLAS for Missing Data Analysis:

• SOLAS in Use – case study (PDF / 136 KB)
• Essay: “Software for Multiple Imputation”, Donald B. Rubin (PDF / 88 KB)