[jira] [Updated] (MADLIB-987) Mixed Effects Modeling

["Frank McQuillan (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MADLIB-987?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Frank McQuillan updated MADLIB-987:
-----------------------------------
    Attachment: MixedModels_SRS_v01_AC.pdf

> Mixed Effects Modeling
> ----------------------
>
>                 Key: MADLIB-987
>                 URL: https://issues.apache.org/jira/browse/MADLIB-987
>             Project: Apache MADlib
>          Issue Type: New Feature
>            Reporter: Frank McQuillan
>         Attachments: MixedModels_SRS_v01_AC.pdf
>
>
> Description
> A mixed effects model, like many other statistical models, infers the relationship between dependent (or response ) and independent variables. It generalizes the standard regression model by allowing correlation among observations, which may be due to grouping of subjects (e.g. students in different schools), or to repeated measurements made sequentially on the
> same subject (e.g. longitudinal data in biostatistics or panel data in econometrics). A mixed effects model contains fixed effects and random effects components.
> See reference [1] for a requirements doc for mixed effects modeling tailored to MADlib.
> References
> [1] Mixed effects modeling reqts doc
> (attached)



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)