The Mixed-Up Suite

MIXOR, MIXREG, MIXNO and MIXPREG are based on the collaborative effort of Drs. Donald Hedeker and Robert D. Gibbons of the University of Illinois at Chicago. We were privileged to be asked to produce the user interfaces around the computer programs written by Don Hedeker. The work was supported by the National Institute of Mental Health and the MacArthur Foundation, and the programs are available free of charge for download from the MIXOR/MIXREG homepage located at the University of Illinois at Chicago.

While the programs can be run in 'batch mode' using textual input files, the user interfaces guide you through the definition process and provide filter selections based on current settings with bounds and type checking. This allows non-experts to run powerful analyses on their data. The Mixed-up Suite provides mixed-effects regression functionality not available anywhere else — at any price.

Take a look and see if they can help you out.

MIXOR
MIXOR was the first program developed as part of the Mixed-up Suite of applications. MIXOR is a program which provides estimates for a mixed-effects ordinal probit and logistic regression model. This model can be used for analysis of clustered or longitudinal ordinal (and dichotomous) outcome data. For clustered data, the mixed-effects model does not assume that each observation is independent, but does assume data within clusters are dependent to some degree. The degree of this dependency is estimated along with estimates of the usual model parameters, thus adjusting these effects for the dependency resulting from the clustering of the data. Similarly, for longitudinal data, the mixed-effects approach can allow for individual-varying intercepts and slopes across time, and can estimate the degree to which these time-related terms vary.

MIXREG
MIXREG is a program that provides estimates for a mixed-effects regression model (MRM) including autocorrelated errors. This model can be used for analyses of unbalanced longitudinal data, where individuals may be measured at different number of timepoints, or even at different timepoints. Autocorrelated errors of a general form or following an AR(1), MA(1), or ARMA(1,1) form are allowable. This model can also be used for analysis of clustered data, where the mixed-effects model does not assume that each observation is independent, but does assume data within clusters are dependent to some degree. The degree of this dependency is estimated along with with estimates of usual model parameters, thus adjusting these effects for the dependency resulting from the clustering of the data.

MIXNO
MIXNO provides maximum marginal likelihood for mixed-effects nominal logistic regression analysis. These models can be used for analysis of correlated nominal response data, for example, data arising from a clustered design. For clustered data, the mixed-effects model assumes that data within clusters are dependent. The degree of dependency is jointly estimated with the usual model parameters, thus adjusting for dependence resulting from clustering of the data. MIXNO uses marginal maximum likelihood estimation, utilizing a Fisher-scoring solution. For the scoring solution, the Cholesky factor of the random-effects variance-covariance matrix is estimated, along with the effects of model covariates.

MIXPREG
MIXPREG provides maximum marginal likelihood estimates for mixed-effects Poisson regression analysis. These models can be used for analysis of correlated count data, for example, data arising from a clustered design. For clustered data, the mixed-effects model assumes that data within clusters are dependent. The degree of dependency is jointly estimated with the usual model parameters, thus adjusting for dependence resulting from clustering of the data. MIXPREG uses marginal maximum likelihood estimation, utilizing a Newton-Raphson iterative solution. Specifically, the Cholesky factor of the random-effects variance-covariance matrix is estimated along with the effects of model covariates.