Shrinkage Pre-Test Estimator of the Intercept Parameter for a Regression Model with Multivariate Student-t Errors

In the presence of an uncertain prior information about the value of the slope parameter, the estimation of the intercept parameter of a simple regression model with a multivariate Student-t error distribution is investigated. The unrestricted, restricted and shrinkage preliminary test maximum likelihood estimators are defined. The expressions for the bias and the mean square error of the three estimators are provided and the relative efficiences are analyzed. A maximin criterion is established, and graphs are constructed for an arbitrary number of degrees of freedom (D.F.) as well as sample sizes. A criterion to select optimal significance level is also discussed.     

Charge heterogeneity of cholera toxin and its subunits

Abstract Analytical isoelectric focusing (IEF) in thin layers of polyacrylamide gels resolved cholera toxin into 3 isomeric forms differing in charge (isoelectric points 6.80, 6.65 and 6.55). All these forms had identical molecular weights, and were also antigenically similar, as demonstrated by their reactivity to antisera to cholera toxin. Both the B and A subunits possessed charge heterogeneity. The B subunit was detected in a free form when a solution of cholera toxin was aged for a few days. Antisera to cholera toxin, irrespective of mode of immunisation, contained antibodies to both the intact cholera toxin and the free B subunit as demonstrated by the immunoblotting technique based on IEF.