On the limitations of comparing mean square forecast errors

Linear models are invariant under non-singular, scale-preserving linear transformations, whereas mean square forecast errors (MSFEs) are not. Different rankings may result across models or methods from choosing alternative yet isomorphic representations of a process. One approach can dominate others for comparisons in levels, yet lose to another for differences, to a second for cointegrating vectors and to a third for combinations of variables. The potential for switches in ranking is related to criticisms of the inadequacy of MSFE against encompassing criteria, which are invariant under linear transforms and entail MSFE dominance. An invariant evaluation criterion which avoids misleading outcomes is examined in a Monte Carlo study of forecasting methods.     

Timed bromocriptine administration reduces body fat stores in obese subjects and hyperglycemia in type II diabetics.

Obese postmenopausal female volunteers were given timed daily oral dosages of bromocriptine, and tested for reduction of body fat stores. This dopamine agonist has been shown to reset circadian rhythms that are altered in obese animals and to reduce body fat levels in several animal models. The participants were instructed not to alter their existing exercise and eating behavior during treatment. Skinfold measurements were taken on 33 subjects as indices of body fat. The measurements (e.g., suprailiac) were reduced after six weeks by about 25%, which represents a reduction of 11.7% of the total body fat. These dramatic decreases in body fat, which are equivalent to that produced by severe caloric restriction, were accompanied by more modest reductions of body weight (2.5%), indicating a possible conservation of protein that is usually lost as a consequence of such caloric restriction. The effects of bromocriptine treatment on body fat and hyperglycemia were also examined in non-insulin dependent diabetics being treated with oral hypoglycemics (7 subjects) or insulin (7 subjects). Total body fat was reduced by 10.7% and 5.1% in diabetics on oral hypoglycemics and insulin, respectively, without any significant reductions in body weight. Hyperglycemia was reduced in most of the 15 diabetic subjects treated leading to euglycemia and even cessation of hypoglycemic drugs in 3 of the 7 subjects during 4-8 weeks of bromocriptine treatment. These findings support the hypothesis that obesity and type II diabetes may be treated effectively with bromocriptine when administered at the proper times and dosages.