The prediction of minimum temperatures at Alice Springs aerodrome, using data for 1800 CST the previous night, is examined by two methods. Simple linear regression on air temperature yielded useful results on development data, but was unstable on independent data. No improvement was associated with the inclusion of additional parameters. The second method, based on radiative flux at 1800 CST, gave approximately the same accuracy as the regression technique but was more stable for independent data. It has the added advantage of being more easily adapted to cloudy nights.
E. A. Mizon (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: