Abstract: The assessment of probability gains K of earthquake prediction methods with different time and space scales is one of basic studies on synthetical probability gain model of earthquake prediction. The earthquake prediction methods with different time and space scales can be divided into tree types,i.e. time gain, space gain and time-space gain. In this paper, the assessment methods of probability gains are studied, the relations between prediction state probability and probability gain and between prediction R value and probability gain are given. On the basis of synthetical probability gain model, it is possible to make the better medium- and long term earthquake predictions at present time.

Key words: Earthquake prediction, Prediction effect evaluation, Probability gain model