Parameter estimation and its application using non-statistical theory

In many social science and natural science research, researchers are often faced with the problems such as: the system have no enough system distinctive features, the probability distribution unknown and the number of data very small. It is hard to use statistical theory in these researches. A novel method called non-statistical method is proposed in this paper to obtain more characteristics and further information in the system. The method permits the probability distribution unknown and the number of data very small, because of some deficiency of classical statistics in system and information science. In virtue of the special strategy of data processing, more characteristics and further information in the system could be obtained using this method. Some basic concepts, characteristics and elements of non-statistical method including point estimation, interval estimation, optimum level, practicable interval and systemic characteristic mapping etc, are also recommended. The validity of this method is examined by some measurement cases and practical engineering examples.