No database exists suitable for disclosing the effect on economic activities caused by dementia. Thus, our aim is to obtain the principles to resolve its influence by analyzing related databases. In order to detect the capability deterioration of economic activity for elderly people at the age of sixty-five years or older, we have employed an anonymous data set obtained from the National Survey of Family Income and Expenditure carried out by Ministry of Internal Affairs and Communications. So far, a method has been developed to detect dissaving risk of elderly people using clustering. Nevertheless, the initial approach showed limitations for further improvement. Therefore, we have proposed another method to detect dissaving risk by dividing analysis data into test/training data. The test data are then processed through primary/secondary determination in accordance with the amount of savings. In secondary determination, three methods are performed in terms of income and savings. In the previous analysis, as an initial step, merely the data of single females were analyzed. In this paper, the data of single males and two-person households are analyzed. As a result of the analysis, the correct judgment rate exceeds 80% for single males, while it attains as high as approximately 70% for two-person households.