Further Assessment of Dissaving Risk against Life Expectancy for Elderly People through Anonymous Data and Random Data for Social Implementation

With a view to detecting the deterioration of economic activity for elderly people at age of sixty-five or over, we have used an anonymous data (AD) obtained from the National Survey of Family Income and Expenditure (NSFIE) carried out by Ministry of Internal Affairs and Communications (MIC). So far we have developed a method to detect dissaving risk of elderly people by using AD and random data (RD). Here, AD was used as test data, while RD was set as training data. Then three methods were performed in terms of savings and income. The analysis result has shown that using both RD and AD would be as effective as using only AD for evaluating the performance of the proposed method. However, with a view to social implementation, it is required to investigate if our methods could be applicable when house types, rent or own house, are disregarded. This paper firstly shows that our method can be extensive with disregarding house types. Then these results can be improved for single males and females through altering two experimental parameters. For the purposes of performance evaluation, four sets of conditions are considered and analyzed. The best when the threshold to determine AD as dissaving risk might depend on the training data. Moreover, the best criterion to detect dissaving risk of elderly males was not the same for elderly females.