The literature on investor’s behavior and performance mainly focused on ordinary individual investors instead of elderly individual investors. This research surveys 196 elderly individual investors, reading the tape at securities trading houses in Taipei City by stratified random sampling. As data collected by questionnaires in this research are ordinal scale data, chi-squared test and ordered logit model are used in exploring the relationship of investor's personal features and his performance. The variables illustrating personal features include “gender”, “education attainment”, “investor's experience”, “number of stocks held”, and “change in the number of stocks held”, “disposition effect”, and “overconfidence”.
The statistical result shows, “education attainment” has significantly positive impact on “investment performance”. However, three variables, i.e., “number of stocks held”, “disposition effect” and “overconfidence” have significantly negative impacts on “investment performance”.
The research suggests, future studies on individual investors may have to separate investors into different groups by age, education attainment or other personal features to get a more precise profile of each group. That is, if an investor’s performance can be approximately expected with the personal features data, then the securities firm can use this information to provide suitable customized advices to help the investor.
