The robustness of point optimal testing for rosenberg random regression coefficients
研究了罗森伯格随机回归系数模型中点最优检验对两种偏离情况的稳健性，发现该方法对希尔德雷思-霍克备择假设和回归扰动非正态性都相当稳健。
The literature on testing for the presence of Rosenberg's (1973) return to normalcy random coefficient model is well developed with both Shively (1988) and Brooks (1993) advocating the use of point optimal tests. This paper explores the robustness of point optimal testing for the Rosenberg alternative to two departures: the special case HildrethHouck (1968) alternative and non-normality in regression disturbances, finding the point optimal testing approach to be fairly robust to both departures.