The Behavior of Unemployment and Vacancies in Sweden: An Alternative Test
检验了瑞典1970-1974年月度数据中失业率与职位空缺率的关系，质疑了普通最小二乘法估计的可靠性，为劳动市场研究者提供了替代性检验视角。
where v V/L, u UIL, V is registered job vacancies, L is the labor force, U is survey unemployment, and a, b, and c are unknown parameters to be estimated.' On the basis of the theory set out by Phelps and restated by Holmlund (1975, pp. 65-67), the a priori sign expectations for b and c in models (1) and (2) are b >0, c <0 and b, c<0 respectively. To test these implications empirically, Holmlund estimated these models with monthly data from Sweden over the period 1970-1974,2 using ordinary least-squares techniques (OLS). His results, which are reproduced in Table 1 for convenience, conformed to the alleged predictions of Phelps' model, thus providing some apparent support for the theoretical framework utilized. Upon further scrutiny, however, the above regression specifications and the resulting empirical estimates are troublesome in several respects. First, the use of OLS to estimate the parameters of equations (1) or (2) is of questionable