Employees are the most valuable assets of an organization. Their significance to organizations calls for not only the need to attract the best talents but also the necessity to retain them for a long term. This paper focuses on reviewing the findings of previous studies conducted by various researchers with the aim to identify determinants factors of employee retention. This research closely looked at the following broad factors: development opportunities, compensation, work-life balance, management/leadership, work environment, social support, autonomy, training and development. The study reached the conclusion that further investigations need to be conducted regarding employee retention to better comprehend this complex field of human resource management.

No abstract

This paper investigates if bankruptcy of Japanese listed companies can be predicted using data from 1992 to 2005. We find that the traditional measures, such as Altman's (J Finance 23:589-609, 1968) Z-score, Ohlson's (J Accounting Res 18:109-131, 1980) O-score and the option pricing theory-based distance-todefault, previously developed for the U.S. market, are also individually useful for the Japanese market. Moreover, the predictive power is substantially enhanced when these measures are combined. Based on the unique Japanese institutional features of main banks and business groups (known as Keiretsu), we construct a new measure that incorporates bank dependence and Keiretsu dependence. The new measure further improves the ability to predict bankruptcy of Japanese listed companies.

In this paper, we study normal homogeneous Finsler spaces. We first define the notion of a normal homogeneous Finsler space, using the method of isometric submersion of Finsler metrics. Then we study the geometric properties. In particular, we establish a technique to reduce the classification of normal homogeneous Finsler spaces of positive flag curvature to an algebraic problem. The main result of this paper is a classification of positively curved normal homogeneous Finsler spaces. It turns out that a coset space G/H admits a positively curved normal homogeneous Finsler metric if and only if it admits a positively curved normal homogeneous Riemannian metric. We will also give a complete description of the coset spaces admitting non-Riemannian positively curved normal homogeneous Finsler spaces.Mathematics Subject Classification (2000): 22E46, 53C30. Key words: Normal homogeneous spaces; isometric submersion; flag curvature. RésuméDans cet article, nousétudions les espaces de Finsler homogènes normales. Nous définissons d'abord la notion d'un espace homogène Finsler normale, en utilisant la mthode de l'immersion isométrique de métriques de Finsler. Ensuite, nousétudions les propriétés géométriques. En particulier, nousétablissons une technique pour réduire la classification des espaces de Finsler homognes normales de courbure du pavillon positifà un problme algébrique. Le résultat principal de cet article est une classification des espacesà courbure positive de Finsler homognes normales. Il se trouve que d'un espace de coset G/H admet une métrique homogène normale courbure positive Finsler si et seulement si il admet une courbure positive normale homogène métrique riemannienne. Nous allons aussi donner une description complète des espaces de coset admettant non-riemanniennes espacesà courbure positive de Finsler homogènes normales.

In this paper, we study Clifford-Wolf translations of Finsler spaces. We give a characterization of those Clifford-Wolf translations generated by Killing vector fields. In particular, we show that there is a natural interrelation between the local one-parameter groups of Clifford-Wolf translations and the Killing vector fields of constant length. In the special case of homogeneous Randers spaces, we give some explicit sufficient and necessary conditions for a Killing vector field to have a constant length, in which case the local one-parameter group of isometries generated by the Killing field consist of Clifford-Wolf translations. Finally, we construct explicit examples to explain some of the results of this paper.

In this paper, we introduce new types of Finsler metrics, called (α 1 , α 2 )-metrics. We define the notion of the good datum of a homogeneous (α 1 , α 2 )-metric and use that to study the geometric properties. In particular, we give a formula for the S-curvature and deduce a condition for the S-curvature to be vanishing identically. Moreover, we consider the restrictive Clifford-Wolf homogeneity of left invariant (α 1 , α 2 )-metrics on compact connected simple Lie groups. We prove that, in some special cases, a restrictively Clifford-Wolf homogeneous (α 1 , α 2 )-metric must be Riemannian. An unexpected interesting observation contained in the proof reveals the fact that S-curvature may play an important role in the study of Clifford-Wolf homogeneity in Finsler geometry. on spheres have been classified recently by the authors in [27]. Therefore our next step is to classify left invariant CW-homogeneous Finsler metrics on compact Lie groups. In our previous works, we have performed this program for Randers metrics and (α, β)-metrics (see [12,25]). Therefore we study CW-homogeneity of left invariant (α 1 , α 2 )-metrics on compact Lie groups in this paper. However, the general case seems to be very involved, so we will confine ourselves to the case where in the decomposition g = V 1 + V 2 of T G e = g, the subspace V 2 is a commutative subalgebra of g. In particular, we will discuss the following two cases:Case 1 G is a compact connected simple Lie group, and V 2 is a Cartan subalgebra. Case 2 G is a compact connected simple Lie group, and V 2 is 2-dimensional commutative subalgebra.In the study of the restrictive CW-homogeneity of left invariant non-Riemannian (α 1 , α 2 )-metrics in Case 1, the S-curvature plays an important role. The main results are the following two theorems.The study of the restrictive CW-homogeneity of left invariant non-Riemannian (α 1 , α 2 )-metrics of Case 2 is a generalization of our work on (α, β)-metrics [25]. The systematic technique we have developed in the study of Killing vector fields of constant length of left invariant Randers and (α, β)-metrics also works in this case. (α 1 , α 2 )-metric on a compact connected simple Lie group G with a decomposition g = V 1 + V 2 such that V 2 is a 2-dimensional commutative subalgebra. If F is restrictively CW-homogeneous, then it must be Riemannian. Theorem 1.2 Let F be a left invariantTheorems 1.1 and 1.2 suggest the following conjecture: Conjectute 1.3 Let F be a left invariant (α 1 , α 2 )-metric on a compact connected simple Lie group G with a decomposition g = V 1 + V 2 such that V 2 is a commutative subalgebra. If F is restrictively CW-homogeneous, then it must be Riemannian.More generally, we can make a stronger one: (α 1 , α 2 )-metric on a compact connected simple Lie group G with a dimension decomposition (n 1 , n 2 ), where n 1 ≥ n 2 > 1. If F is restrictively CW-homogeneous, then it must be Riemannian. Conjectute 1.4 Let F be a left invariantThe paper is organized as follows. In Sect. 2, we recall some definitions and known res...

In this paper, I study the isoparametric hypersurfaces in a Randers sphere (S n , F ) of constant flag curvature, with the navigation datum (h, W ). I prove that an isoparametric hypersurface M for the standard round sphere (S n , h) which is tangent to W remains isoparametric for (S n , F ) after the navigation process. This observation provides a special class of isoparametric hypersurfaces in (S n , F ), which can be equivalently described as the regular level sets of isoparametric functions f satisfying −f is transnormal. I provide a classification for these special isoparametric hypersurfaces M , together with their ambient metric F on S n , except the case that M is of the OT-FKM type with the multiplicities (m 1 , m 2 ) = (8, 7). I also give a complete classificatoin for all homogeneous hypersurfaces in (S n , F ). They all belong to these special isoparametric hypersurfaces. Because of the extra W , the number of distinct principal curvature can only be 1,2 or 4, i.e. there are less homogeneous hypersurfaces for (S n , F ) than those for (S n , h).

No abstract