| Summary: | 碩士 === 國防管理學院 === 資源管理研究所 === 90 === Abstract
Most discussions and researches found on turnover focused primarily on the reasoning and causes for the resignation rather than a systematic ratio in determining the rating and the numbers of the turnover through out the studies. However, turnover rate is one of the key issues worth probing into as part of the human resources management. Appropriate number of turnover in any organization is necessary through out the growing processes; on the other hand, any inappropriate or unplanned turnover will not only have negative effect on the growth of any organization, but also brings down the quality of the workforce which inevitably disrupts the structure of the organization. Therefore, having to focus studies on the rate of turnover is worth instigating, using the statistics incurred as one of the source in the human resources administrative policy.
This research is based on and with the comptroller and financial officer in the military as main factor. Using both the Logistic Regression Model and Back-Propagation neural Network as guidelines in building this study. The studies confirmed that based upon age, salary, seniority, rank, education, and gender played a very relevant degree of turnover rate within an organization. Through the studies, it is also found that by using the above said methods, the results found are as close to becoming fact and high degree of acceptance.
Keyword：turnover、turnover rate、forecast、logistic regression
model、back-propagation neural network
|
|---|
