國內博達科技無預警聲請重整，對會計師業及資本市場均造成嚴重之衝擊，有些投資人甚至質疑財務報表審計存在的必要性及其價值。財務報表審計有無價值主宰會計師業是否有存在之必要，其影響不容小覷。本研究乃依據Wallace(1980，1987)提出之代理、資訊及保險等三假說，藉會計師審計失敗之事件，觀察其他簽證客戶股價之反應，以衡量財務報表審計是否有價值。 本研究以企業及審計均失敗的公司為實驗組，企業及審計均未失敗的公司作為控制組，期衡量財務報表審計是否有價值。依據研究個案篩選標準，本研究針對87至92年間，各選取五家公司作為實驗及控制組之研究個案。 實證結果顯示，在五個研究個案中，中友的簽證會計師事務其他簽證客戶之股價在事件期間，有顯著為負的異常報酬，亦即，中友之簽證會計師提供之財務報表審計確有價值；然國揚、和旺、桂宏及茂矽等四家公司的簽證會計師事務所審計失敗時，因市場出現其他影響股價變動之事件，故無法衡量此四家公司之審計是否有價值。 若僅自中友之案例，類推所有事務所之財務報表審計均有價值，恐不足以採信，故本研究推論，會計師事務所提供的財務報表審計服務可能有價值，惟其是否確有價值，則無法自實證結果提出佐證。 / PROCOMP INFORMATICS LTD affects accountant industry and the capital market seriously. Some investors even question both the necessity of existence and the value of financial statement audit. Whether financial statement audit has value decides the necessity of accountant industry’s existence and its influence is significant . This research is based on agency, information and insurance hypothesis proposed by Wallace (1980, 1987). In this paper, we choose the event of audit failure and observe the stock price of CPA firm other audit clients to measure the value of financial statement audit. This research designs both the experimental group and the control group to measure the value of financial statement audit. We choose companies which occurred business failure and audit failure as the experimental group. In stead, we choose companies which didn’t occur business failure and audit failure as the control group. Based on the research screening standard, we are aimed at five companies respectively taken for the experimental and the control group. Results show that the stock price of other audit clients of Chungyo’s CPA firm will have significant negative abcdrmal return when CPA firm occurs audit failure. Namely, financial statement audit provided by Chungyo’s CPA firm has value. However, when the other company’s CPA firm occur audit failure, some other event that affect stock price to vibrate happen simultaneously, we can’t measure the value of financial statement audit provided by the four company’s CPA firms. On the whole, if we use the finding to analogize that financial statement audit provided by all CPA firms has value , it maybe not to be persuasive. Consequently, this research infers that the financial statement audit services provided by CPA firms may has value. Whereas, we can’t conclude that financial statement audit has value truly.

Cette thèse se donne pour objectif de contribuer à la compréhension des tendances démographiques en Afrique subsaharienne par l’examen de l’influence de la mortalité infantile sur les comportements de fécondité. Dans le premier chapitre, j’examine la relation entre mortalité infantile et fécondité à l’échelle micro-économique. Des modèles de données de comptage sont utilisés pour analyser les déterminants de la fécondité complète de femmes d’une communauté rurale sénégalaise. Je montre que l’effet global de la mortalité infantile est positif tant sur la fécondité totale que sur la fécondité nette. De plus, j’identifie une relation en U inversé entre mortalité infantile et fécondité nette. Dans le second chapitre, j’analyse les effets de la mortalité infantile sur les comportements reproductifs dans un modèle de fécondité endogène où la survie infantile est stochastique. J’adopte une forme fonctionnelle de coût des enfants englobant quatre scénarios différents, chacun représentant un contexte socio-économique distinct. Mon modèle peut prédire des réponses positives et négatives de la fécondité, selon que les enfants sont respectivement “intensifs en temps” ou “pourvoyeurs en temps”. Finalement, le troisième chapitre analyse les effets de la mortalité et de la morbidité infantiles sur le processus de décision de fécondité des femmes rurales sénégalaises. J’estime des modèles dynamiques non linéaires de données de panel. Je montre que la mortalité et la morbidité palustres à l’échelle de la communauté, ont un effet positif sur les décisions ultérieures de fécondité. Cet effet est d’autant plus fort que la maladie est létale pour les enfants infectés. / This dissertation attempts to contribute to the understanding of current demographic trends in sub-Saharan Africa by examining the role of child mortality in shaping fertility behavior. In the first chapter of this dissertation, I examine the relationship between child mortality and fertility at the micro level. Count data models are employed to investigate the determinants of completed fertility of women from a Senegalese rural community. The global effect of child mortality on total and net fertility is found to be positive. I also identify an inverted-U shaped relationship between child mortality and net fertility. In the second chapter of this dissertation, I analyze the effects of child mortality changes on fertility behaviors in an endogenous fertility model where child survival is stochastic. I adopt a functional form for the cost of children that allows for four different scenarios, each of which is representative of a particular socio-economic setting. My model can predict both positive and negative fertility responses to child mortality depending on whether children are “time-intensive” or “time-supplying”, respectively. Finally, the third chapter analyzes the effects of childhood mortality and morbidity on the fertility decision-making process among rural Senegalese women. I estimate nonlinear dynamic panel data models of fertility behavior. I find that community child mortality and morbidity attributable to malaria exert a joint influence on fertility behaviors. Community-level malaria incidence among children has a positive effect on subsequent fertility choices, and this positive effect is stronger the more the disease is fatal to children who are infected.