由於開放總體經濟理論，在在著許多爭論，往往因其所強調重點不同而有不同的結論 。這些爭論最後須藉助實證之證據來支持某種理論，以了解經濟變數間可能的影響途 徑。 然由迴歸所觀察到兩變數間有強烈相關，並不能代表此二變數間存在有因果關係。故 本文乃擬利用Granger 因果關係的概念與時間序列之統計方法，在儘可能避免作太多 先驗理論之限制下，使觀察值之統計資料能充分表現其一般之情況，來探討台灣重要 經濟變數間可能的影響途徑與因果關係。 本文共分五章： 第一章「緒論」：就有關國際金融理論作一概要性探討。 第二章「因果關係檢定」：介紹因果關係之概念、基本假設檢定方法與檢定結果之解 釋。 第三章「模型選擇之方法」：介紹ＶＡＲ模型的區塊排除性檢定，客觀貝氏ＶＡＲ模 型，Hsiao 之ＶＡＲ模型認定，與多元時間序列ＶＡＲＭＡ模型。 第四章「台灣之實證分析」：乃就第三章所論之模型，以台灣有關重要變數作實證上 之分析。 第五章「結論與建議」：對全文作綜合性總結與建議。

《金瓶梅》為中國四大奇書之一，廣泛而深入地描寫人們對財色的欲求，透過其中所程現之人情事理的普遍性，可以體會人生的道理。如今，《金瓶梅》一書普行於世，已有多種版本而張竹坡評本《金瓶梅》對於中國內外流傳貢獻最大。本文旨在了解張竹坡利用評點形式分析《金瓶梅》時，是如何認識「小說」和「人生」的，試圖再詮釋張竹坡的小說觀和人生觀，希望藉由研究張批《金瓶梅》，得以更了解「小說」之本質，並且從而體會人生的道理。全文共分六章。第壹章緒論，概述研究旨趣、文獻檢討及使用版本。第貳章則從小說評點的發達情況、《金瓶梅》問世以後世人之評述與清初文字獄及禁燬小說等三方面，來探討張批小說理論之形成背景。第參章概述張竹坡生平及其評點《金瓶梅》之版本問題。第肆章探討張竹坡論小說之本質，創作及閱讀態度。第伍章探討張竹坡有關小說的人物，小說的情節結構，小說的主題，小說的語言，小說的細節，小說的環境構設等等的小說理論。第陸章結語，綜論張批小說理論之價值。

Field and laboratory studies were used to elucidate the factors affecting temporal and spatial variations of species abundance and structure of macroalgal assemblage and environmental variables between fish farming (FFA) and non-fish farming (NFFA) areas in Hsiao-Lu-Chiao island, a coral island in southwestern Taiwan. Four experiments have been approached: 1. field surveys of macroalgal assemblage structure on 5-m and 10-m depth at 3 sampling sites at FFA (FFA1, FFA2 and FFA3) and 1 sampling site at NFFA from September 2004, January 2005 and April 2005; 2.the relationship between abiotic (monthly maximum air temperature, monthly minimum air temperature, monthly mean air temperature, monthly cumulative precipitation, monthly cumulative irradiance, seawater temperature, light extinction coefficient, water motion, and nutrient (NO3-, NO2-, NH4+, SRP, DON, and DOP) and biotic (seaurchin density) factors and spatio-temporal variations in macroalgal structures analyzed by non-parametric multivariate model; 3. Factors affecting macroalgal abundance and structure: (1). Comparison of growth temperature ranges in different species to field temperature fluctuation; (2).Comparison of growth irradiance ranges in different species to field irradiance fluctuation; 4.Herbivore pressure: (1). Spatio-temporal variations of sea urchin abundance and structure of assemblage; (2). Gut contents and food preference of sea urchin experiment; (3). Herbivore exclusion experiment. Macroalgal %cover, biomass, species richness, diversity (H¡¦) and evenness (J¡¦) showed temporal and spatial variations, low values in January 2005 and also low values in the 5 m- and 10 m-depth areas of FFA1 and the 10 m-depth areas of FFA2. The data of k-dominance curve, hierarchical cluster and ANOSIM tests indicate that macroalgal assemblage is different between 4 sampling sites, between 2 depths and between 3 seasons. Ceratodictyon spongiosum is the most important species that separates September and January assemblages from April assemblage and separates the FFA1 and FFA2 assemblages from the FFA3 and NFFA assemblages. BVSTEP analysis shows that nutrients (NO3-, DON, DOP), temperature, monthly cumulative precipitation, and sea urchin density are the factors corresponding to variations of macroalgal assemblages, this correlation is more significant for 5 m-depth assemblage. Fish farming area FFA1and FFA2 assemblage are affect by sea urchin density, temperature and DON. Sea urchin influnces macroalgal abundance and assemblage structure in FFA1 and FFA2. Macroalgal %cover in 5 m-depth area shows a reversal relationship with sea urchin density; however, this relationship is not observed for 10 m-depth area. FFA1 and FFA2 are belong to high grazing pressure sites as indicated by high sea urchin density and exclusion experiment. Sea urchin gut contents and feeding preference test show that sea urchin has strong food selectivity with Hypnea charoides and Gracilaria coronopifolia as the most preferred species. Herbivore exclusion experiment shows that Hypnea charoides and Gracilaria coronopifolia are the species recruited in the cages. Ceratodictyon spongiosum had high biomass in FFA1 and FFA2 in January, which was ¡¥low DOP/high DON¡¦. The coindicence of temporal variations in FFA3 assemblage structure with a change from ¡¦Halimeda opuntia and Boodlea compostia¡¦ ¡÷ ¡¦Amphiroa fragilissima, Corallina phhulifera and Galaxaura oblongata¡¦ ¡÷¡¦Halimeda opuntia and Boodlea compostia¡¦ with low nitrogen/ high phosphorous¡¦ ¡÷¡¦ high nitrogen/ low phosphorous¡¦ ¡÷ ¡¦low nitrogen/ high phosphorous¡¦ suggest a role of ¡¦low nitrogen (NO3-)/high phosphorous (DOP)¡¦ for FFA3 structure modification. NFFA assemblage is controlled by temperature and monthly cumulative precipitation. Monthly cumulative precipitation in September was higher than January and April, in which Boodlea compostia and Gracilaria coronopifolia were dominant algae in September. The temperature growth responses of algae using the continuous-flow outdoor laboratory tank culture system fit their seasonal growth, reflecting the temperature-dependent manner of seasonal variations in abundance. It could be concluded from the present investigation that the structure of benthic macroalgal assemblage in Hsiao-Lu-Chiao island in southwestern Taiwan is affected by predicted natural and pulse disturbances. Temperature fluctuations involve in overall temporal variations in structure. Sea urchin herbivory and nutrient as pulse nutrient modulate the structure in fish farming area while monthly cumulative precipitation is associated with algal structure in non-fish farming area.