In this study, the rainfall, yield distributed water and water sold¡K¡Ketc., 41 parameters from 1974 to 2006 were assessed the reasonable water rate adjustment. At first, 41 parameters were analyzed by SPSS software for descriptive statistiics, Pearson relational analyzing the data of input/output of correlation with a £^ value and screening of variables. Then, actual water price and designed water price will be the out variable. Try to find the optimal neural network structure and try to analyze and produce the water pricing structure. The results show that the unit profit/loss from sales of Taiwan Water Corporation(TWC) for 33 years from 1974-2006, there are 11 years positive and 22 years negative, especially the past 20 years only on 1990, 2003, and 2005 are positive, the others are negative. TWC had not obtained the reasonable profit. Because since 1996, the range of return on water investment and return on equity are -0.98%-0.1% and 0.07%-0.52% lower than legal standard 5%, respectively. Moreover, the rate of water price for the household consumption expenditure from 0.791% in 1982 decrease to 0.39% in 2006. To compare with the rational level for World Health Organization asserted 2%-4%. The water price of Taiwan is only 10%-20% of the level. Furthermore, the rate of water price for disposable income is from 0.606% in 1982 drops to 0.305% in 2006 and the rate for GDP is 0.18%-0.2% in the past 10 years. In this study, the actual water price and designed water price were set as output parameter. The input variables divide to 29 and 19 units and hidden layer is set 1 or 2 layers. BPN(Back-Propagation Network) were through trial and error method to training, testing, and production the output. The training results show that 19 variables is better than 29 variables while we use actual water price and 2 hidden layers is better than 1 layer. However, when we use designed water price, 19 variables is still better than 29 variables, but 1 hidden layer is better than 2 hidden layers. The best production of water price of 1981, 1991, 1996, 2001, 2006 are 9.20, 12.62, 20.09, 23.07, 24.39 NT$, respectively. The values are close to designed water price 9.0, 12.6, 19.1, 22.2, 25.5 NT$. Whether we use 29 or 19 variables, 1 or 2 hidden layers, the training results indicated that the water price designed by household consumption expenditure is better than actual water price. Thus, the historical water price did not correspond to real operating costs for TWC in the past. In addition, the designed water price in this research can more correlated with the operating cost and efficiency of TWC.

Submitted by Sara Ribeiro (sara.ribeiro@ucb.br) on 2017-04-05T12:06:21Z No. of bitstreams: 1 IvaneteFernandesdoPradoDisserta????o2017.pdf: 4250965 bytes, checksum: 8e414ff4c2763ba7d4918d1a888abacd (MD5) / Made available in DSpace on 2017-04-05T12:06:21Z (GMT). No. of bitstreams: 1 IvaneteFernandesdoPradoDisserta????o2017.pdf: 4250965 bytes, checksum: 8e414ff4c2763ba7d4918d1a888abacd (MD5) Previous issue date: 2017-02-23 / Among diseases that happen during pregnancy, hypertension is considered to have the most detrimental effects on the mother's, fetus and newborn's body. Among the repercussions that this disease can bring to the fetus are prematurity and low birth weight, which are considered the main determinants of the risk of death in the neonatal period and of the developmental problems in childhood, besides increasing the probability of occurrence of various diseases in adult life. This study objective to identify the prevalence of arterial hypertension during pregnancy and to associates it with prematurity and low birth weight. It is a case-control study carried out in the Public hospital in the municipality of Guanambi-Bahia a, using the statements of live births and medical records of pregnant women. The sample was composed of 207 pregnant women and 212 live newborns, divided into case-group and control group. A research protocol was used to record newborn data, maternal data, gestation/childbirth, and gestational risk factors: hemorrhage, early placental abruption, placenta previa, premature amniorrexis, oligodendrogram, TORCHS, urinary tract infection, alcoholism , diabetes, hypertension, smoking, medication use, drugs and childbirth. The study was approved by the Research Ethics Committee of the Catholic University of Bras??lia and Secretary of Health of the State of Bahia. The associations of the possible variables that predict prematurity and low birth weight were analyzed using hierarchical models of multiple logistic regression. The prevalence of pregnant women with arterial hypertension during admission to labor in the studied sample was 13.5%. Hypertension was associated with both prematurity and low birth weight. There was a significant association (p <0.05) between gestational risk factors and prematurity. Gestational risk factors and prematurity were also associated with low birth weight (p <0.05). Mothers with hypertension presented 3.47 (95% CI: 1.37-8.81) times more chance of having a preterm child and 2.55 (1.03-6.32) times more chance of having a child with low birth weight (P <0.05); mothers who presented other risk factors that aren't the hypertension had 2.21 (95% CI: 1.13-4.31) times more chance of having preterm (p <0.05) and 1.79 (95% CI: 96-3,36) time more change of having children with low birth weight (p = 0.0681), when they are compared to mothers without gestational risk factors. Mothers with less than seven prenatal visits had 2.31 (1.22-4.34) times more chance of having a preterm child than those with at least 7 visits (p<0.05). Low birth weight was not associated with the number of prenatal visits. Gestational risk factors may induce prematurity and low birth weight. Faced with this outcome, it is advised that pregnant women in this situation should be referred and ensured high-risk follow-up, as well as guided on healthy lifestyles and systematic practice of physical activity. / Dentre as doen??as que acontecem a mulher durante a gravidez, a hipertens??o arterial ?? considerada a que traz mais efeitos prejudiciais para o organismo da m??e, do feto e do rec??m-nascido. Entre as repercuss??es que essa doen??a pode trazer para o feto, encontram-se a prematuridade e o baixo peso ao nascer, que s??o considerados os principais determinantes do risco de morte no per??odo neonatal e dos problemas de desenvolvimento na inf??ncia, al??m de aumentarem a probabilidade de ocorr??ncia de v??rias doen??as na vida adulta. Este estudo objetiva identificar a preval??ncia de hipertens??o arterial durante a gesta????o e associ??-la com a prematuridade e o baixo peso ao nascer. Trata-se de estudo tipo caso-controle realizado no hospital p??blico do munic??pio de Guanambi-Bahia, utilizando as declara????es de nascidos vivos e prontu??rios das gestantes. A amostra foi composta por 207 gestantes e 212 rec??m-nascidos vivos, divididos em grupo-caso e grupo-controle. Foi utilizado um protocolo de pesquisa para registro dos dados do rec??m-nascido, dados m??e, gesta????o/parto e fatores de risco gestacional: hemorragia, deslocamento prematuro de placenta, placenta pr??via, amniorrexe prematura, oligodr??minio, TORCHS, infec????o do trato urin??rio, alcoolismo, diabetes, hipertens??o, tabagismo, uso de medica????es/drogas e parto. O estudo teve aprova????o do Comit?? de ??tica em Pesquisa da Universidade Cat??lica de Bras??lia e Secretaria de Sa??de do Estado da Bahia. As associa????es das poss??veis vari??veis preditoras da prematuridade e baixo peso ao nascer foram analisadas por meio de modelos hierarquizados de regress??o log??stica m??ltipla. A preval??ncia de gestantes com hipertens??o arterial durante a interna????o para o parto na amostra estudada foi de 13,5%. A hipertens??o arterial esteve associada tanto com a prematuridade, quanto com o baixo peso ao nascer. Houve associa????o significativa (p<0,05) entre os fatores de risco gestacional e a prematuridade. Os fatores de risco gestacional e a prematuridade tamb??m estiveram associados com o baixo peso do rec??m-nascido (p<0,05). M??es com hipertens??o apresentaram 3,47 (IC95%: 1,37-8,81) vezes mais chance de ter filho prematuro e 2,55 (1,03-6,32) vezes mais chance de ter filho com baixo peso, m??es que apresentaram outros fatores de risco que n??o a hipertens??o tiveram 2,21 (IC95%:1,13-4,31) vezes mais chance de ter filho prematuro (p<0,05) e 1,79 (IC95%: 0,96-3,36) vez mais chance de terem filhos com baixo peso (p=0,0681), quando comparadas com m??es sem fatores de risco gestacional. As m??es com menos de sete consultas pr??-natais tiveram 2,31 (1,22-4,34) vezes mais chance de ter filho prematuro do que as com pelo menos 7 consultas (p<0,05). O baixo peso n??o esteve associado com o n??mero de consultas de pr??-natal. Fatores de risco gestacional podem induzir a prematuridade e o baixo peso ao nascer. Frente a esse desfecho, orienta-se que gestantes nessa situa????o sejam encaminhadas e asseguradas do acompanhamento de alto risco, bem como orientadas sobre estilos de vida saud??veis e pr??tica de atividade f??sica sistem??tica.

Tese (doutorado)—Universidade de Brasília, Faculdade de Ciências da Saúde, Programa de Pós-Graduação em Ciências da Saúde, 2016. / Submitted by Camila Duarte (camiladias@bce.unb.br) on 2017-02-03T12:31:46Z No. of bitstreams: 1 2016_PriscillaPerezdaSilvaPereira.pdf: 2295378 bytes, checksum: e974f7bba042b2833aeaacc85005b532 (MD5) / Approved for entry into archive by Ruthléa Nascimento(ruthleanascimento@bce.unb.br) on 2017-02-17T19:23:30Z (GMT) No. of bitstreams: 1 2016_PriscillaPerezdaSilvaPereira.pdf: 2295378 bytes, checksum: e974f7bba042b2833aeaacc85005b532 (MD5) / Made available in DSpace on 2017-02-17T19:23:30Z (GMT). No. of bitstreams: 1 2016_PriscillaPerezdaSilvaPereira.pdf: 2295378 bytes, checksum: e974f7bba042b2833aeaacc85005b532 (MD5) / Introdução: O baixo peso ao nascer (BPN) é um indicador de saúde pública que reflete as condições de vida da mãe bem como a qualidade dos cuidados prestados durante o ciclo gravídico. Entre os fatores associados ao BPN estão os hábitos de vida da gestante. O consumo de cigarro e bebidas alcoólicas durante a gestação pode interferir no crescimento e desenvolvimento fetal. Objetivos: Analisar a associação entre o consumo materno de cigarro e de bebidas alcóolicas durante a gestação e o BPN nos países das Américas e em uma região do Nordeste brasileiro. Métodos: Foram realizadas duas revisões sistemáticas da literatura com metanálise incluindo estudos de coorte e caso-controle conduzidos nos países das Américas. A primeira revisão sistemática investigou a associação entre o fumo materno durante a gestação e o BPN e incluiu estudos publicados entre os anos de 1986 e 2016. E, a segunda revisão sistemática analisou a associação entre o uso de bebidas alcóolicas e o BPN com estudos publicados entre os anos de 1983 e 2016. A qualidade metodológica dos estudos incluídos foi avaliada por meio da escala Newcastle-Ottawa. Nas duas revisões sistemáticas foram realizadas metanálises com efeito randômico para obtenção da medida de Odds Ratio (OR) e de seu respectivo Intervalo de Confiança (IC 95%). A heterogeneidade entre os estudos foi investigada por meio de metarregressões e análises por subgrupo. O terceiro estudo desenvolvido nesta tese foi do tipo caso-controle e incluiu 1.612 puérperas e seus respectivos recém-nascidos na região do Vale do São Francisco, Nordeste do Brasil. Foram realizadas análises de regressão linear simples, regressão logística e utilizado um modelo aditivo para analisar a interação entre o cigarro e o álcool. Resultados: O fumo materno foi associado ao baixo peso ao nascer nos países das Américas (OR 2,00; IC 95%: 1,77; 2,26; I2: 66,3%). As metarregressões e análises de subgrupo indicaram que essa associação independe de aspectos metodológicos ou região das Américas. O consumo materno de álcool não foi associado ao BPN nas Américas (OR: 1,06; IC 95%: 0,94; 1,20; I2: 73,9%). No Nordeste brasileiro o consumo materno de mais de 10 cigarros ao dia foi associado ao BPN (OR: 3,87; IC 95%: 1,85-8,10), assim como o alto consumo de álcool (OR: 2,10; IC 95%: 1,23; 3,58). A interação entre o álcool e o fumo consumidos durante a gestação não foi estatisticamente significativa. Conclusão: Esta tese reforça a evidência que o BPN está associado ao consumo materno de cigarro e bebidas alcóolicas durante a gestação. Apesar de não ter sido encontrada associação para o baixo a moderado consumo isso não é indício que o uso dessas substâncias seja seguro. A recomendação de se evitar o uso do álcool e cigarros durante a gestação ainda é a melhor opção para as gestantes. / Introduction: Low birth weight is a public health indicator that reflects maternal health conditions and the healthcare quality delivered during pregnancy. The maternal lifestyle is one of the factors associated with low birth weight. Both alcohol and cigarettes consumption during pregnancy may interfere in fetal growth and development. Aim: To analyze the association between maternal consumption of cigarettes and alcohol beverages during pregnancy and the low birth weight in countries of the Americas and in a region located in the Northeast of Brazil. Methods: Two systematic reviews with meta-analyses were conducted including cohort and case-control studies carried out in countries of the Americas. The first systematic review investigated the association between maternal smoking during pregnancy and low birth weight and included studies published between the years 1986 and 2016. The second one analyzed the association between the consumption of alcoholic beverages and low birth weight with included studies published between the years 1983 and 2016. The methodological quality of the studies was assessed using the Newcastle-Ottawa scale. In both meta-analyses, the random-effects model was used to obtain the Odds Ratio with its confidence interval. The between-study heterogeneity was investigated through meta-regressions and subgroup analyses. The third study developed in this thesis was a case-control, including 1,612 mothers and their newborns in the region of Vale do São Francisco, in the Northeast region of Brazil. In the data analysis, simple and logistic regression models were conducted; besides, an interaction model was carried out to analyze effect modification between cigarettes and alcohol consumption during pregnancy. Results: Maternal smoking was associated with low birth weight in the countries of the Americas (OR 2.00; CI 95%: 1.77; 2.26; I2: 66.3%). Meta-regressions and subgroup analyses indicate that this association occurs regardless of methodological aspects and the region of the Americas. The consumption of alcoholic beverages was not associated with low birth weight in the Americas (OR: 1.06; CI 95%: 0.94; 1.20; I2: 73.9%). In the Brazilian Northeast, the maternal consumption of more than 10 cigarettes a day was associated with low birth weight (OR: 3.87; CI 95%: 1.85; 8.10), as well as the high level of alcohol consumption (OR: 2.10; IC 95%: 1.23; 3.58). The interaction between alcohol and smoking were not statistically significant. Conclusion: This thesis reinforces that low birth weight is associated with the maternal consumption of cigarettes and alcoholic beverages during pregnancy. The lack of association between low birth weight and a low to moderate consumption of substances does not suggest that this consumption is secure. Avoiding the consumption of these substances by pregnant women is still the best recommendation.

Submitted by Kamila Costa (kamilavasconceloscosta@gmail.com) on 2015-07-30T20:30:54Z No. of bitstreams: 2 Dissertação - Jamile L Freire.pdf: 613265 bytes, checksum: 6717901329ef990dbfa3295cede3d795 (MD5) ficha catalografica.pdf: 1849 bytes, checksum: c0de4bb17c675bdb8d657be247620a06 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-08-04T14:03:15Z (GMT) No. of bitstreams: 2 Dissertação - Jamile L Freire.pdf: 613265 bytes, checksum: 6717901329ef990dbfa3295cede3d795 (MD5) ficha catalografica.pdf: 1849 bytes, checksum: c0de4bb17c675bdb8d657be247620a06 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2015-08-04T14:07:45Z (GMT) No. of bitstreams: 2 Dissertação - Jamile L Freire.pdf: 613265 bytes, checksum: 6717901329ef990dbfa3295cede3d795 (MD5) ficha catalografica.pdf: 1849 bytes, checksum: c0de4bb17c675bdb8d657be247620a06 (MD5) / Made available in DSpace on 2015-08-04T14:07:46Z (GMT). No. of bitstreams: 2 Dissertação - Jamile L Freire.pdf: 613265 bytes, checksum: 6717901329ef990dbfa3295cede3d795 (MD5) ficha catalografica.pdf: 1849 bytes, checksum: c0de4bb17c675bdb8d657be247620a06 (MD5) Previous issue date: 2012-11-01 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Low Birth Weight (LBW), multifactorial event and implicated as a major contributor to the rise in infant mortality in developing countries. The objective of this study was to investigate the determinants of LBW. Cross-sectional study of descriptive population-based involved 384 mothers attended in a public hospital. Forms with closed questions were used on the socioeconomic, demographic condition of pregnancy and nutritional standard features, and anthropometric measurements of prenatal card. The results showed an incidence of LBW infants 6.5%, not registering this association influence of socioeconomic variables (p> 0.05). In relation to maternal characteristics, the number of consultations in prenatal care, the previous history of abortion and the type of delivery showed a statistically significant relationship in determining the BPN (p <0.05). Among LBW infants 7.3% were children of adolescents (p> 0.05), and zero the occurrence of this event in mothers with older age (≥ 35 years). Analysis of these results shows a low occurrence of LBW in the universe studied, however it reinforces the need to implement a program of assistance and guidance to more efficient prenatal care for pregnant women teens. / O Baixo Peso ao Nascer (BPN), evento de etiologia multifatorial e incriminado como um dos principais responsáveis da elevação da mortalidade infantil nos países em desenvolvimento. O objetivo deste trabalho foi investigar os fatores determinantes do BPN. Estudo transversal de base populacional descritivo envolveu 384 puérperas atendidas em uma maternidade pública. Foram utilizados formulários com perguntas fechadas sobre a condição socioeconômica, demográfica, características da gestação e padrão nutricional, sendo as medidas antropométricas obtidas do cartão do pré-natal. A análise dos resultados evidenciou uma ocorrência de RNBP de 6,5%, não se registrando nesta associação influencia das variáveis socioeconômicas (p>0,05). Em relação às características maternas, o número de consultas realizadas no pré-natal, a história pregressa de aborto e o tipo de parto apresentaram relação estatisticamente significativa na determinação do BPN (p<0,05). Dentre os RNBP 7,3% eram filhos de adolescentes (p>0,05), sendo nula a ocorrência deste evento nas mães com idade mais avançada (≥ 35 anos). A análise destes resultados evidencia uma baixa ocorrência de BPN no universo estudado, entretanto reforça a necessidade de implementação de um programa de assistência e orientação ao pré-natal mais eficiente para as gestantes adolescentes.

Philosophiae Doctor - PhD / Adolescent hookah pipe use is a public health concern because it poses several health, environmental, and economic risks. Self-determination theory (SDT) posits that people are motivated to engage in certain behaviours in an attempt to satisfy their basic psychological

本研究計畫探討以RN神經網路模型預測國內債券價格的效度。目前一般用於財務預測的神經網路論著主要為BPN模型，惟BPN模型有其限制，所以本研究計畫將（1）分析比較統計計量模型，BPN神經網路，RN神經網路系統對國內公債價格之預測績效。（2）分析不同時期的預測能力，找出景氣和預測變數的關係，同時將比較各個時期統計計量模型和神經網路模型是否同時有效， 抑或有些有效， 有些無效，以探討各工具是否具有互補性或替代性。並探討預測績效是否受到背後經濟環境的影響。 我們研究對象為國內公債，其每日交易資料取樣時間自民國八十一年開始。影響債券價格的因素可拆解成實質利率，預期通貨膨脹率和風險貼水三層面，本研究總體變數之選取，亦循此三項範疇以求周延。 本研究之研究成果對理論及實務應用將有下列三項預期貢獻：（1）比較不同其常的債券在不同景氣狀況下，各不同預測模型的預測效度差異，探討各時期各工具之預測能力，可提供投資實務界對預測工具之選擇，應用與搭配。（2）對債券報酬率預測研究，分析總體變數，利率風險等變數對債券報酬率的影響，可進一步暸解影響債券價格的相關因素及程度。（3）以往神經網路應用在財務預測領域上， 皆以BPN 神經網路為主，此處引進RN神經網路，比較兩者的表現，可提供學術理論界之驗證。 / This research project empirically investigates the accuracy of Reasoning Neural Networks (RN) in forecasting Taiwan's bond prices. We explore (1) the relative predictive abilities of Vector Error Correction Model (VECM), which serve as a representative econometric model, Back Propagation Neural Networks (BPN), which is adopted by most current studies in the application of neural networks in finance, and RN, and (2) th3 potential variations in the three models' predictive power in different phases of economic cycle. Specifically, we aim to study if the three models substitute or complementone another. In addition, we explore the extent to which the relativepredictive abilities of the three models varies with underlying macroecomonic factors. The explanatory variables adopted in this study include all potential drives to (real) risk-free rate, expected inflation rate, and riskspremiums. In this study, we examine the government bond terms to maturity,coupon rate, and prices of government bonds during 1992-1995. This project would contribute to both academic and application researchin the following three aspects : (1) Few, if any , prior study explores whether and how various neuralnetworks and/or eco- nomic models perform under different macro-economicvariables. Our empirical results may indicate an appropriate model ( ormodels ) to improve forecasting of bond prices. (2) This study shows how RN, BPN, and VECM models perform in forecastinggovernment bonds yields to maturity. (3) The BPN model prevails in financial forecasting. Nevertheless, BPNis subject to a few short comings and may thus be a sub-optimal model. This study analyzes if RN is more cost-effective in forecasting bond prices than BPN.

本研究考慮重要且對台股大盤指數走勢有連動影響的因素，主要納入對台股有領頭作用的美國三大股市，那斯達克(NASDAQ)指數、道瓊工業(Dow Jones)指數、標準普爾500(S&P500)指數;其他對台股緊密連動效果的國際股票市場，香港恆生指數、上海證券綜合指數、深圳證券綜合指數、日經225指數;以及納入左右國際經濟表現的國際原油價格走勢，美國西德州原油、中東杜拜原油和歐洲北海布蘭特原油;在宏觀經濟因素方面則考量失業率、消費者物價指數、匯率、無風險利率、美國製造業重要指標的存貨/銷貨比率、影響貨幣數量甚鉅的M1B;在技術分析方面則納入多種重要的指標，心理線 (PSY) 指標、相對強弱(RSI) 指標、威廉(WMS%R) 指標、未成熟隨機(RSV) 指標、K-D隨機指標、移動平均線(MA)、乖離率(BIAS)、包寧傑%b和包寧傑帶狀寬度(BandWidth%);所有考量因素共計35項，因為納入重要因子比較多，所以完備性較高。 本研究先採用的贏者全拿(Winner-Take-All) 競爭學習策略的自組織映射網路(Self-Organizing Feature Maps, SOM)，藉由將相似資料歸屬到已身的神經元萃取出關聯分類且以計算距離來衡量神經元的離散特徵，對於探索大量且高維度的非線性複雜特徵俱有優良的因素相依性投射效果，將有利於提高預測模式精準度。在線性擬合部分則結合倒傳遞(Back-Propagation, BP)、Elman反饋式和徑向基底函數類網路(Radial-Basis-Function Network, RBF)模式為指數預測輸出，並對台股加權指數隔日收盤指數進行預測和評量。而在傳統的Elman反饋式網路只在隱藏層存在反饋機制，本研究則在輸入層和隱藏層皆建立反饋機制，將儲存在輸入層和隱藏層的過去時間資訊回饋給網路未來參考。在徑向基底函數網路方面，一般選取中心聚類點採用隨機選取方式，若能有效降低中心點個數，可降低網路複雜度，本研究導入垂直最小平方法以求取誤差最小的方式強化非監督式學習選取中心點的能力，以達到網路快速收斂，提昇網路學習品質。 研究資料為台股指數交易收盤價，日期自2001/1/2，至2011/10/31共2676筆資料。訓練資料自2001/1/2至2009/12/31，共2223筆;實證測試資料自2010/1/4至2011/10/31，計453個日數。主要評估指標採用平均相對誤差(AMRE)和平均絕對誤差 (AAE)。在考慮因子較多的狀況下，實證結果顯示，在先透過SOM進行因子聚類分析之後，預測因子被分成四個組別，分別再透過BP、Elman recurrent和RBF方法進行線性擬合，平均表現方面，以RBF模式下的四個群組因子表現最佳，其中RBF模式之下的群組4，其AMRE可達到0.63%，最差的AMRE則是群組1，約為1.05%;而Elman recurrent模式下的四組群組因子之ARME則介於1.01%和1.47%之間;其中預測效果表現最差則是BP模式的預測結果。顯示RBF具有絕佳的股價預測能力。最後，在未來研究建議可以運用本文獻所探討之其他數種類神經網路模式進行股價預測。 / In this study, we considering the impact factors for TWSE index tendency, mainly aimed at the three major American stock markets, NASDAQ index, Dow Jones index, S&P 500, which leading the Taiwan stock market trend; the other international stock markets, such as the Hong Kong Hang-Seng Index, Shanghai Stock Exchange Composite Index, Shenzhen Stock Exchange Composite Index, NIKKEI 225 index, which have close relationship with Taiwan stock market; we also adopt the international oil price trend, such as the West Texas Intermediate Crude Oil in American, the Dubai crude oil in Middle Eastern, North Sea Brent crude oil in European, which affects international economic performance widely; On the side of macroeconomic factors, we considering the Unemployed rate, Consumer Price Index, exchange rate, riskless rate, the Inventory to Sales ratio which it is important index of American manufacturing industry, and the M1b factor which did greatly affect to currency amounts; In the part of Technical Analysis index, we adopt several important indices, such as the Psychology Line Index (PSY), Relative Strength Index (RSI), the Wechsler Memory Scale—Revised Index (WMS%R), Row Stochastic Value Index (RSV), K-D Stochastics Index, Moving Average Line (MA), BIAS, Bollinger %b (%b), Bollinger Band Width (Band Width%);All factors total of 35 which we have considered the important factor is numerous, so the integrity is high. In this study, at first we adopt the Self-Organizing Feature Maps Network which based on the Winner-Take-All competition learning strategy, Similar information by the attribution to the body of the neuron has been extracted related categories and to calculate the distance to measure the discrete characteristics of neurons, it has excellent projection effect by exploring large and complex high-dimensional non-linear characteristics for all the dependency factors , would help to improve the accuracy of prediction models, would be able to help to improve the accuracy of prediction models. The part of the curve fitting combine with the back-propagation (Back-Propagation, BP), Elman recurrent model and radial basis function network (Radial-Basis-Function Network, RBF) model for the index prediction outputs, forecast and assessment the next close price of Taiwan stocks weighted index. In the traditional Elman recurrent network exists only one feedback mechanism in the hidden layer, in this study in the input and hidden layer feedback mechanisms are established, the previous information will be stored in the input and hidden layer and will be back to the network for future reference. In the radial basis function network, the general method is to selecting cluster center points by random selection, if we have the effectively way to reduce the number of the center points, which can reduces network complexity, in this study introduce the Orthogonal Least Squares method in order to obtain the smallest way to strengthen unsupervised learning center points selecting ability, in order to achieve convergence of the network fast, and improve network learning quality. Research data for the Trading close price of Taiwan Stock Index, the date since January 2, 2001 until September 30, 2011, total data number of 2656. since January 2, 2001 to December 31, 2009 a total number of 2223 trading close price as training data; empirical testing data, from January 4, 2010 to September 30, 2011, a total number of 433. The primary evaluation criteria adopt the Average Mean Relative Error (AMRE) and the Average Absolute Error (AAE). In the condition for consider more factors, the empirical results show that, by first through SOM for factor clustering analysis, the prediction factors were divided into four categories and then through BP, Elman recurrent and RBF methods for curve fitting, at the average performance , the four group factors of the RBF models get the best performance, the group 4 of the RBF model, the AMRE can reach 0.63%, the worst AMRE is group 1, about 1.05%; and the four groups of Elman recurrent model of ARME is between 1.01% and 1.47%; the worst prediction model is BP method. RBF has shown excellent predictive ability for stocks index. Finally, the proposal can be used in future studies of the literatures that we have explore several other methods of neural network model for stock trend forecasting.