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.

思及當前捉襟見肘的財政窘境，無可避免地，債台高築的臺灣實陷入飲鴆止渴般以債養債之無限迴圈中，導致政府政策效能不彰、社會福利運作生弊亦無可厚非；於『公共債務法』之財政規範下，臺灣業已瀕臨法定舉債門檻，故不論是對短期政府支出之排擠、扭曲性稅率之稽徵抑或對長期經濟成長的斲傷，皆是身為中華民國國民真正惶悚不安之所在。 職是之故，本研究係採用一納入政府財政部門及貨幣當局之擴充『實質景氣循環模型』，藉以Sidrauski（1967）所提出的貨幣效用函數為出發點，將實質餘額引進理論模型，並透過計量操作捕捉實證期間起於西元1971年第一季迄至2007年第四季之政府政策函數，過程中，我們不難發現政府購買性支出及稅率皆存在相當的持續性，且對政府未償公債餘額之高低作出某種程度的反應。亦即，若政府實施公債融通政策，俾使期初公債餘額較高之際，則本期甚或往後各期的政府支出將遭受抑制和排擠，尤有甚者，政府勢必擬以提高未來稅率以茲挹注該債務之還本付息所造成的財政缺口；是以，本研究著眼於引進公債餘額對政府支出及稅率存在反饋作用下，財政政策與貨幣政策之總體效果及各總體變數之動態調整過程的風貌。即便公債發行或賒借為政府提供一財務週轉工具以裨益財政政策保有更靈活之彈性，然據模型所產生的結果顯示，就長期而論，政府必須維持一穩定之未償公債餘額，即公債水準具備『均數復歸』性質，而該財政目標係透過削減未來政府支出、調整扭曲性稅率及鑄幣稅融通政策方得以達成預算平衡，準此，該設定將造成公債融通之減稅政策對經濟體系具有實質效果，『公債融通』管道亦『非中立性政策』，從而傳統『李嘉圖等值定理』於本模型中無法成立。 就政策面層次而言，本研究試圖放寬『反饋法則』與政策係數之設定，以檢視透過不同程度之政府支出、稅率甚至貨幣供給途徑的改變來平衡因增加公債發行所造成的財政赤字，對經濟體系之長短期效果有何迥異處；是文亦藉由衝擊反應函數分別探討於政府支出增加、減稅措施及貨幣擴張之下，政策的傳遞機制與各總體變數之動態性質，顯然地，就高債務比率前提下，當政府戮力於刺激景氣而欲積極實施立竿見影的總體經濟政策之際，卻常因狃於急效而欲速不達，非但政策效果有限，亦可能使體系落入更為不景氣的田地，從而，財政惡化不啻為經濟危機的導火線也就不言而喻。再者，貨幣政策對體系之實質變數具有一定程度的作用，是故，本模型於短期內無法一窺『貨幣中立性』之堂奧，唯長期始得以復見。總括言之，政府亟須奉『健全財政』為圭臬，擬定政策時更得戒慎恐懼，並適切權衡利弊得失，以茲裨益有更具信心的經濟表現。 此外，本研究亦透過『效準』實驗以評估模型『配適度』之良窳，即便於反覆疊代法下，該模擬表現係瑕瑜互見而不盡完美，卻也大抵符合景氣循環之『典型化特徵』；然就實質景氣循環模型所為人詬病之勞動市場一隅而論，引進公債之反饋法則下的財政政策操作，無疑地改善了傳統工時與工資率動輒高度正相關之本質，從而獲致相對較低之理論相關係數，亦朝實證資料所呈現工時與工資率存在幾近零相關甚或低度負相關之表徵更邁進一大步。 / With current financial difficulties beyond government capability, it is inevitable that the already deep-in-debt Taiwan opted for momentary relief by paying debt through debt financing and ended up in an infinite loop, causing spiral-down performances in government policies and faulty operations of social welfare instruments. Taiwan has been on the verge of reaching the statutory upper limit of debt financing according to “The Public Debt Act” regulations and all nationals are becoming anxious about such impacts as crowding out of short-run government spending, levying of distorting taxes, and damages on long-run economic growth. To better understand the debt’s impacts, this research uses the “Real Business Cycle Model” extended by taking government treasury agency and monetary institution into account. Starting with Money In Utility Function (MIUF) as proposed by Sidrauski (1967) to introduce real money balance into the theoretical model and, in the process of econometric manipulation, to detect empirical governmental policy functions in the period between the first quarter, 1971 and the fourth quarter, 2007, it is not hard to discover that there are considerable persistence in both government purchases and tax rates, with manifestation of certain degree of responses to the total amount of outstanding bonds the government has yet to pay. In other words, a governmental bond financing policy designed to render high initial bonds outstanding tends to cause suppression and crowding out of government spending in current and even later periods. Furthermore, the government is bound to plan on raising taxes in the future in order to cut financial deficit gap caused by paying back the principles and interests of the debt. Therefore, this study focuses on presenting the macroeconomic effects of fiscal policies and monetary policies, as well as the dynamic adjustment processes of macroeconomic variables based on the impact of feedback effect of bonds outstanding on government spending and tax rates. Even thought public bonds issuance or debt financing serves as a governmental fiscal instrument for financial turnover to ensure flexibility of fiscal policies, our model shows that the government should, from a long-run perspective, maintain a stable amount of bonds outstanding. Put in a different way, the level of bonds outstanding shows “mean-reverting” characteristics which rely on future government spending cut, distorting tax adjustment and seigniorage financing policy to achieve balance of budget. As a result, such setup would cause the bond-financing backed tax deduction policies to create practical effects on economies and, as the bond financing instruments are “Non-Neutrality” policies, would render the “Ricardian Equivalence Theorem” invalid in our model. In the policy aspect, this study tries to relax both “feedback rules” and setup of policy parameters for investigating the differences between long-run and short-run effects on the economy by different degrees of changes in government spending, tax rates and even money supply channels which are used to balance the fiscal deficit caused by increased bond issuance. This article also studies, through the impulse response function, the policy propagation mechanism and the dynamics of key macroeconomic variables under the situation of government spending increase, tax deduction and monetary expansion. It is obvious that the government, in the case of high debt ratios and when making all endeavors to spur economy by implementing macroeconomic policies aimed for instant results, is accustomed to seeking quick fixes only to achieve very limited effects, sometimes even to drive the economy into further recession. It is therefore evident that fiscal degradation could lead to economic disaster. Moreover, as the monetary policies have certain degrees of influence on real variables of the economy, this model will not be able to clearly analyze the “neutrality of money” in such a short period of time. The effect will only reveal in the long run. In summary, the government should keep “sound finance” as the highest guiding principle and be extremely cautious in formulating policies in order to weigh all pros and cons discreetly, thus help to achieve a benefiting economic performance that generates more confidence. Furthermore, this study assesses “goodness of fit” of the model through a “calibration” experiment. Although the simulation results show, under recursive method, intermingled good and poor occasions that are beyond satisfaction, they generally agree with the “typical characteristics” of business cycles. However, in the aspect of long-criticized labor market of the real business cycle model, the fiscal policy operation under feedback rules with introduction of public debts for sure has greatly improved on the conventional intrinsic property of high correlation between labor hours and real wage rates, by delivering a relatively low theoretical correlation coefficient, which is a big step towards the empirical results of almost zero or even weakly negative correlation between labor hours and real wage rates.