近年來由於亞洲金融風暴的肆虐，產生經濟不景氣，使得失業的問題逐漸受到社會所關注，本論文企圖以三個時間序列方法：1.單變量ARIMA模型；2.轉換函數(TF)模型；3.向量自迴歸(VAR)模型來建立台灣地區的失業率時間序列預測模型。資料則是利用台灣地區民國75年1月至民國87年12月的失業率月資料作實證預測分析，為了知道資料是否來自時間趨勢模型，測試是否經過差分消掉一部份的記憶會發生預測的誤差，所以先以多步(multi-step)預測和一步(one-step)預測的方法計算出民國88年1月至88年12月預測值，而預測評估準則則採用(1)MAPE、RMSPE、MPE及泰爾不等係數(THEIL)；(2)變化方向誤差與趨勢變化誤差兩大方向來做預測比較。最後將算出的12期預測值與行政院主計處整體統計資料庫中所得到的失業率實際值利用預測評估準則做比較，結果發現一步預測法較多步預測法準確；而向量自迴歸模型(VAR)在大部份的預測期數上有較小的MAPE、RMSPE、MPE及THEIL值，因為此VAR模型考慮了在變數之間的共整合現象，有助於模型的預測，所以有較好預測的能力；反而是較複雜的ARIMA模型及轉換模型預測能力稍差一點。 / In this thesis, we plan to construct three time series models to forecast the Taiwan unemployment Rate. These time series models are ARIMA model、transfer function (TF) model and Vector Autoregressive (VAR) model. The data set consists of monthly observations for the period 75:1-87:12 for unemployment rate. We want to know if the data came from time trend model. First, we use multi-step forecasting and one-step forecasting to calculate 12 forecasted values from 88:01-88:12. Then We compare the prediction performance of these two methods by using：(1) MAPE、RMSPE、MPE and Theil’s Inequality Coefficient (THEIL)；(2) Direction of Change Error and trend Change Error etc. It is found that one-step forecasting is more correct than multi-step forecasting and the forecasting performance of VAR model is improved by explicitly taking account of cointegration between the variables in the model，so VAR model has lower MAPE、RMSPE、MPE and THEIL for most horizons. However，the more parsimonious ARIMA and transfer function models have higher MAPE、RMSPE、MPE for most horizons.

本篇論文擴充Ang et al. (2007)之基本架構,分別建構台灣各式月資料與季資料的物價指數預測模型,並進行預測以及實證分析。我們用以衡量通貨膨脹率的指標為 CPI 年增率與核心CPI 年增率。我們比較貨幣模型、成本加成模型、6 種不同設定的菲力浦曲線模型、3 種期限結構模型、隨機漫步模型、 AO 模型、ARIMA 模型、VAR 模型、主計處(DGBAS)、中經院(CIER) 及台經院(TIER) 之預測。藉由此研究,我們可以完整評估出文獻上常用之各式月資料及季資料預測模型的優劣。 我們實證結果顯示,在月資料預測模型樣本外預測績效表現方面, ARIMA 模 型對 2 種通貨膨脹率指標的樣本外預測能力表現最好。至於季資料預測模型樣本外預測績效表現, ARIMA 模型對未來核心 CPI 年增率的樣本外預測能力表現最好; 然而,對於 CPI 年增率為預測目標的預測模型則不存在最佳的模型。此外,實證分析中我們也發現本研究所建構的模型預測表現仍遜於主計處的預測,但部份模型的樣本外預測能力表現則比中經院與台經院的預測為佳。 / This paper compares the forecasting performance of inflation in Taiwan. We conduct various inflation forecasting methods (models) for two inflation measures(CPI growth rate and core-CPI growth rate) by using monthly and quarterly data. Besides the models of Ang et al. (2007), we also consider some macroeconomic models for comparison. We compare some Monetary models, Mark-up models, six variants of Phillips curve models, three variants of term structure models, a Random walk model, an AO model, an ARIMA model, and a VAR model. We also compare the forecast ability of these model with three different survey forecasts (the DGBAS, CIER, and TIER surveys). We summarized our findings as follows. The best monthly forecasting model for both inflation measures is ARIMA model. For quarterly core-CPI inflation, ARIMA model is also the best model; however, when comparing the quarterly forecasts for CPI inflation, there does not exist the best one. Besides, we also found that the DGBAS survey outperforms all of our forecasting methods/models, but some of our forecasting models are better than the CIER and TIER surveys in terms of MAE.

附最低保證投資型保險商品的特色在於無論投資者的投資績效好壞，保險金額皆享有一最低投資保證，過去關於此類商品的研究皆假設標的資產為單一資產，或依固定比例之投資組合，並沒有考慮到投資人自行配置投資組合的效果，但大部分市售商品中，投資人可以自行配置投資標，此情況之下，保險公司如何衡量適當的保證成本即為一相當重要之課題。 本研究假設投資人風險偏好服從冪次效用函數，並假設與保單所連結之投資標的有兩種資產，一為具有高風險高報酬的資產，另一為具有低風險低報酬之資產，在每個保單年度之初，投資人可以選擇配置在兩種資產之比例，我們運用黃迪揚(2009)所提出的動態規劃數值解之方法，計算出在考慮投資人自行配置資產之下，保證成本將會比固定比例之投資高出12個百分點。 此外，為了瞭解在不同資產報酬率的模型之下，保證成本是否會有不一樣的結論，除了對數常態模型之外，我們假設高風險資產與低風險資產服從ARIMA-GARCH(Autoregressive Integrated Moving Average-Generalized Autoregressive Conditional Heteroscedastic )模型，並得到較高的保證成本。 / The main characteristic of variable annuities (VA) with minimum benefits is that the benefit will be guaranteed. Previous literatures assume a specific underling asset return process when considering the guaranteed cost of VA; but they do not consider the portfolio choice opportunity of the policyholders. However, it is common for policyholders to rebalance his portfolio in many types of VA products. Therefore it’s important for insurance companies to apply an approximate method to measure the guaranteed cost. In this research, we assume that there are two potential assets in policyholders’ portfolio; one with high risk and high return and the other one with low risk and low return. The utility function of the policyholder is assumed to follow a power utility. We consider the asset allocation effect on the guaranteed cost for a VA with guaranteed minimum withdrawal benefits, finding that the guaranteed cost will increase 12% compared with a specific underling asset. The model effect of the asset return process is also examined by considering two different asset processes, the lognormal model and ARIMA-GARCH model. The solution of dynamic programming problem is solved by the numerical approach proposed by Huang (2009). Finally we get the conclusion which the guaranteed cost given by the ARIMA-GARCH model is greater than the lognormal model.

Abstracts in English, Afrikaans and Northern Sotho / More time series forecasting methods were researched and made available in recent years. This is mainly due to the emergence of machine learning methods which also found applicability in time series forecasting. The emergence of a variety of methods and their variants presents a challenge when choosing appropriate forecasting methods. This study explored the performance of four advanced forecasting methods: autoregressive integrated moving averages (ARIMA); artificial neural networks (ANN); support vector machines (SVM) and regression models with ARIMA errors. To improve their performance, bagging was also applied. The performance of the different methods was illustrated using South African air passenger data collected for planning purposes by the Airports Company South Africa (ACSA). The dissertation discussed the different forecasting methods at length. Characteristics such as strengths and weaknesses and the applicability of the methods were explored. Some of the most popular forecast accuracy measures were discussed in order to understand how they could be used in the performance evaluation of the methods. It was found that the regression model with ARIMA errors outperformed all the other methods, followed by the ARIMA model. These findings are in line with the general findings in the literature. The ANN method is prone to overfitting and this was evident from the results of the training and the test data sets. The bagged models showed mixed results with marginal improvement on some of the methods for some performance measures. It could be concluded that the traditional statistical forecasting methods (ARIMA and the regression model with ARIMA errors) performed better than the machine learning methods (ANN and SVM) on this data set, based on the measures of accuracy used. This calls for more research regarding the applicability of the machine learning methods to time series forecasting which will assist in understanding and improving their performance against the traditional statistical methods / Die afgelope tyd is verskeie tydreeksvooruitskattingsmetodes ondersoek as gevolg van die ontwikkeling van masjienleermetodes met toepassings in die vooruitskatting van tydreekse. Die nuwe metodes en hulle variante laat ʼn groot keuse tussen vooruitskattingsmetodes. Hierdie studie ondersoek die werkverrigting van vier gevorderde vooruitskattingsmetodes: outoregressiewe, geïntegreerde bewegende gemiddeldes (ARIMA), kunsmatige neurale netwerke (ANN), steunvektormasjiene (SVM) en regressiemodelle met ARIMA-foute. Skoenlussaamvoeging is gebruik om die prestasie van die metodes te verbeter. Die prestasie van die vier metodes is vergelyk deur hulle toe te pas op Suid-Afrikaanse lugpassasiersdata wat deur die Suid-Afrikaanse Lughawensmaatskappy (ACSA) vir beplanning ingesamel is. Hierdie verhandeling beskryf die verskillende vooruitskattingsmetodes omvattend. Sowel die positiewe as die negatiewe eienskappe en die toepasbaarheid van die metodes is uitgelig. Bekende prestasiemaatstawwe is ondersoek om die prestasie van die metodes te evalueer. Die regressiemodel met ARIMA-foute en die ARIMA-model het die beste van die vier metodes gevaar. Hierdie bevinding strook met dié in die literatuur. Dat die ANN-metode na oormatige passing neig, is deur die resultate van die opleidings- en toetsdatastelle bevestig. Die skoenlussamevoegingsmodelle het gemengde resultate opgelewer en in sommige prestasiemaatstawwe vir party metodes marginaal verbeter. Op grond van die waardes van die prestasiemaatstawwe wat in hierdie studie gebruik is, kan die gevolgtrekking gemaak word dat die tradisionele statistiese vooruitskattingsmetodes (ARIMA en regressie met ARIMA-foute) op die gekose datastel beter as die masjienleermetodes (ANN en SVM) presteer het. Dit dui op die behoefte aan verdere navorsing oor die toepaslikheid van tydreeksvooruitskatting met masjienleermetodes om hul prestasie vergeleke met dié van die tradisionele metodes te verbeter. / Go nyakišišitšwe ka ga mekgwa ye mentši ya go akanya ka ga molokoloko wa dinako le go dirwa gore e hwetšagale mo mengwageng ye e sa tšwago go feta. Se k e k a le b a k a la g o t šwelela ga mekgwa ya go ithuta ya go diriša metšhene yeo le yona e ilego ya dirišwa ka kakanyong ya molokolokong wa dinako. Go t šwelela ga mehutahuta ya mekgwa le go fapafapana ga yona go tšweletša tlhohlo ge go kgethwa mekgwa ya maleba ya go akanya. Dinyakišišo tše di lekodišišitše go šoma ga mekgwa ye mene ya go akanya yeo e gatetšego pele e lego: ditekanyotshepelo tšeo di kopantšwego tša poelomorago ya maitirišo (ARIMA); dinetweke tša maitirelo tša nyurale (ANN); metšhene ya bekthara ya thekgo (SVM); le mekgwa ya poelomorago yeo e nago le diphošo tša ARIMA. Go kaonafatša go šoma ga yona, nepagalo ya go ithuta ka metšhene le yona e dirišitšwe. Go šoma ga mekgwa ye e fepafapanego go laeditšwe ka go šomiša tshedimošo ya banamedi ba difofane ba Afrika Borwa yeo e kgobokeditšwego mabakeng a dipeakanyo ke Khamphani ya Maemafofane ya Afrika Borwa (ACSA). Sengwalwanyaki šišo se ahlaahlile mekgwa ya kakanyo ye e fapafapanego ka bophara. Dipharologanyi tša go swana le maatla le bofokodi le go dirišega ga mekgwa di ile tša šomišwa. Magato a mangwe ao a tumilego kudu a kakanyo ye e nepagetšego a ile a ahlaahlwa ka nepo ya go kwešiša ka fao a ka šomišwago ka gona ka tshekatshekong ya go šoma ga mekgwa ye. Go hweditšwe gore mokgwa wa poelomorago wa go ba le diphošo tša ARIMA o phadile mekgwa ye mengwe ka moka, gwa latela mokgwa wa ARIMA. Dikutollo tše di sepelelana le dikutollo ka kakaretšo ka dingwaleng. Mo k gwa wa ANN o ka fela o fetišiša gomme se se bonagetše go dipoelo tša tlhahlo le dihlo pha t ša teko ya tshedimošo. Mekgwa ya nepagalo ya go ithuta ka metšhene e bontšhitše dipoelo tšeo di hlakantšwego tšeo di nago le kaonafalo ye kgolo go ye mengwe mekgwa ya go ela go phethagatšwa ga mešomo. Go ka phethwa ka gore mekgwa ya setlwaedi ya go akanya dipalopalo (ARIMA le mokgwa wa poelomorago wa go ba le diphošo tša ARIMA) e šomile bokaone go phala mekgwa ya go ithuta ka metšhene (ANN le SVM) ka mo go sehlopha se sa tshedimošo, go eya ka magato a nepagalo ya magato ao a šomišitšwego. Se se nyaka gore go dirwe dinyakišišo tše dingwe mabapi le go dirišega ga mekgwa ya go ithuta ka metšhene mabapi le go akanya molokoloko wa dinako, e lego seo se tlago thuša go kwešiša le go kaonafatša go šoma ga yona kgahlanong le mekgwa ya setlwaedi ya dipalopalo. / Decision Sciences / M. Sc. (Operations Research)

本文旨在檢定台電公司現行季節電價月份劃分之合理性，並探討影響用戶別售電量與電費收入之經濟因素。為達成此目的，本文先就負載觀點與成本觀點進行群集分析，以檢定季節電價是否具統計意義之正當性；其次建立經濟計量模型，分別採用戶別之總售電量與總電費收入做為被解釋變數，運用民國88年1月至民國91年12月之月資料進行實證分析。本文建立之經濟模型有二，分別為時間序列以及複迴歸方程式模型。經檢定分析後，本文就各實證參數之經濟意涵加以闡示，最後並提出結論以及未來研究之方向。 本文透過月資料之群集分析，顯示夏月相對於非夏月之群集差異與台電公司現行季節電價夏月與非夏月之月份相一致，證實台電公司季節電價月份劃分之合理性。其次，透過ARIMA時間序列建立之短期電力需求預測模型，經實證結果顯示：電燈與電力用戶別之售電量均逐年增加，預測民國93年1月至民國99年12月，電燈用戶之年售電量平均成長率為3.33%、電力用戶為3.23%。再者，利用複迴歸模型進行實證分析之結果發現：(一)影響售電量之主要變數為溫度。惟因電燈用戶每隔兩月抄表一次，與電力用戶按月抄表之作業方式不同，故電燈用戶每月售電量係受前期(月)溫度影響，而電力用戶則受當期(月)溫度影響。(二)各用戶別之總電費收入與售電量有明顯相關，且經估算出各月售電量之電費收入彈性顯示：電燈用戶約為0.5，電力用戶約為1。由於總電費收入為總售電量與平均電價之乘積，故電燈用戶之電費收入增加1% 時，其售電量僅增加0.5%，顯示總電費的收入增加係有部分來自於平均電價的提高；換言之，就電燈用戶別而言，其電費收入增減變化之百分比除了會受到售電量增減幅度之影響外，亦反映了平均電價變化的情形。同理，對電力用戶來說，其各月售電量之電費收入彈性接近於1，表示電費收入變化1% 時，售電量亦增加1%，即電費收入之增減變化比例主要受到售電量之同向等幅變化所影響。 至於各用戶別之電費收入方面，電燈與電力兩類用戶自民國88年初至91年底四年期間均有逐年增加之趨勢，惟電力用戶之年增加幅度有隨時間遞減之現象，且歷年大抵以7-10月份較高，2月份最低。此外，影響用戶別電費收入之解釋變數中，各類用戶之售電量最為顯著，其參數值係隱示每增加一度售電量對其電費收入之影響。其中，電燈用戶之估計參數值為2.69，而電力用戶則為1.35。再者，由其電費收入之售電量彈性係數可以發現：電燈用戶約為1.2，電力用戶約為0.7，顯示電燈用戶總售電量增加1%時，總電費收入增加的幅度大於1%，而電力用戶則相反。推估電力用戶此一彈性係數較電燈用戶低之原因在於：電力用戶與電燈用戶之電價結構不同，前者係採需量電費與能量電費之兩部電價制，而後者僅包含流動電費之一部電價。最後，實證結果亦顯示電力系統之尖峰負載與負載率會影響電費收入，惟其影響幅度不大。 / A Study on Customer-by-Category Energy Sales and Power Sales Revenue Model: The Case of Taiwan Power Company Abstract The main purposes of this study are to examine the rationality of the seasonal pricing scheme defined by summer and non-summer months and to identify economic factors influencing customer-by-category energy sales and power sales revenue, utilizing the data of Taiwan Power Company (Taipower) as an empirical case. In order to achieve this objective, the cluster analysis from the perspective of load pattern and cost pattern are examined respectively to see if the seasonal pricing scheme has statistical meaning in its pattern differences in terms of summer vs. non-summer season. Second, two economic models including time-series analysis and multiple regression equations are formulated for the empirical case study. The subtotal energy sales and the subtotal power sales revenue by different type of customer categories, i.e. lighting and industrial customers, are set to be the explained variables. Data from January 1999 to December 2002 are collected for modeling simulation tests. The economic meanings and policy implications of the modeling results are elaborated on. And conclusions with directions for further research are presented. Through the cluster analysis utilizing monthly data within the time frame mentioned above, empirical research results on the grouping cluster of summer vs. non-summer months shows a consistent trend with those defined by Taipower’s present seasonal pricing scheme. Second, the empirical results of ARIMA time-series model show that the forecasted energy sales of both lighting and industrial customers will be gradually increasing through January 2004 to December 2010, and the average annual growth rate of energy sales for the lighting customer is 3.33%, and for the industrial customer is 3.23%. On the other hand, the empirical research results through the multiple regression equations show that the main factor affecting the energy sales is temperature. Due to the different time schedules for reading electricity meters between the lighting customer and the industrial customer, i.e. the time interval for reading the meter of lighting customers is every two months and for industrial customers is every month, the monthly energy sales of the lighting customer are directly related to the temperature of the previous month, while the monthly sales of the industrial customer are directly related to the temperature of the present month. In addition, for each type of customers, there is an obvious correlation between the total power sales revenue and the total energy sales. Furthermore, the estimated elasticity of the total power sales revenue versus total energy sales is about 0.5 for the lighting customer, and about 1 for the industrial customer. Since the total power sales revenue is the product of total energy sales times the average electricity price, when the total power sales revenue increases 1% with the total energy sales only increases 0.5%, it implies that the increase of total power sales revenue not just only comes from the increase of energy sales, but also partially affected by the increase of average electricity price. Similarly, for the industrial customer, when the elasticity of their monthly total power sales revenue versus total energy sales is close to 1, it implies that when the total power sales revenue increases 1%, the total energy sales also increase about 1%. In other words, the change of percentage of the total power sales revenue is mostly attributed to the variation of total energy sales, not because of the average electricity price. As for the simulation results of the total power sales revenue, those of the lighting and industrial customers are both gradually increasing between the years 1999 to 2002. However, the increasing pace of the industrial customer tended to slow down. Moreover, both types of the customers possess a similar trend that their total power sales are higher in statistical meaning for the months from July to October, and lower for February, for those above three years. Besides, among the variables affecting each type of customer’s power sales revenue, the energy sales is the most significant one, its parameter implies that whenever the total energy sales increases one unit, i.e. one kwh, it would affect the total power sales revenue by that amount equivalent to the figure of the parameter. According to the empirical results, the estimated parameter mentioned-above of the lighting customer is 2.69, and 1.35 of the industrial customer respectively. That implies one kwh unit price for the lighting customer is 2.69 N.T. dollars, and 1.35 N.T. dollars for the industrial customer. Moreover, from the elasticity of the total energy sales versus the total power sales revenue, it shows that the elasticity of the lighting customer is around 1.2, and the elasticity of the industrial customer is around 0.7. The underlining reason of the difference between the two figures could be that the electricity pricing structure of the lighting and industrial customers are quite different. The industrial customer is charged by two-part tariff including a demand charge for the capacity use and an energy charge for the kwh use. While the lighting customer is charged simply by a single rate, i.e. the energy use. Finally, the empirical results also show that the magnitude of the peak load and the load factor of the whole electricity system also affect the total power sales revenue of each type of the customer, though with much less effect.

Pension schemes are facing more difficulties on matching their underlying liabilities with assets, mainly due to faster mortality improvements for their underlying populations, better environments and medical treatments and historically low interest rates. Given most of the pension schemes are relatively much smaller than the national population, modelling and forecasting the small populations' longevity risk become urgent tasks for both the industrial practitioners and academic researchers. This thesis starts with a systematic analysis on the influence of population size on the uncertainties of mortality estimates and forecasts with a stochastic mortality model, based on a parametric bootstrap methodology with England and Wales males as our benchmark population. The population size has significant effect on the uncertainty of mortality estimates and forecasts. The volatilities of small populations are over-estimated by the maximum likelihood estimators. A Bayesian model is developed to improve the estimation of the volatilities and the predictions of mortality rates for the small populations by employing the information of larger population with informative prior distributions. The new model is validated with the simulated small death scenarios. The Bayesian methodologies generate smoothed estimations for the mortality rates. Moreover, a methodology is introduced to use the information of large population for obtaining unbiased volatilities estimations given the underlying prior settings. At last, an empirical study is carried out based on the Scotland mortality dataset.

Submitted by Milena Rubi (milenarubi@ufscar.br) on 2017-08-08T16:27:16Z No. of bitstreams: 1 VIEIRA_Joao_2016.pdf: 19409044 bytes, checksum: 86307e07137eea13b51469dd588082b5 (MD5) / Approved for entry into archive by Milena Rubi (milenarubi@ufscar.br) on 2017-08-08T16:27:25Z (GMT) No. of bitstreams: 1 VIEIRA_Joao_2016.pdf: 19409044 bytes, checksum: 86307e07137eea13b51469dd588082b5 (MD5) / Approved for entry into archive by Milena Rubi (milenarubi@ufscar.br) on 2017-08-08T16:27:32Z (GMT) No. of bitstreams: 1 VIEIRA_Joao_2016.pdf: 19409044 bytes, checksum: 86307e07137eea13b51469dd588082b5 (MD5) / Made available in DSpace on 2017-08-08T16:27:37Z (GMT). No. of bitstreams: 1 VIEIRA_Joao_2016.pdf: 19409044 bytes, checksum: 86307e07137eea13b51469dd588082b5 (MD5) Previous issue date: 2016-05-10 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Historically, in Brazil, small farms were used as a source of livelihood or recreation areas, currently also as a way to supplement income or a way to ensure security. These small investors, for lack of knowledge often end up leaving to make economic analyzes and market behavior before engaging in forestry projects. The work aimed to study the behavior of prices of eucalyptus wood and gum rosin of Pinus elliottii and the economic viability of traditional manner and project alternative that aim to produce these two products. For this we used econometric univariate models to forecast the prices of the products studied, finding a maximum error of 4.65%. Conditional heteroscedasticity models were applied to predict the volatility of resin prices and eucalyptus wood prices for the process, which were 5.72% and 4.28% per month respectively. The bootstrap simulation method was used to determine the volatility of pine resin prices, and obtained a value of 8.76%. The theory of real options was used to determine the economic viability of projects, had their results compared with traditional methodologies. Furthermore, the volatility obtained by the conditional heterocedasticity model and the bootstrap alternative approaches were applied to this theory, no difference in volatility analysis obtained by the bootstrap method. It is indicated greater care in feasibility studies for forest projects across the long term that they demand and due to different results according to each methodology as demonstrated in this work. / Historicamente, no Brasil, pequenas propriedades rurais eram utilizadas como fonte de subsistência ou áreas de lazer, atualmente também como uma maneira de complementar a renda ou garantir uma forma de previdência. Esses pequenos investidores, por falta de conhecimento, muitas vezes acabam deixando de fazer análises econômicas e de comportamento de mercado antes de se envolverem em projetos florestais. O trabalho teve como objetivo estudar o comportamento dos preços da madeira de eucalipto e goma resina de Pinus elliottii e a viabilidade econômica de forma tradicional e alternativa de projetos que visam produzir esses dois produtos. Para isso foram utilizados modelos econométricos univariados para fazer a previsão dos preços dos produtos estudados, encontrando um erro máximo de 4,65%. Modelos de heterocedasticidade condicional foram aplicados para prever a volatilidade dos preços de resina e dos preços de madeira de eucalipto para processo, que foram de 5,72% e 4,28% ao mês respectivamente. O método de simulação bootstrap foi utilizado para verificar a volatilidade dos preços de resina pinus, sendo obtido um valor de 8,76%. A teoria das opções reais foi utilizada para verificar a viabilidade econômica de projetos, teve seus resultados comparados com metodologias tradicionais. Além disso, com a volatilidade obtida pelo modelo de heterocedasticidade condicional e pelo método bootstrap foram aplicadas abordagens alternativas dessa teoria, havendo diferença na análise com a volatilidade obtida pelo método bootstrap. Indicam-se maiores cuidados nos estudos de viabilidade para projetos florestais frente ao longo prazo que os mesmos demandam e devido a diferentes resultados de acordo com cada metodologia conforme foi demonstrado nesse trabalho.

The economic growth and the increase in population has led to an increased demand for water for various purposes such as domestic consumption, irrigation, industrial use, power generation, navigation, recreation, and ecological requirements. With the increase in population, the per-capita water availability is continuously decreasing. Due to increase in demand and accompanying scarcity of water the conflict among the potential users of the resource is on raise. Hence, the allocation of the available water resource is a big challenge as the intersect oral and inter-regional water allocation is often competing and conflicting in nature. In the above context a good model to manage the available water resources would require reliable inputs on the available water resources. In the first part of this thesis we compare different techniques that are typically used for modeling the river water flow. Time series analysis (ARIMA) is compared with machine learning techniques such as support-vector regression (SVR) and neural network models. The performance of these techniques is compared by applying them to a long-term time-series data of the inflows of three tributaries of the river Cauvery into the Krishnaraja Sagar reservoir (KRS). Flow data over a period of 30 years from three different observation points established in upper Cauvery river sub-basin is analyzed. Specifically, a multi-layer feed forward network trained with a back-propagation algorithm and support vector regression with epsilon-insensitive loss function is compared with the ARIMA models. It is found that the performance of support vector regression model is superior to those of the other techniques considered. The second part of our thesis is to develop a model for optimal water allocation to the different sectors with the aim of maximizing the total utility of available water resource in a river basin. A hydro-economic modeling framework is developed that incorporates the economic assessment of the value of water. This inter-sectoral allocation problem is studied in the context of enforcing certain minimum water rights to every person for domestic use and a certain minimum irrigation need set out by the contingency plans of the state agriculture department in Cauvery river basin. A non-linear optimization model is built to obtain an optimal inter-sectoral water allocation policy. The study evaluates the economic impact of different parameters of competing demands such as water availability, population, basic water right (quantity), ground water contribution, and crop benefit. The optimal policies that implements the water allocation priorities as set out by the National Water Policy (2012) are compared. Further, results show that the basic water right can be secured for essential needs with optimal management of available surface and ground water resources. In the third part of thesis, we study the conflict of water sharing that arises between sectors/regions. We consider the river water-sharing problem between two agents along a river. Each agent has a stated claim to the river water. The Absolute Territorial Sovereignty (ATS) and Absolute Territorial Integrity (ATI) principles are promoted by different agents along the river as a means to maximize their individual benefit. However, these principles are invariably considered to be unjust by one or more of the other agents. Hence, it is preferred to have a negotiated water treaty that is perceived to be equitable and just by all. A one way downstream stream bilateral bargaining model can be used to guide the negotiated water treaty between the agents. In this bargaining framework we introduce the issue of negative externalities imposed by the upstream agent on the downstream agent/s in the form of pollution and/or flooding. This imposes a cost on the downstream agent to mitigate losses due to the negative externalities. A bargaining model that incorporates the impact of negative externalities is developed to guide the negotiated treaties. We identify individually rational bargaining strategies for a two agents transferable utility one way downstream river water sharing problem. The results characterize the agreement and disagreement points for bilateral trading

[pt] Esta tese propõe a utilização do DBSCAN (Density Based Spatial Clustering of Applications with Noise) para separar os componentes de ruído na fase de agrupamento das autotriplas da Análise Singular Espectral (SSA) de Séries Temporais. O DBSCAN é um método moderno de clusterização (revisto em 2013) e especialista em identificar ruído através de regiões de menor densidade. O método de agrupamento hierárquico até então é a última inovação na separação de ruído na abordagem SSA, implementado no pacote R- SSA. No entanto, o método de agrupamento hierárquico é muito sensível a ruído, não é capaz de separá-lo corretamente, não deve ser usado em conjuntos com diferentes densidades e não funciona bem no agrupamento de séries temporais de diferentes tendências, ao contrário dos métodos de aglomeração à base de densidade que são eficazes para separar o ruído a partir dos dados e dedicados para trabalhar bem em dados a partir de diferentes densidades. Este trabalho mostra uma melhor eficiência de DBSCAN sobre os outros métodos já utilizados nesta etapa do SSA, garantindo considerável redução de ruídos e proporcionando melhores previsões. O resultado é apoiado por avaliações experimentais realizadas para séries simuladas de modelos estacionários e não estacionários. A combinação de metodologias proposta também foi aplicada com sucesso na previsão de uma série real de velocidade do vento. / [en] This thesis proposes using DBSCAN (Density Based Spatial Clustering of Applications with Noise) to separate the noise components of eigentriples in the grouping stage of the Singular Spectrum Analysis (SSA) of Time Series. The DBSCAN is a modern (revised in 2013) and expert method at identify noise through regions of lower density. The hierarchical clustering method was the last innovation in noise separation in SSA approach, implemented on package R-SSA. However, is repeated in the literature that the hierarquical clustering method is very sensitive to noise, is unable to separate it correctly, and should not be used in clusters with varying densities and neither works well in clustering time series of different trends. Unlike, the methods of density based clustering are effective in separating the noise from the data and dedicated to work well on data from different densities This work shows better efficiency of DBSCAN over the others methods already used in this stage of SSA, because it allows considerable reduction of noise and provides better forecasting. The result is supported by experimental evaluations realized for simulated stationary and non-stationary series. The proposed combination of methodologies also was applied successfully to forecasting real series of wind s speed.

Mise en oeuvre de techniques de modélisation récentes pour la prévision statistique et économique. / Doctorat en Sciences / info:eu-repo/semantics/nonPublished