晚近金融環境快速變化，金融商品不斷的開發及付款方式不斷的革新，逐漸改變消費者持有現金作為支付工具的習性，使得傳統上只用所得和利率來解釋貨幣需求函數已不太具解釋能力。因此本文將「金融創新」的影響納入貨幣需求函數。 本文實證的變數選擇方面，採用通貨和信用卡的使用分別代表貨幣需求和金融創新以作為分析的重點。實證方法方面，同時採用一般常用的「聯立方程式」－Johansen Procedure和「單迴歸方程式」－OLS、動態OLS(Saikkonen's OLS及Phillips and Loretan's NLS)及FM-OLS來估計信用卡和通貨的關係，並比較這些方法得出的結果。 實證結果比較，發現使用OLS所估計的係數值及圖型均較採用動態OLS及Johansen為佳。在OLS估計時，發現季節性虛擬變數對於估計有重要的影響。 由於使用OLS估計法，在變數為單根時，無法使用一般統計檢定係數顯著性，因此多加入FM-OLS和OLS實證結果的比較，以達能充分了解各變數對於通貨需求的真實影響。 採用FM-OLS實證結果，得到信用卡卡數cn、利率r和實質所得rgnp對於實質通貨淨額rcur的影響顯著，其估計值分別為-0.0025、-0.22與0.60左右。結果顯示當信用卡使用增加、利率增加與所得減少，均會減少通貨的持有，與理論相符。 在採用FM-OLS實證時，發現匯率ex對於通貨需求rcur的影響不顯著，與以往實證結果不符。推測可能原因為(1)通貨持有動機為交易目的，甚少當做投資的工具；(2)過去實證所選取的貨幣需求量為更廣義的M1A、M1B及M2。 第一章　導論 1 第一節 研究動機 1 第二節 研究方法與目的 4 第三節 研究架構 6 第二章 信用卡發展、意義及影響 7 第一節 信用卡的發展 7 第二節 信用卡的意義 9 第三節 信用卡對經濟的影響 11 第三章 貨幣需求理論與文獻回顧 14 第一節 貨幣需求理論 14 1.Baumol and Tobin存貨理論 14 2.Hafiz Akhand and Ross Milbourne 16 第二節 國外實證文獻 21 第三節 國內實證文獻 26 第四章 研究方法 33 第一節 單根檢定 33 1.ADF Test 33 2.Phillips-Perron Test 35 3.Perron Test 36 第二節 共積理論－Johansen Procedure估計法 39 第三節 動態OLS估計法 43 1.Saikkonen's OLS估計法 43 2.Phillips and Loretan's NLS估計法 44 第四節 Fully Modified OLS(FM-OLS)估計法 45 第五章 實證結果 48 第一節 資料說明與處理 48 第二節 單根檢定 50 第三節 Johansen Procedure估計法 53 第四節 OLS及動態OLS的估計及比較 63 1.OLS估計法 63 2.Saikkonen's OLS估計法 69 3.Phillips and Loretan's NLS估計法 78 4.小結 83 第五節 Johansen 與OLS估計結果比較 85 第六節 FM-OLS實證結果與OLS比較 87 第七節 本章實證結果 93 第六章 結論與建議 94

Cette thèse s'inscrit dans le domaine des problèmes inverses en traitement du signal. Elle est consacrée à la conception d'algorithmes de restauration et de séparation de signaux parcimonieux et à leur application à l'approximation de courbes de forces en microscopie de force atomique (AFM), où la notion de parcimonie est liée au nombre de points de discontinuité dans le signal (sauts, changements de pente, changements de courbure). Du point de vue méthodologique, des algorithmes sous-optimaux sont proposés pour le problème de l'approximation parcimonieuse basée sur la pseudo-norme ℓ0 : l'algorithme Single Best Replacement (SBR) est un algorithme itératif de type « ajout-retrait » inspiré d'algorithmes existants pour la restauration de signaux Bernoulli-Gaussiens. L'algorithme Continuation Single Best Replacement (CSBR) est un algorithme permettant de fournir des approximations à des degrés de parcimonie variables. Nous proposons aussi un algorithme de séparation de sources parcimonieuses à partir de mélanges avec retards, basé sur l'application préalable de l'algorithme CSBR sur chacun des mélanges, puis sur une procédure d'appariement des pics présents dans les différents mélanges. La microscopie de force atomique est une technologie récente permettant de mesurer des forces d'interaction entre nano-objets. L'analyse de courbes de forces repose sur des modèles paramétriques par morceaux. Nous proposons un algorithme permettant de détecter les régions d'intérêt (les morceaux) où chaque modèle s'applique puis d'estimer par moindres carrés les paramètres physiques (élasticité, force d'adhésion, topographie, etc.) dans chaque région. Nous proposons finalement une autre approche qui modélise une courbe de force comme un mélange de signaux sources parcimonieux retardées. La recherche des signaux sources dans une image force-volume s'effectue à partir d'un grand nombre de mélanges car il y autant de mélanges que de pixels dans l'image.

本研究以日本央行干預日圓匯率對新台幣匯率的波及效果為研究主題，選擇樣本期間為1999年1月5日至2003年12月31日之日資料做實證分析。本研究選擇日本央行干預外滙的貨幣數量、新台幣與日圓間前一日及前二日的匯率變動率、日本國定假日及日本重大經濟政策訊息為自變數，分析對因變數，也就是新台幣與日圓間匯率變動率有何影響。我們分別採用最小平方法(OLS)迴歸分析與一般化自我迴歸條件異質變異數GARCH(1,1)模型來進行分析，並發現GARCH(1,1)模型在估計上優於最小平方法(OLS) 迴歸分析；另外我們的結果也證實日本央行干預日圓與美元間匯率，在日圓外匯市場大量購買外匯，確實會對新台幣與日圓間匯率變動率造成影響，亦其的確對新台幣與日圓間匯率變動程度有波及效果。 一直以來，國內有關央行干預行為的文獻中，大多探討台灣中央銀行的干預行為，對新台幣與美元間匯率變動的影響，較少有文獻在討論當日本央行透過干預日圓外匯市場，意圖使日圓相對於美元貶值時，新台幣與日圓間匯率的變動程度會因此受到波及。而本文實證研究發現，日本央行在干預日圓匯率時，的確會對新台幣與日圓間的匯率變動率造成影響，也就是說，日本中央銀行干預日圓匯率對新台幣匯率的波及效果是存在的，這對未來在估計新台幣與日圓間匯率變動時，將更能精確的估計新台幣與日圓間匯率變動的程度。 關鍵詞:匯率干預(exchange rate intervention)、日本央行(Japan Central Bank)、 最小平方法(OLS)、一般化自我迴歸條件異質變異數(GARCH)

No description available.

The ordinary least squares (OLS) is one of the most widely used methods for modelling the functional relationship between variables. However, this estimation procedure counts on some assumptions and the violation of these assumptions may lead to nonrobust estimates. In this study, the simple linear regression model is investigated for conditions in which the distribution of the error terms is Generalised Logistic. Some robust and nonparametric methods such as modified maximum likelihood (MML), least absolute deviations (LAD), Winsorized least squares, least trimmed squares (LTS), Theil and weighted Theil are compared via computer simulation. In order to evaluate the estimator performance, mean, variance, bias, mean square error (MSE) and relative mean square error (RMSE) are computed.

In this work, we study and extend algorithms for Sparse Regression and Causal Inference problems. Both the problems are fundamental in the area of Data Science. The goal of regression problem is to nd out the \best" relationship between an output variable and input variables, given samples of the input and output values. We consider sparse regression under a high-dimensional linear model with strongly correlated variables, situations which cannot be handled well using many existing model selection algorithms. We study the performance of the popular feature selection algorithms such as LASSO, Elastic Net, BoLasso, Clustered Lasso as well as Projected Gradient Descent algorithms under this setting in terms of their running time, stability and consistency in recovering the true support. We also propose a new feature selection algorithm, BoPGD, which cluster the features rst based on their sample correlation and do subsequent sparse estimation using a bootstrapped variant of the projected gradient descent method with projection on the non-convex L0 ball. We attempt to characterize the efficiency and consistency of our algorithm by performing a host of experiments on both synthetic and real world datasets. Discovering causal relationships, beyond mere correlation, is widely recognized as a fundamental problem. The Causal Inference problems use observations to infer the underlying causal structure of the data generating process. The input to these problems is either a multivariate time series or i.i.d sequences and the output is a Feature Causal Graph where the nodes correspond to the variables and edges capture the direction of causality. For high dimensional datasets, determining the causal relationships becomes a challenging task because of the curse of dimensionality. Graphical modeling of temporal data based on the concept of \Granger Causality" has gained much attention in this context. The blend of Granger methods along with model selection techniques, such as LASSO, enables efficient discovery of a \sparse" sub-set of causal variables in high dimensional settings. However, these temporal causal methods use an input parameter, L, the maximum time lag. This parameter is the maximum gap in time between the occurrence of the output phenomenon and the causal input stimulus. How-ever, in many situations of interest, the maximum time lag is not known, and indeed, finding the range of causal e ects is an important problem. In this work, we propose and evaluate a data-driven and computationally efficient method for Granger causality inference in the Vector Auto Regressive (VAR) model without foreknowledge of the maximum time lag. We present two algorithms Lasso Granger++ and Group Lasso Granger++ which not only constructs the hypothesis feature causal graph, but also simultaneously estimates a value of maxlag (L) for each variable by balancing the trade-o between \goodness of t" and \model complexity".

O objetivo deste trabalho foi estudar a aplicação de reator anaeróbio serial (RAS) no processamento de vinhaça de cana-de-açúcar para a remoção de matéria orgânica e produção de biogás, submetido ao aumento gradativo de carregamento orgânico. Para tanto, foi utilizado um reator composto por quatro câmaras em série denominadas C1, C2, C3 e C4, com capacidade de 5 L cada, totalizando um volume de 20 L. A primeira etapa do trabalho (Etapa 1) consistiu na operação e monitoramento do reator RAS partindo-se da aplicação da carga orgânica volumétrica (COV) de 0,5 kg DQO m-3.d-1 até a obtenção da COV de 2,5 kg DQO m-3.d-1. Na segunda etapa (Etapa 2) foram operados reatores em batelada sequencial, simulando a operação contínua do reator RAS, visando estudar o efeito do aumento de taxas de carregamento orgânico aplicado. O reator RAS operou sob baixos valores de COVglobal e COALglobal aplicados na Etapa 1, obtendo-se máximas de 2,50 kg DQO m-3.d-1 e 0,167 kg STV m-3.d-1, respectivamente. Os resultados obtidos da Etapa 1 indicam satisfatória remoção global média de matéria orgânica na forma de DQO, superiores a 90 %, propiciada pela adaptação da biomassa ao longo dos reatores sequenciais. O metano apresentou-se como o gás predominante na composição do biogás produzido, com valores acima de 75 %, que confirmam o estabelecimento do sistema metanogênico pleno. Em todos os compartimentos dos reatores sequenciais verificou-se maior abundância relativa para os domínios Archaea e Bacteria os gêneros Methanosaeta, produtores de metano e, T78, responsável pela degradação de açúcares, respectivamente. Principalmente para os reatores C3 e C4, observou-se adaptação de algumas populações para ambos os domínios Bacteria e Archaea indicado pelo menor índice de diversidade e aumento da dominância. O aumento das taxas de carregamento orgânico aplicado ao lodo, estipuladas nos ensaios em batelada na Etapa 2, propiciou o aumento dos valores de carga orgânica aplicada para todos os reatores. Com o aumento do carregamento orgânico aplicado, observa-se o aumento e acúmulo de ácidos orgânicos para todas as fases dos ensaios, principalmente o acético. A produção de metano foi significativa na operação dos reatores em batelada, principalmente em C1. De forma geral, os resultados obtidos demonstraram o bom desempenho do processamento da vinhaça de cana-de açúcar em reator compartimentado tipo RAS para a remoção de matéria orgânica e produção de biogás. / The aim of this study was to investigate the application of an anaerobic serial reactor (ASR) in the sugarcane stillage processing for the removal of organic matter and biogas production, submitted to the gradual increase in organic loading rate. ASR reactor consisted of four reactors operated in series (C1, C2, C3 and C4), each with a reaction volume of 5 L (total volume of 20 L). The first step (Step 1) consisted of reactor operation and monitoring; organic loading rate was gradually increased from 0.5 kg COD m-3d-1 up to 2.5 kg COD m-3 d-1. In the second step (Step 2), sequential batch reactors were operated, simulating the continuous ASR reactor operation, aiming to study the effect of the increase of organic load applied to sludge (OLS). The ASR reactor operated under low values of COVglobal and COALglobal applied in Step 1, obtaining a maximum 2.50 kg COD m-3d-1 and 0.167 kg TVS/m-3d-1, respectively. The results obtained from Step 1 indicated satisfactory removal of the global average organic matter in the form of COD more than 90 %, caused by adaptation of the biomass along the sequential reactors. Methane appeared to be the predominant gas in biogas composition, with values above 75%, which confirmed the establishment of the full methanogenic system. In all compartments of the sequential reactors, there was a higher relative abundance for the domains Archaea and Bacteria of the genera Methanosaeta, methane producers, and T78, which is responsible for degradation of sugars, respectively. Especially for the reactors C3 and C4, the adaptation of some populations for both domains Bacteria and Archaea indicated was observed by the lower diversity index and increase in dominance. The increase of the organic loading rates applied to the sludge, as established in the batch tests, increased the organic load values applied to all reactors. With the increase of organic loading applied, an increase in the accumulation of organic acids, mainly acetic, for all stages and the batch tests. Methane production was significant in the operation of batch reactors, mainly in the C1. In general, the results obtained showed good performance of the processing of the sugarcane stillage in serial reactor (ASR) for the removal of organic matter and biogas production.

MCom / Department of Accountany / The relationship between the Equity Market Risk Premium (MRP) and macroeconomic variables has been a subject of extensive discussion in the finance literature. The MRP is a central component of the main asset pricing models which are used to estimate the cost of equity which is mainly used in investment appraisal, performance measurement and valuation of equity assets. Past studies have identified inflation rate, interest rate, foreign exchange rate and political risk as the key macroeconomic variables that determine the size of the MRP. The test of the impact of these variables on the MRP have however been based mainly on data from developed countries and a few emerging countries. To the researcher’s knowledge, there are no studies that have investigated the impact of these macroeconomic variables on the MRP in South Africa. It is necessary to test the impact of these variables in the context of South Africa as these variables vary across countries. Using time series secondary data that was obtained from the SARB database, JSE database and World Bank database for the period 2002 to 2017, this study investigated the impact of these variables on the MRP in South Africa. A total of 192 observations per series of the inflation rate, interest rate, foreign exchange rate, political risk, JSE-ALSI and 91-days Treasury bill was used in the study. The data used were tested for possible misspecification errors that could arise from using a time series secondary data and the regression model was fitted using the Ordinary Least Square (OLS) estimator. The misspecification tests and models were both implemented on STATA 15 software. The results shows that inflation rate, interest rate and foreign exchange rate have a negative impact on the MRP whilst political risk has a positive impact on the MRP. Furthermore, the result shows that the inflation rate is the only variable amongst other variable tested that has a significant influence on the MRP for the study period. The study, therefore, concludes that inflation rate has the highest impact on the MRP in the context of South Africa. The study recommends that inflation rate should be monitored and kept within its target of 3-6% amongst other variables tested in order to increase investors’ confidence in the security market and also foster economic growth. The main limitations to the study were the limited data sources and insufficient funds. / NRF

Abstracts Does post-industrialized countries face a second Kuznets curve with the IT revolution? By: Rickard Sandqvist, Supervisor: Ulf Jansson, Urban and regional planning, advanced level, Master thesis for master exam in Urban and regional planning, 30 ECTS credits, Language: English, Key words: recent income inequality trends, the creative class, kuznets hypothesis and theories of income inequality, income inequality with the IT revolution in post-industrialized countries, SURE estimation, OLS estimation, income inequality measures.  The aim of the thesis is to analyze the development of income inequality for countries that have longer time series available. I use theories and hypothesises of income inequality and do some development of Kuznets hypothesis. My research question is: does post-industrialized countries face a second Kuznets curve with the IT revolution? I use OLS (ordinary least square) and SURE (seemingly unrelated regression equation) estimations to answer my research question. I use data from the world bank and I analyze data from mainly countries in europe and north and south america. My results is that there is evidence that post industrialised countries face a second Kuznets curve and I do the conclusion that this is depending on the IT revolution.

本研究分為兩個部分，第一個部分為探討1990年來英美等國央行貨幣政策操作改革方向，期望貨幣市場的金融同業隔夜拆款利率，沿著隔夜拆款目標利率微幅波動。 英國央行原採零準備率制度，不易估測貨幣市場資金，其隔夜拆款利率波動幅度較美國為劇，為有效控制操作目標， 2006年5月起實施「自願準備金制度」，有利英國央行進行公開市場操作，達成穩定利率的效果。 第二部分參考Nadja(2006)一文，探討我國央行貨幣政策操作對短期利率之影響，係以隔夜拆款利率與目標利率的利差為利率函數模型之因變數，其中以重貼現率為目標利率，並以超額準備為主要的操作變數。 本文以最小平方估計法（OLS）實證結果發現，央行貨幣政策操作有效地影響隔夜拆款利率；惟2003年起央行不以重貼現率為隔夜拆款利率的底限，貨幣政策操作對隔夜拆款利率與重貼現率之間的利率變動並無顯著性的影響，亦即央行已放棄重貼現率為隔夜拆款利率之目標利率，而係積極地進行之公開市場操作，穩定短期利率。