[pt] Este artigo investiga a natureza das assimetrias de informação no mercado brasileiro de empréstimos consignados de 2013 a 2021. Desenvolvemos um modelo de demanda que leva em consideração o efeito da informação privada dos consumidores nas decisões de empréstimo. A novidade do modelo é sua capacidade de extrair informações sobre características não observáveis usando dados públicos a nível dos bancos. Empiricamente, utilizamos a variação das participações de mercado e das taxas de inadimplência dos bancos para estimar um parâmetro de utilidade que representa o sinal de seleção no mercado. Nossa análise revela evidências empíricas de seleção vantajosa dentro do mercado, indicando que os tomadores de empréstimos mais seguros estão mais inclinados a solicitar empréstimos. Além disso, expandimos o modelo para incorporar um parâmetro distinto para a Caixa Econômica, um banco estatal que exibiu comportamento diferente em comparação com outras instituições financeiras durante o mesmo período. Nossa análise revela um parâmetro de seleção significativamente menor em magnitude para a Caixa. / [en] This paper investigates the nature of information asymmetries in the Brazilian payroll loan market from 2013 to 2021. We develop a demand model that accounts for the effect of consumers private information on borrowing decisions. The novelty of the model is its ability to extract information on unobservable characteristics using publicly available firm-level data. Empirically, we use the variation of market shares and default rates within banks to estimate a utility parameter that represents the sign of selection in the market. Our analysis reveals empirical evidence of advantageous selection within the market, indicating that safer borrowers are more inclined to apply for loans. Additionally, we expand the model to incorporate a distinct parameter for Caixa Econômica, a state-owned bank that exhibited different behavior compared to other financial institutions during the same period. Our analysis reveals a significantly lower selection parameter in magnitude for Caixa.

No description available.

No description available.

The problem of determining the probability of default or credit risk for companies is crucial when providing financial services. This problem is often modeled based on snapshot data that does not take the time dimension into account. Instead, we approach the problem with enterprise resource planning data in time series. With the added complexity the time series introduce, we pose that deep learning models could be suitable for the task. A comparison of a fully convolutional network and a transformer encoder was made to the current state-of-the-art model for the probability of default problem, XGBoost. The comparison showed that XGBoost generalized very well to the time series domain, even well enough to beat the deep learning models across all evaluation metrics. Furthermore, time series data with monthly, quarterly and yearly timestamps over three years was tested. Also, public features that could be extracted from quarterly and annual financial reports were compared with internal enterprise resource planning data. We found that the introduction of time series to the problem improves the performance and that models based on internal data outperform the ones based on public data. To be more precise, we argue that the dataset being based on small to medium-sized companies lessens the impact of highly granular data, and makes the selection of what features to include more prominent. This is something XGBoost takes advantage of in a very efficient way, especially when extracting features that capture the behavior of the time series, causing it to beat the deep learning competitors even though it does not pick up on the sequential aspect of the data.

The aim of this thesis is to analyse the potential effects of the planned Brenner Pass opening in 2028. The thesis is divided into 3 studies: in the first, we will provide a methodological description of possible approaches to be used in this type of analysis. In the second, we will propose an experiment to understand if it is possible to exploit the default effect to induce a modal shift towards the most sustainable means of transportation: the train. In the third study, we will conduct a simulation by applying the model constructed by Monte et al in 2018. The simulation results demonstrate that the opening of the Brenner Base Tunnel will lead to an increase in welfa re in the Trentino Alto Adige region (+0.2%), along with other socio economic changes such as increased commuting and housing costs. Theresults of the second study highlight the potential of a simple policy based on the default effect, pushing over half ( 54%) of the participants to repeatedly choose the most sustainable option.

This thesis investigates correlations and linkages in credit risk that widely exist in all sectors of the financial markets. The main body of this thesis is constructed around four empirical chapters. I started with extending two main issues focused by earlier empirical studies on credit derivatives markets: the determinants of CDS spreads and the relationship between CDS spreads and bond yield spreads, with a special focus on the effect of the subprime crisis. By having observed that the linear relationship can not fully explain the variation in CDS spreads, the third empirical chapter investigated the dependence structure between CDS spread changes and market variables using a nonlinear copula method. The last chapter investigated the relationship between the CDS spread and another credit spread - the TED spread, in that a MVGARCH model and twelve copulas are set forth including three time varying copulas. The results of this thesis greatly enhanced our understanding about the effect of the subprime crisis on the credit default swap market, upon which a set of useful practical suggestions are made to policy makers and market participants.

過去國內之抵押貸款提前清償與逾期還款之相關研究，在實證研究上最主要利用邏輯斯迴歸或是比例轉機模型( Proportional hazard model )分析影響一般住宅抵押貸款人提前清償與逾期還款之因素，並估計一般住宅抵押貸款人提前清償之機率。本文選擇採用研究抵押貸款時，國內未曾使用之卜瓦松迴歸( Poisson regression model )來估計比例轉機模型假設下影響提前清償與違約變數之參數，以研究影響抵押貸款借款人之提前償還與違約因素。 本研究結合比例轉機模型與卜瓦松迴歸模型，目的在結合兩模型之優點，在處理時間相依之共變數效率提高，並且在處理多重時間尺度的方程式較偏最大概似估計法直接，以得到較佳的研究成果。另外，過去國內提前清償與違約之文獻中並未加入利率走勢之變數，本研究加入再融資利率對31∼90天期商業本票利率之比率與再融資利率波動性兩變數，以考慮利率走勢對貸款者提前清償及違約行為之影響。 模型中的解釋變數包括地區、季節、抵押貸款年齡、貸款成數、貸款人年齡、性別、婚姻狀況、教育程度、職業、屋齡、房屋坪數、所得、貸款金額、月付額對薪資比、再融資利率/31∼90天期商業本票利率、再融資利率波動性等十六項。實證結果在提前清償部份，顯著正向之變數有貸款年齡、屋齡、房屋坪數、所得、月付額與薪資比，顯著負向之變數包括季節、再融資利率對31∼90天期商業本票利率之比率、貸款金額。在違約部份，顯著正向之變數包括貸款年齡、貸款成數、年齡、所得、月付額與薪資比、再融資利率對31∼90天期商業本票利率之比率；顯著負向之變數包括季節、教育程度及貸款金額。

金融機構在承做住宅抵押貸款時，面臨兩種風險，分別是提前清償及違約。這兩種借款人風險行為對金融機構的資產管理產生相當大的影響，尤其在不動產證券化的推動上，都是評價證券價格的關鍵因子，因此，討論借款者提前還款與違約行為，是近年不動產證券化領域中重大議題。 借款人決定提前清償及違約與否，除了與借款人特性之外尚有房屋特性與財務選擇上的特性，且有許多影響因子並非維持在貸款起始的狀態，而是會在貸款存續期間內隨著時間遷移有所改變，因此，本文在進一步研究影響借款人行為時，處理時間相依變數，利用動態調整過後的變數來分析借款人提前清償及違約風險行為，觀察借款人特徵、房屋型態及貸款條件等變數與借款人風險行為的關係並進行證券價格MBS的評價。 實證結果顯示，借款人特徵部分並不會影響提前清償但會影響違約風險。且借款者在財務選擇上面，會有落後反映的現象，亦即隨著時間的經過，借款者才會選擇有利於自己的財務決策，且雖然本國貸款為浮動利率貸款，但是影響借款者最深的仍然是利率相關因素，且觀察到借款者對於財務上面的好處比壞處有更敏感的現象，顯示出借款者比較趨近於風險愛好者。 在MBS評價上發現，影響價格的最重要因素在於放款的品質，因此，要真正落實抵押住宅證券化的實行，關鍵在於金融機構必須篩選優良放款進行證券化，才能吸引投資人，增加住宅抵押貸款證券化發行的可行性。

房屋抵押貸款的評價，因為需考慮到貸款人的提前清償及違約風險造成現金流量之不確定性，決定房屋抵押貸款的價格比決定一般違約證券的價格更具困難度。因此，如何合理評估抵押貸款證券的價值實為一值得深入探討之課題，本文即針對此議題進行研究。傳統文獻在進行房屋抵押貸款的評價方法，主要可區分為兩種：結構模型(structural-form approach)及縮減式模型(reduced-form approach)。目前的文獻上，其評價的封閉解只存在於結構式模型，但在此模型下的評價，存在著違約與提前清償條件的設定問題，這將對評價的準確性造成很大的影響，在實務的應用上有一定的限制。再者，結構式模型在處理多變數且變數間具相關性的情況，存在一定的複雜性與困難度，而縮減式模型在此情況的處理上是較容易的。本研究將從縮減式模型的角度，引入 Jarrow (2001)的概念，在包含多重變數並考慮變數間相關係數之縮減模型下，進行房屋抵押貸款封閉解的推導。透過此方法可協助資產管理者從事投資組合配置最適化與避險策略的分析，亦期望能提供實務界一個更具可行性與效率性之房屋抵押貸款評價模型。 / Valuing mortgage-related securities is more complicated than valuing regular defaultable claims due to the borrower’s prepayment behavior as well as the possibility of default. In general, the methods that are applied to investigate mortgage value and termination risk can be divided into two categories: a structural-form approach and a reduced-form approach. Some researchers use a structural-form model to obtain the closed-form formulae for the mortgage value. With this method, however, it is difficult to identify the critical region of early exercise and deal with the situation including multivariable and their correlations correlation among variables. As an alternative, the reduced-form model developed in this study is able to value the mortgage without setting boundary conditions, and can thereby accurately handle the multi-dimensional space of correlated state variables. This study extends Jarrow’s (2001) model to examine mortgage valuations. The purpose of this article is to derive a closed-form solution of the mortgage valuation equation under a general reduced-form model that embeds relevant economic variables. This new approach enables portfolio managers to undertake sophisticated portfolio optimization and hedging analyses, and makes it possible to more accurately and efficiently value the complicated mortgage.

本研究以還款績效的觀點，分析小型信用貸款中申貸者的特性，讓銀行放款的依據除了判斷正常戶與否之外，進一步以還款績效與風險區分出不同群組的申貸者，以期作不同的放款策略；同時將個人基本變數 、該銀行內徵信資料以及聯合徵信資料變數 作統計性分類，篩選出代表性因素，研究這些因素如何影響各還款績效群組。 研究發現，申貸者可以區分為「還款能力平穩—逾期風險低」、「還款能力優良—逾期風險中」、「還款能力低下—逾期風險高」這三群。而從影響各群組的因素中可以瞭解，「還款能力平穩—逾期風險低」群組，多為各方面信用持平良好的申貸者；「還款能力優良—逾期風險中」群組，多為具有理財管理特質、財務狀況良好的申貸者；「還款能力低下—逾期風險高」群組，多為具有債務因素、信用表現不佳、申貸動作頻繁的申貸者。 在放款利潤與風險方面，對三個群組之申貸戶分別採用不同方法放款，可以作到讓銀行對較少申貸戶放款並且可提升利潤並且改善損失。進行多元羅吉斯迴歸模型分析可以發掘出具影響力的因素，針對這些因素來進行分群後並採差異化放款方法，也可以作到對較少申貸戶放款並且能提升利潤以及降低損失的效果。由於因素代表具解釋性變數的歸納，配合這些具預測機能的因素及變數分群訂定差異化授信政策，有助於防範風險於未然。 / This research analyses the characteristics of small-scale credit loan applicants on the persepective of repay performances，allowing the banks not only to discriminate between good and bad applicants but also to establish different lending tatics for applicants of different repay performance groups。We also analyse the personal characteristics and joint credit informantion of these applicants to sieve out the representative factors，and study how these factors affect the repay performance groups。 Our research discovers that the applicants can be discriminanted into three groups：「low but steady repay ability—low overdue loss」、「good repay ability— acceptable overdue loss」、「very low repay ability—high overdue loss」。We can learn from those factors，that most applicants grouped as 「low but steady repay ability— low overdue loss」also have good credit qualities in other aspect；applicants grouped as 「good repay ability—acceptable overdue loss」 have finance management concept and good financial condition；applicants grouped as 「very low repay ability—high overdue loss」have debt burdens and bad credit qualities。 As for the revenues and riks，we can improve the profit and loss with fewer applicants by taking differenct lending policies to those three groups。By using multinomial logistic regression，we can discover those factors who has significant effects and use these factors to cluster applicants into groups and to adopt different lending policies for these groups。Because those factors represent the induction of the variables which can explain the applicants’ behaviors，we can somehow prevent the risks by establishing different policies with the coordination of these factors and clusters。