No / Motivated by an extensive literature showing that government bond yields exhibit a strong non-Markov property, in the sense that moving averages of long-lagged yields significantly improve the predictability of excess bond returns. We then develop a systematic approach of constructing non-Markov Gaussian dynamic term structure models (GDTSMs) under the Heath-Jarrow-Morton (HJM) framework. Compared to the current literature, our approach is more flexible and parsimonious, enabling us to estimate an economically significant non-Markov effect that helps predict excess bond returns both in-sample and out-of-sample.

<p>Diss. (sammanfattning) Stockholm : Handelshögskolan, 2006 [6], xiii, [1] s.: sammanfattning, s. 1-259, [5] s.: 4 uppsatser. Spikblad saknas</p>

2014年保險法第146條之4修正，增列保險業依保險法規定投資於國內證券市場上市或上櫃買賣之外幣計價股權或債券憑證之投資金額，可不計入其國外投資限額。本研究探討台灣壽險業投資外幣計價國際債券不納入國外投資限額對於台幣公債市場籌資之影響，並分析壽險業投資國際債券之贖回風險。 主要研究結果發現：(1)開放投資國際債券後，壽險業資金運用增加國外投資，但減持公債及國庫券。依據統計分析，顯示壽險業資金運用於國外投資佔比大幅增加時，除專案運用及公共投資外，其餘項目之佔比皆減少。其中台灣公債及國庫券佔比與國外投資佔比呈高度負相關。(2)壽險業對公債需求程度影響國庫籌資之成本，需求程度越低，國庫長天期籌資成本越高。透過複迴歸模型分析發現，壽險持券比(即壽險業持有公債餘額佔公債未償還餘額比例)越低，30年期公債殖利率越高。 透過本文模型，投資國際債券時，應考量可贖回國際債券相對公債之加碼、閉鎖期、國際債券再度發行之可能性與未來市場利率可能走低之幅度。以投資30年期債券為例，當可贖回國際債券相對公債之加碼減少，閉鎖期縮短，利率走跌幅度增加時，贖回風險將增加。因可贖回國際債券之高利率僅為收益率錯覺，利率走低時之贖回風險將抵銷此高利率。依據2014年至2016年債券市場資訊，本文模型評估投資人提前贖回風險為52.45bps。 / The 2014 amendment of Article 146-4 of Insurance Act extending the overseas investment ceiling to the value for foreign currency denominated listed or over-the-counter certificates of domestic stocks or bonds that are invested in by insurance enterprises in accordance with provisions of Insurance Act. This paper investigates the impact of funding in Taiwan government bond market under the overseas investment ceiling with the exclusion of international bonds investment in Taiwan life insurance industry, and analyzes the call risk of international bonds. The main results show that: (1)After the 2014 amendment of Article 146-4 of Insurance Act, foreign investments are increasing, while government bonds holdings are decreasing in investment portfolio of life insurance industry. Based on statistical analysis, as the ratio of foreign investments surging, only the ratio of authorized projects or public investment is increasing, others are falling. Especially, the ratio of government bonds and that of foreign investment are strongly negative correlation. (2) Demand of government bond of life insurance industry has impact on the funding cost of Taiwan government. The lower the demand, the higher the funding cost. Through multiple regression model, the result shows, the lower the bond holding ratio of life insurance industry, the higher the yield of Taiwan 30-year government bonds. According to the model in this paper, spread between callable international bond and government bond, lock-up period, the probability of re-issuance in international bond market, and the downtrend of interest rate should be all considered when investing in international bonds. The high yield of callable international bond is yield illusion to investors,since is largely offset by call risk. According to the model with bond market data between 2014 to 2016, the assessment of call risk is 52.45bps.

The aim of this thesis is to model the dynamics of international term structure of interest rates taking into consideration several dependence channels.Thanks to a new international Treasury yield curve database, we observe that the explained variability decision criterion, suggested by the literature, is not able to select the best combination of factors characterizing the joint dynamics of yield curves. We propose a new methodology based on the maximisation of the likelihood function of a Gaussian state-space model with common and local factors. The associated identification problem is solved in an innovative way. By estimating several sets of countries, we select two global (and three local) factors which are also useful to forecast macroeconomic variables in each considered economy.In addition, our method allows us to detect hidden factors in the international bond returns. They are not visible through a classical principal component analysis of expected bond returns but they are helpful to forecast inflation and industrial production. Keywords: International treasury yield curves, common and local factors, state-space models, EM algorithm, International bond risk premia, principal components.

Cette thèse concerne la modélisation de la dynamique des courbes des taux internationales avec prise en compte de plusieurs canaux de dépendance. A l’aide d’une nouvelle base de données des taux souverains internationaux, nous observons que le critère de la variabilité expliquée, proposé par la littérature, n’est pas capable de sélectionner une meilleure combinaison des facteurs décrivant la dynamique jointe des courbes des taux. Nous proposons une méthode nouvelle de section des facteurs fondée sur la maximisation de vraisemblance d’un modèle espace-état linéaire gaussien avec facteurs communs et locaux. Le problème d’identification associée est résolu d’une façon novatrice. En estimant différents combinaisons de pays, nous sélectionnons des deux facteurs globaux et trois locaux ayant un pouvoir prédictif des variables macro-économiques (activité économique et taux d’inflation) dans chaque économie considérée. Notre méthode nous permet aussi de détecter des facteurs cachés dans les rendements obligataires. Ils ne sont pas visibles à travers une analyse classique en composant principales des rendements obligataires et ils contribuent à la prévision du taux d’inflation et du taux de croissance de la production industrielle. / The aim of this thesis is to model the dynamics of international term structure of interest rates taking into consideration several dependence channels.Thanks to a new international Treasury yield curve database, we observe that the explained variability decision criterion, suggested by the literature, is not able to select the best combination of factors characterizing the joint dynamics of yield curves. We propose a new methodology based on the maximisation of the likelihood function of a Gaussian state-space model with common and local factors. The associated identification problem is solved in an innovative way. By estimating several sets of countries, we select two global (and three local) factors which are also useful to forecast macroeconomic variables in each considered economy.In addition, our method allows us to detect hidden factors in the international bond returns. They are not visible through a classical principal component analysis of expected bond returns but they are helpful to forecast inflation and industrial production. Keywords: International treasury yield curves, common and local factors, state-space models, EM algorithm, International bond risk premia, principal components.