本文選題：短期利率 + 參數(shù)估計(jì)　； 參考：《西南財(cái)經(jīng)大學(xué)》2014年碩士論文

【摘要】：利率是金融市場(chǎng)中的核心變量。近年來,隨著中國利率市場(chǎng)化進(jìn)程的不斷加快,利率研究的理論和實(shí)用價(jià)值日益凸顯,本文以短期利率作為研究對(duì)象,探討了單因素短期利率模型參數(shù)估計(jì)的方法,提出一種基于廣義自助法的偏差修正方法。最后以三個(gè)月拆借期限的Shibor為樣本進(jìn)行實(shí)證分析。 對(duì)于參數(shù)估計(jì),本文對(duì)比分析了歐拉似然法和模擬似然法(SMLE),這兩種方法都屬于偽似然估計(jì)方法,其中,模擬似然法(SMLE)是歐拉似然法和蒙托卡羅方法的結(jié)合,具有更好的統(tǒng)計(jì)性質(zhì),但是計(jì)算復(fù)雜,歐拉似然方法則更簡單和易于計(jì)算。 當(dāng)樣本量不夠大時(shí),兩種估計(jì)方法都存在較大的偏差。因此,我們提出一種新的基于廣義Bootstrap的偏差修正方法。其基本思想是對(duì)似然函數(shù)賦予隨機(jī)權(quán)重,構(gòu)造加權(quán)似然函數(shù),得到基于加權(quán)似然函數(shù)的新的參數(shù)估計(jì)值,以此估計(jì)出參數(shù)估計(jì)的偏差,最后進(jìn)行偏差校正。通過隨機(jī)模擬實(shí)驗(yàn),我們發(fā)現(xiàn)這種方法在一定程度上能夠降低歐拉似然估計(jì)的偏差。 此外,本文對(duì)Shibor利率進(jìn)行實(shí)證研究,估計(jì)出了8個(gè)常見短期利率模型的參數(shù)值,并進(jìn)行了偏差校正。最后運(yùn)用CIR模型對(duì)零息債券定價(jià),分析利率期限結(jié)構(gòu),得出價(jià)格變動(dòng)曲線和收益率變動(dòng)曲線。
[Abstract]:Interest rate is the core variable in the financial market. In recent years, with the accelerating process of interest rate marketization in China, the theoretical and practical value of interest rate research has become increasingly prominent. This paper takes short-term interest rate as the object of study, and discusses the method of parameter estimation of single-factor short-term interest rate model. A deviation correction method based on generalized self-help method is proposed. Finally, the three-month loan term of Shibor as a sample for empirical analysis. For parameter estimation, Euler likelihood method (Euler likelihood method) and simulated likelihood method (SMLE), both of which belong to pseudo likelihood estimation method, are compared and analyzed in this paper. Among them, simulated likelihood method (SMLE) is a combination of Euler likelihood method and Monto Carlo method. It has better statistical properties, but the calculation is complicated, and the Euler likelihood method is simpler and easier to calculate. When the sample size is not large enough, there is a big deviation between the two estimation methods. Therefore, we propose a new bias correction method based on generalized bootstrap. The basic idea is to give random weight to likelihood function, construct weighted likelihood function, obtain new parameter estimation value based on weighted likelihood function, estimate the deviation of parameter estimation, and correct the deviation finally. We find that this method can reduce the deviation of Euler likelihood estimation to some extent. In addition, this paper makes an empirical study on Shibor interest rate, estimates the parameter values of eight common short-term interest rate models, and corrects the deviation. Finally, the CIR model is used to price zero interest bond, and the term structure of interest rate is analyzed, and the price change curve and yield change curve are obtained.
【學(xué)位授予單位】：西南財(cái)經(jīng)大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2014
【分類號(hào)】：F224;F832.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 林海,鄭振龍;中國利率動(dòng)態(tài)模型研究[J];財(cái)經(jīng)問題研究;2005年09期

2 胡瑾瑾;陳淼W，

本文編號(hào)：2099975