A nonlinear cointegration model for US bond Armillotta E. A nonlinear cointegration model for US bond Introduction Aim of the thesis Literature Emanuele Armillotta review Linear Model Alternative UNIVERSIT` A POLITECNICA DELLE MARCHE Models Nonlinear Model Department of Economics and Social Sciences Preliminary Ph.D. in Economics XV cycle analysis Data set Out-of-sample October 2015 analysis Work in progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 1 / 16
A nonlinear cointegration Introduction 1 model for US bond Armillotta E. Aim of the thesis 2 Introduction Literature review 3 Aim of the thesis Linear Model Literature Alternative Models review Nonlinear Model Linear Model Alternative Models Nonlinear Model Preliminary analysis 4 Preliminary analysis Data set Data set Out-of-sample analysis Out-of-sample analysis Work in Work in progress progress 5 Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 2 / 16
Nonlinear model A nonlinear cointegration model for US bond Recently a stand of literature has been interested in the use Armillotta E. of nonlinear model in different macroeconomic areas Introduction Aim of the Power purchasing parity: thesis Literature Lo e Zivot(2001), Taylor e Siklos(2001), Kapetanios e review Shin(2002), Lundberg e Ter¨ asvirta(2006), Nam(2011), Linear Model Alternative Models Beckmann(2012) Nonlinear Model Preliminary Employment rate: analysis Skalin e Ter¨ asvirta(2002), Caner e Hansen(2003), Data set Out-of-sample analysis Akram(2005), P´ erez-Alonso e Di Sanzo(2010) Work in progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 3 / 16
Aim of the thesis A nonlinear cointegration model for US bond Armillotta E. 1 Could nonlinear adjustment mechanism capture Introduction macroeconomic dynamics in the term of structure? Aim of the thesis 2 Could the nonlinear framework represent in a stylised way Literature review the American monetary policy? Linear Model Alternative Models Nonlinear Model 3 Which approximation function is best suited for this Preliminary analysis purpose? Data set Out-of-sample analysis Work in progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 4 / 16
Linear Model A nonlinear cointegration model for US Campbell e Shiller(1987) were the first who used a bond cointegration model (VECM) in which the spread represents a Armillotta E. linear stationary combination of two processes I(1) Introduction Aim of the thesis Problems Literature 1 Spreads aren’t stationary because of: review Linear Model time-varying risk premium Alternative Models failure rational expectations hypothesis Nonlinear Model Preliminary correlation with macro factors analysis 2 Tests reject the hypothesis that there is only one common Data set Out-of-sample analysis trend Work in 3 Restrictions imposed by linear model progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 5 / 16
Linear Model A nonlinear cointegration model for US Campbell e Shiller(1987) were the first who used a bond cointegration model (VECM) in which the spread represents a Armillotta E. linear stationary combination of two processes I(1) Introduction Aim of the thesis Problems Literature 1 Spreads aren’t stationary because of: review Linear Model time-varying risk premium Alternative Models failure rational expectations hypothesis Nonlinear Model Preliminary correlation with macro factors analysis 2 Tests reject the hypothesis that there is only one common Data set Out-of-sample analysis trend Work in 3 Restrictions imposed by linear model progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 5 / 16
Linear Model A nonlinear cointegration model for US Campbell e Shiller(1987) were the first who used a bond cointegration model (VECM) in which the spread represents a Armillotta E. linear stationary combination of two processes I(1) Introduction Aim of the thesis Problems Literature 1 Spreads aren’t stationary because of: review Linear Model time-varying risk premium Alternative Models failure rational expectations hypothesis Nonlinear Model Preliminary correlation with macro factors analysis 2 Tests reject the hypothesis that there is only one common Data set Out-of-sample analysis trend Work in 3 Restrictions imposed by linear model progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 5 / 16
Linear Model A nonlinear cointegration model for US Campbell e Shiller(1987) were the first who used a bond cointegration model (VECM) in which the spread represents a Armillotta E. linear stationary combination of two processes I(1) Introduction Aim of the thesis Problems Literature 1 Spreads aren’t stationary because of: review Linear Model time-varying risk premium Alternative Models failure rational expectations hypothesis Nonlinear Model Preliminary correlation with macro factors analysis 2 Tests reject the hypothesis that there is only one common Data set Out-of-sample analysis trend Work in 3 Restrictions imposed by linear model progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 5 / 16
Linear Model A nonlinear cointegration model for US Campbell e Shiller(1987) were the first who used a bond cointegration model (VECM) in which the spread represents a Armillotta E. linear stationary combination of two processes I(1) Introduction Aim of the thesis Problems Literature 1 Spreads aren’t stationary because of: review Linear Model time-varying risk premium Alternative Models failure rational expectations hypothesis Nonlinear Model Preliminary correlation with macro factors analysis 2 Tests reject the hypothesis that there is only one common Data set Out-of-sample analysis trend Work in 3 Restrictions imposed by linear model progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 5 / 16
Linear Model A nonlinear cointegration model for US Campbell e Shiller(1987) were the first who used a bond cointegration model (VECM) in which the spread represents a Armillotta E. linear stationary combination of two processes I(1) Introduction Aim of the thesis Problems Literature 1 Spreads aren’t stationary because of: review Linear Model time-varying risk premium Alternative Models failure rational expectations hypothesis Nonlinear Model Preliminary correlation with macro factors analysis 2 Tests reject the hypothesis that there is only one common Data set Out-of-sample analysis trend Work in 3 Restrictions imposed by linear model progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 5 / 16
Linear Model A nonlinear cointegration model for US Campbell e Shiller(1987) were the first who used a bond cointegration model (VECM) in which the spread represents a Armillotta E. linear stationary combination of two processes I(1) Introduction Aim of the thesis Problems Literature 1 Spreads aren’t stationary because of: review Linear Model time-varying risk premium Alternative Models failure rational expectations hypothesis Nonlinear Model Preliminary correlation with macro factors analysis 2 Tests reject the hypothesis that there is only one common Data set Out-of-sample analysis trend Work in 3 Restrictions imposed by linear model progress Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 5 / 16
Alternative Models A nonlinear cointegration Literature has tried to verify empirically the expectations model for US bond hypothesis through different methods: Armillotta E. Introduction 1 By including exogenous macroeconomic variables (inflation Aim of the and real activity indicator) thesis Ang e Piazzesi (2003), Carriero et al.(2004), Valente et Literature review al.(2004) Linear Model Alternative 2 By empirical analysis on yields with shorter maturity (daily Models Nonlinear Model or weekly maturity) Preliminary analysis Longstaff(2000), Della Corte et al.(2007) Data set Out-of-sample 3 By using nonlinear models analysis Work in Blake e Fomby(1992), Enders e Granger(1998), Hansen e progress Seo(2002) Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 6 / 16
Alternative Models A nonlinear cointegration Literature has tried to verify empirically the expectations model for US bond hypothesis through different methods: Armillotta E. Introduction 1 By including exogenous macroeconomic variables (inflation Aim of the and real activity indicator) thesis Ang e Piazzesi (2003), Carriero et al.(2004), Valente et Literature review al.(2004) Linear Model Alternative 2 By empirical analysis on yields with shorter maturity (daily Models Nonlinear Model or weekly maturity) Preliminary analysis Longstaff(2000), Della Corte et al.(2007) Data set Out-of-sample 3 By using nonlinear models analysis Work in Blake e Fomby(1992), Enders e Granger(1998), Hansen e progress Seo(2002) Armillotta E. (UNIVPM) A nonlinear cointegration model for US bond October 2015 6 / 16
Recommend
More recommend
