Numerical issues in threshold autoregressive modeling of time series

This paper analyses the contribution of various numerical approaches to making the estimation of threshold autoregressive time series more efficient. It relies on the computational advantages of QR factorizations and proposes Givens transformations to update these factors for sequential LS problems. By showing that the residual sum of squares is a continuous rational function over threshold intervals it develops a new fitting method based on rational interpolation and the standard necessary optimality condition. Taking as benchmark a simple grid search, the paper illustrates via Monte Carlo simulations the efficiency gains of the proposed tools.     

关于复数运算在高次方程中的运用

随着office2010版开发,复数函数的运算的完善为高次方程的完整解提供了技术上的有力保障。     

Volatility Estimation with Price Quanta

Volatility estimators based on high, low, opening and closing prices have been developed, and perform well on simulated data, but on real data they frequently give lower values for volatility than the simple open–close estimator. This may be due to the fact that for real data, the maximum (or minimum) price is often at the beginning or end of the day. While this could not happen if the observed process was log Brownian, it could happen if the observed process were log Brownian, but observed only to the nearest penny. We develop the theory of such approximations to derive the corrected versions of the basic estimators.     

Sunspots and Hopf bifurcations in continuous time endogenous growth models

First we treat a three-dimensional continuous time abstract stationary model that includes one predetermined variable and two non-predetermined variables. We construct stationary sunspot equilibria in this model under the following two alternative conditions: (i) a steady state has two stable roots and one unstable root; and (ii) A closed orbit has a two-dimensional manifold on which it is asymptotically stable. Next, we apply these results to the models due to Lucas and Romer that undergo Hopf bifurcations for some parameter values. We construct sunspot equilibria in these models.     

Distributive cycles and endogenous technical change in a BoPC growth model

Our purpose in this paper is to expand Goodwin's (1967) distributive cycle model to an open economy framework in a way that incorporates the balance-of-payments constraint on growth. We do so by allowing technical change to be endogenous to the cyclical dynamics of the system and by adopting an independent investment function. We show that a Hopf-Bifurcation analysis establishes the possibility of persistent and bounded cyclical paths both for a 3D and a 4D extension of the model. Some numerical simulations are performed based on the analytical models developed. Motivational empirical evidence is also provided for Thirlwall's law using a sample of 16 OECD countries.     

基于因子分解的代理多重盲签名方案

代理多重签名是指两个以上的原始签名者将他们的签名权同时指派给一个代理签名者,代理签名者代表多个原始签名者实现他们的多重签名。盲签名是指签名人不知道所签消息的具体内容。文章结合代理多重签名和盲签名的优点,提出了一种新的签名方案,并对其有效性进行验证,最后指出进一步的研究方向。     

On perpetual American put valuation and first-passage in a regime-switching model with jumps

In this paper we consider the problem of pricing a perpetual American put option in an exponential regime-switching Lévy model. For the case of the (dense) class of phase-type jumps and finitely many regimes we derive an explicit expression for the value function. The solution of the corresponding first-passage problem under a state-dependent level rests on a path transformation and a new matrix Wiener–Hopf factorization result for this class of processes. Research supported by the Nuffield Foundation, grant NAL/00761/G, and EPSRC grant EP/D039053/1.     

基于logistic回归的上市公司财务危机预警模型

国内外所有关于财务危机预警问题的研究,主要涉及到两个方面的问题：一是财务危机概念的界定;二是预测变量（判别指标）的选择和预警模型的建立（企业财务危机是一个连续的动态过程,并直接表现为绩效指标的恶化,因此,可通过一定的财务指标来构造企业的财务预警模型）。运用中国上市公司的财务指标数据及因子分析和Logistic回归等方法构建基于上市公司的财务危机预警的Logistic模型,经过检验,具有一定的实际应用价值。     

The informative sample size for dynamic multiple equation systems with moving average errors

Abstract  The problem considered here, is that of finding suitable conditions for dynamic economic systems that exclude the existence of observationally equivalent structures. Here observational equivalence refers to equality of distributions or first and second moments of a small finite sample from the observable process. It is shown, that under these conditions we may act as if the lagged endogenous variables are nonrandom exogenous variables, when global identifiability is investigated.     

Neoclassical growth model with two fixed delays

“Delay” has been considered as one of destabilizing factors in economic dynamics for a long time. Dynamic macroeconomics is concerned with explaining growth and fluctuations. This paper shows how various dynamics involving cyclic fluctuations can emerge in the standard neoclassical growth model when two distinct delays, a delay in production and a delay in depreciation, are explicitly taken into account. We first confirm that the production delay has a stabilizing effect and the depreciation delay has a destabilizing effect in a one‐delay model. We then determine the stability switching curve analytically in the two delay model. It is shown that cyclic fluctuations emerge via Hopf bifurcation when stability is lost. It is also found that stability loss and gain repeatedly occur. Numerical examples verify the theoretical results when the Cobb‐Douglas production function is adopted.