Affinity Class Research Articles

Adaptive output-feedback decentralized control of a class of second order nonlinear systems using recurrent fuzzy neural networks

In this paper the design of an adaptive output-feedback decentralized control for the class of second order nonlinear affine interconnected systems based on recurrent fuzzy neural networks (RFNN) is addressed. First, a centralized control that needs the state measurements of all subsystems is designed. Then a decentralized control using the local state measurements is obtained by adding a control component aimed at compensating for the interconnections. Finally, an adaptive output-feedback decentralized control based on an RFNN is designed. In design of such controller, no separated state estimator is needed, since the controller dynamics is embedded in the recurrent network. Practical tracking is established by invoking Lyapunov stability analysis. Simulation and experimental results are presented to evaluate the performance of the proposed control law.

Neo-Lymphoid Aggregates in the Adult Liver Can Initiate Potent Cell-Mediated Immunity

Author SummaryLymph nodes (LNs) are believed to be the most important tissues initiating immune responses by facilitating the activation of T and B lymphocytes. Mice lacking such LNs (called alymphoplastic) are severely immune compromised and resistant to immunizations. We discovered that the immune-deficiency of such alymphoplastic mice is actually not caused by the loss of LNs, but rather by the underlying genetic lesion. Surprisingly, mice lacking all lymph nodes can still mount potent T cell-mediated immune responses. We also discovered that T and B cells have completely different structural requirements for their activation/maturation. Whereas B cells rely on LNs to become efficient antibody-producing cells, T cells can be activated successfully outside of such dedicated tissues. So—in the absence of LNs—antigens delivered by immunization are actively transported into the liver where cellular immunity is initiated. The mammalian fetal liver is responsible for the early formation of blood and immune cells, and we propose that the adult liver can still provide a niche for T cell–antigen encounters. During evolution, T and B cells emerged simultaneously, allowing cold-blooded vertebrates (which lack LNs) to launch adaptive immune responses. The development of LNs in mammals coincided with a drastic improvement in antibody affinity maturation, whereas T cells remain LN-independent to this day.

The LIBOR Market Model: A Critical Review

This paper presents a critical review of the different versions of the LIBOR market model (LMM). Based on the new taxonomy of the term structure models (see Nawalkha, Beliaeva, and Soto [2007a, 2007b]) the typical application of the LMM are shown to triple-plus type, exposing these to the dangers of “smoothing.” This paper also derives a double-plus version of Jarrow, Li, and Zhao’s (JLZ) [2007] LMM model with stochastic volatility and jumps, which is less exposed to the dangers of “smoothing” compared with the original triple-plus version of this model. Finally, this paper makes a persuasive case for considering high-dimensional double-plus term structure models in the affine and quadratic classes. Fast computational methods make these models powerful alternatives to the LMM for valuing and hedging interest rate derivatives.

Risk Premiums of Excess Bond Returns Have a Unit Root

We develop a term structure model that can match two stylized facts of excess returns on long-term bonds. The first stylized fact is the predictability of excess returns, which requires sufficient volatility in the price of risk. The second stylized fact is that yields are dominated by a level factor, which requires persistence in the spot interest rate or in the price of risk, or both. We calibrate a stochastic discount factor that is consistent with both stylized facts. Doing so we need time series processes that are outside the essentially affine class of term structure models. For parsimony we choose a fractionally integrated process, which leads to tractable analytical solutions. As an important implication we find that risk premiums of excess bond returns are very persistent.

GANP suppresses the arginine methyltransferase PRMT5 regulating IL-4-mediated STAT6-signaling to IgE production in B cells

Antigen (Ag)-driven B cells undergo antibody (Ab) affinity maturation and class switching in germinal center (GC) B cells. GANP is one of the molecules required for Ab affinity maturation. We herein found an increase of IgE in B cell ganp-deficient mice and studied the signal transduction pathway regulated by GANP. GANP suppresses the STAT-mediated transcription activity in GC B cells with the regulation of arginine methyltransferase activity by the interaction with JAK-binding protein arginine methyltransferase (PRMT) 5 and JAK1/JAK3 that are responsible for STAT6 activation. The prmt5 mRNA was up-regulated in B cells after stimulation in vitro and in vivo in GC B cells. The loss of GANP caused up-regulation of phosphorylation and arginine dimethylation of STAT6 in B cells after stimulation with LPS and IL-4 in vitro. On the contrary, GANP over-expressed B cells in ganp gene-transgenic mice showed a low STAT6 phosphorylation after stimulation. The over-expression of PRMT5 caused the up-regulation of STAT6-mediated gene transcription, which was also suppressed by the co-transfection of GANP, in luciferase reporter assay. GANP down-regulates JAK1/JAK3 to STAT6-signaling with regulation of arginine methylation activity, which might be responsible for the B cell endogenous suppressive mechanism of hyper-IgE.

Forecasting Yield Volatility with Arbitrage-Free Nelson-Siegel Models

The existing literature on interest rate volatility using standard affine, arbitrage-free mod- els of the term structure is still inconclusive as to which factors are causing interest rate volatility. This is partly due to the latent nature of the factors in such models, and partly due to the difficulty in estimating their parameters. In this paper, to address both issues simulta- neously, we choose to focus on the widely used Nelson-Siegel yield curve model with its three factors representing the level, slope, and curvature of the yield curve. Christensen, Diebold, and Rudebusch (CDR, 2007) introduced the affine arbitrage-free class of Nelson-Siegel models and showed that it is easy to estimate and delivers robust estimates of the model parameters in addition to fitting the yield curve well in sample and forecasting yields well out of sample. However, their model class is characterized by constant volatility and, hence, has little to bear on the important issue of interest rate volatility. We introduce three new classes (with a total of five subclasses) of affine arbitrage-free models that incorporate stochastic volatility fac- tors, while preserving as much of the Nelson-Siegel factor loading structure as possible. We examine the in-sample and out-of-sample properties of various specifications, ranging from the most flexible to the most parsimonious. The models' yields are compared to those of the benchmark CDR model with constant volatility. We further compare the models' volatility dynamics to realized volatility measures across the maturity spectrum. Our goal is to identify a class of models that preserves the ability to forecast the yield term structure at least on par with the CDR model, while also being able to match yield volatility.

Tracking and synchronisation for a class of PWA systems

In this paper, the tracking problem for a class of discontinuous piecewise affine (PWA) systems is addressed. We propose an observer-based output-feedback control design, consisting of a feedforward, a piecewise affine feedback law and a model-based observer, solving the tracking problem. These synthesis results can also be employed to tackle the master–slave synchronisation problem for PWA systems. It is shown that for certain classes of PWA systems the design is characterised in terms of linear matrix inequalities. The results are illustrated by application to mechanical systems with discontinuous friction characteristics.

Valuation of Credit Contingent Options with Applications to Quanto CDS

We study the valuation of credit-contingent asset or options by modelling the correlation between asset price and credit default. We provide three ways of modelling such correlation: (1) asset value follows a diffusion process with a one-time jump (such as currency devaluation) at the time of credit default; (2) Default intensity and asset price are driven by correlated Brownian motions in addition to the jump; (3) Default time and future asset price are correlated through a copula. When both asset price and credit default are independent of interest rates, such contract can be valued on a two-dimensional lattice (or finite-difference grid) in the second approach. We show that for a large class of one-factor default rate models, the computation can be reduced to one-dimension, a property often reserved for the affine class of models. We also obtain analytical solutions if default hazard rate, asset price return, and the copula are all Gaussian. Our experience shows that valuation is much more sensitive to the first and third type of correlations. We apply the model to the valuation of extinguishable FX swaps that terminate upon a credit event and quanto credit default swaps where premium and protection legs are paid in different currencies.

What does the market price of risk tell us in the single factor interest rate model?

We derive the formulas for single-factor Lévy type-bond pricing by incorporating the stochastic market price of risk, and find a sufficient condition of the market price of risk for producing the well-known affine class or the quadratic class.

A geometric approach for the design of MIMO sliding controllers. Application to a wind‐driven doubly fed induction generator

AbstractThis paper presents a systematic methodology to design controllers for a general class of nonlinear MIMO systems affine in the control in the presence of bounded uncertainties and disturbances. The design method is developed using a theoretical framework based on the combination of a geometric approach and sliding mode techniques. The resulting robust control law guarantees finite time convergence, whereas chattering reduction is achieved by utilizing the minimum discontinuous action required to ensure disturbance rejection. The proposed methodology is applied to the control of a grid‐connected wind energy generation system based on a doubly fed induction generator. The control objectives considered in this paper are maximization of the wind energy conversion and reactive power regulation to minimize machine losses. Copyright © 2008 John Wiley & Sons, Ltd.

A Bayesian Perspective on Stochastic Neurocontrol

Control design for stochastic uncertain nonlinear systems is traditionally based on minimizing the expected value of a suitably chosen loss function. Moreover, most control methods usually assume the certainty equivalence principle to simplify the problem and make it computationally tractable. We offer an improved probabilistic framework which is not constrained by these previous assumptions, and provides a more natural framework for incorporating and dealing with uncertainty. The focus of this paper is on developing this framework to obtain an optimal control law strategy using a fully probabilistic approach for information extraction from process data, which does not require detailed knowledge of system dynamics. Moreover, the proposed control method framework allows handling the problem of input-dependent noise. A basic paradigm is proposed and the resulting algorithm is discussed. The proposed probabilistic control method is for the general nonlinear class of discrete-time systems. It is demonstrated theoretically on the affine class. A nonlinear simulation example is also provided to validate theoretical development.

Adaptive diffusion using Hesse matrix for detecting nonstationary signals

Adaptive diffusion, using a gradient‐square tensor, was proposed to improve the readability of time frequency representations (TFRs) of the Cohen class and the affine class in 2005. It works with a signal‐dependent kernel and can adapt to a wide range of nonstationary signals. Here we substitute the Hesse matrix for the gradient‐square tensor after making an analysis of local diffusion behavior in an adaptive diffusion process, in close conjunction with the local signature of the auto terms and the interference terms. The eigenvectors of the Hesse matrix not only can give the local average gradient direction and thus the local diffusion direction, but also can provide an explicit mathematical explanation for local diffusion behavior. The Hesse method is capable of providing TFRs of better readability than the gradient square tensor. The validity of the new method is verified by computational simulations.

Stochastic Volatility in the Affine Arbitrage-Free Class of Nelson-Siegel Term Structure Models

Christensen, Diebold and Rudebusch (2007) derive the affine arbitrage-free class of dynamic Nelson-Siegel models and find that it performs well in terms of forecasting yields. However, that class of models is limited by the fact that it only allows for Gaussian factors with constant volatility. In this paper, we modify the above class of models to incorporate stochastic volatility factors while preserving as much of the Nelson-Siegel factor loading structure as possible. We introduce five new classes of affine arbitrage-free models. Limiting ourselves to the extended affine risk premium specification introduced in Cheridito, Filipovic and Kimmel (2007), we derive the maximally flexible specification of each class of models. We study the in-sample and out-of-sample properties of the most flexible specification for each class as well as those of the most parsimonious specification where the factors are independent. The results are compared to those reported by Christensen, Diebold and Rudebusch (2007) for the regular class of arbitrage-free dynamic Nelson-Siegel models with constant volatility. The ultimate goal is to find a class of models that preserves the ability to forecast yields at least on par with the Nelson-Siegel model while being able to also forecast yield volatility. This is important when it comes to pricing volatility sensitive products such as interest rate derivatives.

Anti-PDGFR-α antibodies measured by non-bioactivity assays are not specific for systemic sclerosis

Objective:To evaluate the presence of anti-PDGFR-α antibodies by immunological methods in patients with systemic sclerosis (SSc).Methods:Fifty-eight women diagnosed with SSc and 36 healthy women controls were included. IgG anti-PDGFR-α were...

Neural-network adaptive controller for nonlinear systems and its application in pneumatic servo systems

In this paper, a novel control law is presented, which uses neural-network techniques to approximate the affine class nonlinear system having unknown or uncertain dynamics and noise disturbances. It adopts an adaptive control law to adjust the network parameters online and adds another control component according to H-infinity control theory to attenuate the disturbance. This control law is applied to the position tracking control of pneumatic servo systems. Simulation and experimental results show that the tracking precision and convergence speed is obviously superior to the results by using the basic BP-network controller and self-tuning adaptive controller.

LINEAR‐QUADRATIC JUMP‐DIFFUSION MODELING

We aim at accommodating the existing affine jump‐diffusion and quadratic models under the same roof, namely the linear‐quadratic jump‐diffusion (LQJD) class. We give a complete characterization of the dynamics of this class by stating explicitly the structural constraints, as well as the admissibility conditions. This allows us to carry out a specification analysis for the three‐factor LQJD models. We compute the standard transform of the state vector relevant to asset pricing up to a system of ordinary differential equations. We show that the LQJD class can be embedded into the affine class using an augmented state vector. This establishes a one‐to‐one equivalence relationship between both classes in terms of transform analysis.

Recovery of Oligomers and Cooperativity When Monomers of the M2 Muscarinic Cholinergic Receptor Are Reconstituted into Phospholipid Vesicles

FLAG- and HA-tagged M2 muscarinic receptors from coinfected Sf9 cells have been purified in digitonin-cholate and reconstituted into phospholipid vesicles. The purified receptor was predominantly monomeric: it showed no detectable coimmunoprecipitation; it migrated as a monomer during electrophoresis before or after cross-linking with bis(sulfosuccinimidyl)suberate; and it bound agonists and antagonists in a manner indicative of identical and mutually independent sites. Receptor cross-linked after reconstitution or after reconstitution and subsequent solubilization in digitonin-cholate migrated almost exclusively as a tetramer. The binding properties of the reconstituted receptor mimicked those reported previously for cardiac muscarinic receptors. The apparent capacity for N-[3H]methylscopolamine (NMS) was only 60% of that for [3H]quinuclidinylbenzilate (QNB), yet binding at saturating concentrations of [3H]QNB was inhibited fully and in a noncompetitive manner at comparatively low concentrations of unlabeled NMS. Reconstitution of the receptor with a saturating quantity of functional G proteins led to the appearance of three classes of sites for the agonist oxotremorine-M in assays with [3H]QNB; GMP-PNP caused an apparent interconversion from highest to lowest affinity and the concomitant emergence of a fourth class of intermediate affinity. All of the data can be described quantitatively in terms of cooperativity among four interacting sites, presumably within a tetramer; the effect of GMP-PNP can be accommodated as a shift in the distribution of tetramers between two states that differ in their cooperative properties. Monomers of the M2 receptor therefore can be assembled into tetramers with binding properties that closely resemble those of the muscarinic receptor in myocardial preparations.

Stability of piecewise affine systems with application to chaos stabilization

This paper addresses the stability issue for a class of piecewise affine (PWA) systems, where the state spaces are assumed to be dividable into a certain number of hypercuboid subspaces. By constructing appropriate piecewise continuous Lyapunov functions, several numerically tractable stability criteria are developed for four subclasses of such PWA systems, which allow to recast the switching control problem for the PWA systems as a convex optimization problem. Moreover, the proposed method is applied to switching controller design for (globally and locally) stabilizing the unstable equilibrium points of PWA chaotic systems. Numerical simulations on the chaotic Chua's circuit are presented to verify the theoretical results.

Modelling Long Maturity Forward Rates with Overshooting

It is natural to assume that interest rates mean-revert, and natural consequence of this is that long forward rates are asymptotically constant. However, from US Treasury STRIPs data, forward rates slope increasingly downwards, and do not attenuate in volatility, as maturity increases beyond about 15 years.We fit an equilibrium model which reconciles this behavior with mean reversion. The key is to extract a latent factor, which the forward rate follows, but overshoots, causing mean reversion on a time scale in terms of weeks. This allows the forward rates to mean revert over the long term in the objective measure but not in the risk neutral measure. We show that this mean reversion can in principle be exploited for profit.Our model falls into the Essentially Affine class. We finally set our analysis in the context of similar models for shorter maturities, and discuss applications to managing positions in long maturity bonds and associated derivatives.

DNA-dependent protein kinase inhibits AID-induced antibody gene conversion.

Affinity maturation and class switching of antibodies requires activation-induced cytidine deaminase (AID)-dependent hypermutation of Ig V(D)J rearrangements and Ig S regions, respectively, in activated B cells. AID deaminates deoxycytidine bases in Ig genes, converting them into deoxyuridines. In V(D)J regions, subsequent excision of the deaminated bases by uracil-DNA glycosylase, or by mismatch repair, leads to further point mutation or gene conversion, depending on the species. In Ig S regions, nicking at the abasic sites produced by AID and uracil-DNA glycosylases results in staggered double-strand breaks, whose repair by nonhomologous end joining mediates Ig class switching. We have tested whether nonhomologous end joining also plays a role in V(D)J hypermutation using chicken DT40 cells deficient for Ku70 or the DNA-dependent protein kinase catalytic subunit (DNA-PKcs). Inactivation of the Ku70 or DNA-PKcs genes in DT40 cells elevated the rate of AID-induced gene conversion as much as 5-fold. Furthermore, DNA-PKcs-deficiency appeared to reduce point mutation. The data provide strong evidence that double-strand DNA ends capable of recruiting the DNA-dependent protein kinase complex are important intermediates in Ig V gene conversion.