Searcharxiv⌕ Search

arXiv subjects

Akimichi Takemura

Publications and source records attributed to Akimichi Takemura.

At least 37 records · Page 2Linked to original sources
arXiv2014

Schur Complement Based Analysis of MIMO Zero-Forcing for Rician Fading

For multiple-input/multiple-output (MIMO) spatial multiplexing with zero-forcing detection (ZF), signal-to-noise ratio (SNR) analysis for Rician fading involves the cumbersome noncentral-Wishart distribution (NCWD) of the transmit sample-correlation (Gramian) matrix. An \textsl{approximation} with a \textsl{virtual} CWD previously yielded for the ZF SNR an approximate (virtual) Gamma distribution. However, analytical conditions qualifying the accuracy of the SNR-distribution approximation were unknown. Therefore, we have been attempting to exactly characterize ZF SNR for Rician fading. Our previous attempts succeeded only for the sole Rician-fading stream under Rician--Rayleigh fading, by writing it as scalar Schur complement (SC) in the Gramian. Herein, we pursue a more general, matrix-SC-based analysis to characterize SNRs when several streams may undergo Rician fading. On one hand, for full-Rician fading, the SC distribution is found to be exactly a CWD if and only if a channel-mean--correlation \textsl{condition} holds. Interestingly, this CWD then coincides with the \textsl{virtual} CWD ensuing from the \textsl{approximation}. Thus, under the \textsl{condition}, the actual and virtual SNR-distributions coincide. On the other hand, for Rician--Rayleigh fading, the matrix-SC distribution is characterized in terms of determinant of matrix with elementary-function entries, which also yields a new characterization of the ZF SNR. Average error probability results validate our analysis vs.~simulation.

cs.IT↗
arXiv2014

An objective look at obtaining the plotting positions for QQ-plots

Choosing the plotting positions for the QQ-plot has been a subject of much debate in the statistical and engineering literature. This paper looks at this problem objectively by considering three frameworks: distribution-theoretic; decision-theoretic; game-theoretic. In each framework, we derive the plotting positions and show that there are more than one legitimate solution depending on the practitioner's objective. This work clarifies the choice of the plotting positions by allowing one to easily find the mathematical equivalent of their view and choose the corresponding solution. This work also discusses approximations to the plotting positions when no closed form is available.

math.ST↗
arXiv2014

Estimation of exponential-polynomial distribution by holonomic gradient descent

We study holonomic gradient decent for maximum likelihood estimation of exponential-polynomial distribution, whose density is the exponential function of a polynomial in the random variable. We first consider the case that the support of the distribution is the set of positive reals. We show that the maximum likelihood estimate (MLE) can be easily computed by the holonomic gradient descent, even though the normalizing constant of this family does not have a closed-form expression and discuss determination of the degree of the polynomial based on the score test statistic. Then we present extensions to the whole real line and to the bivariate distribution on the positive orthant.

math.ST↗
arXiv2014

Derandomization in Game-theoretic Probability

We give a general method for constructing a deterministic strategy of Reality from a randomized strategy in game-theoretic probability. The construction can be seen as derandomization in game-theoretic probability.

math.PR↗
arXiv2014

Markov degree of configurations defined by fibers of a configuration

We consider a series of configurations defined by fibers of a given base configuration. We prove that Markov degree of the configurations is bounded from above by the Markov complexity of the base configuration. As important examples of base configurations we consider incidence matrices of graphs and study the maximum Markov degree of configurations defined by fibers of the incidence matrices. In particular we give a proof that the Markov degree for two-way transportation polytopes is three.

math.CO↗
arXiv2014

A framework for fitting sparse data

This paper develops a framework for fitting functions with domains in the Euclidean space, when data are sparse but a slow variation allows for a useful fit. We measure the variation by Lipschitz Bound (LB). Functions which admit smaller LB are considered to vary more slowly. Since most functions in practice are wiggly and do not admit a small LB, we extend this framework by approximating a wiggly function, f, by ones which admit a smaller LB and do not deviate from f by more than a specified Bound Deviation (BD). In fact for any positive LB, one can find such a BD, thus defining a trade-off function (LB-BD function) between the variation measure (LB) and the deviation measure (BD). We show that the LB-BD function satisfies nice properties: it is non-increasing and convex. We also present a method to obtain it using convex optimization. For a function with given LB and BD, we find the optimal fit and present deterministic bounds for the prediction error of various methods. Given the LB-BD function, we discuss picking an appropriate LB-BD pair for fitting and calculating the prediction errors. The developed methods can naturally accommodate an extra assumption of periodicity to obtain better prediction errors. Finally we present the application of this framework to air pollution data with sparse observations over time.

stat.ME↗
arXiv2014

Conformal Geometry of Sequential Test in Multidimensional Curved Exponential Family

This article presents a differential geometrical method for analyzing sequential test procedures. It is based on the primal result on the conformal geometry of statistical manifold developed in Kumon, Takemura and Takeuchi (2011). By introducing curvature-type random variables, the condition is first clarified for a statistical manifold to be an exponential family under an appropriate sequential test procedure. This result is further elaborated for investigating the efficient sequential test in a multidimensional curved exponential family. The theoretical results are numerically examined by using von Mises-Fisher and hyperboloid models.

math.ST↗
arXiv2014

Properties of powers of functions satisfying second-order linear differential equations with applications to statistics

We derive properties of powers of a function satisfying a second-order linear differential equation. In particular we prove that the n-th power of the function satisfies an (n+1)-th order differential equation and give a simple method for obtaining the differential equation. Also we determine the exponents of the differential equation and derive a bound for the degree of the polynomials, which are coefficients in the differential equation. The bound corresponds to the order of differential equation satisfied by the n-fold convolution of the Fourier transform of the function. These results are applied to some probability density functions used in statistics.

math.CA↗
arXiv2014

Exact MIMO Zero-Forcing Detection Analysis for Transmit-Correlated Rician Fading

We analyze the performance of multiple input/multiple output (MIMO) communications systems employing spatial multiplexing and zero-forcing detection (ZF). The distribution of the ZF signal-to-noise ratio (SNR) is characterized when either the intended stream or interfering streams experience Rician fading, and when the fading may be correlated on the transmit side. Previously, exact ZF analysis based on a well-known SNR expression has been hindered by the noncentrality of the Wishart distribution involved. In addition, approximation with a central-Wishart distribution has not proved consistently accurate. In contrast, the following exact ZF study proceeds from a lesser-known SNR expression that separates the intended and interfering channel-gain vectors. By first conditioning on, and then averaging over the interference, the ZF SNR distribution for Rician-Rayleigh fading is shown to be an infinite linear combination of gamma distributions. On the other hand, for Rayleigh-Rician fading, the ZF SNR is shown to be gamma-distributed. Based on the SNR distribution, we derive new series expressions for the ZF average error probability, outage probability, and ergodic capacity. Numerical results confirm the accuracy of our new expressions, and reveal effects of interference and channel statistics on performance.

cs.IT↗
arXiv2013

Markov degree of the Birkhoff model

We prove the conjecture by Diaconis and Eriksson (2006) that the Markov degree of the Birkhoff model is three. In fact, we prove the conjecture in a generalization of the Birkhoff model, where each voter is asked to rank a fixed number, say r, of candidates among all candidates. We also give an exhaustive characterization of Markov bases for small r.

math.ST↗
arXiv2013

Optimality of Thompson Sampling for Gaussian Bandits Depends on Priors

In stochastic bandit problems, a Bayesian policy called Thompson sampling (TS) has recently attracted much attention for its excellent empirical performance. However, the theoretical analysis of this policy is difficult and its asymptotic optimality is only proved for one-parameter models. In this paper we discuss the optimality of TS for the model of normal distributions with unknown means and variances as one of the most fundamental example of multiparameter models. First we prove that the expected regret of TS with the uniform prior achieves the theoretical bound, which is the first result to show that the asymptotic bound is achievable for the normal distribution model. Next we prove that TS with Jeffreys prior and reference prior cannot achieve the theoretical bound. Therefore the choice of priors is important for TS and non-informative priors are sometimes risky in cases of multiparameter models.

math.ST↗
arXiv2013

Markov degree of the three-state toric homogeneous Markov chain model

We consider the three-state toric homogeneous Markov chain model (THMC) without loops and initial parameters. At time $T$, the size of the design matrix is $6 \times 3\cdot 2^{T-1}$ and the convex hull of its columns is the model polytope. We study the behavior of this polytope for $T\geq 3$ and we show that it is defined by 24 facets for all $T\ge 5$. Moreover, we give a complete description of these facets. From this, we deduce that the toric ideal associated with the design matrix is generated by binomials of degree at most 6. Our proof is based on a result due to Sturmfels, who gave a bound on the degree of the generators of a toric ideal, provided the normality of the corresponding toric variety. In our setting, we established the normality of the toric variety associated to the THMC model by studying the geometric properties of the model polytope.

math.ST↗
arXiv2013

Properties and applications of Fisher distribution on the rotation group

We study properties of Fisher distribution (von Mises-Fisher distribution, matrix Langevin distribution) on the rotation group SO(3). In particular we apply the holonomic gradient descent, introduced by Nakayama et al. (2011), and a method of series expansion for evaluating the normalizing constant of the distribution and for computing the maximum likelihood estimate. The rotation group can be identified with the Stiefel manifold of two orthonormal vectors. Therefore from the viewpoint of statistical modeling, it is of interest to compare Fisher distributions on these manifolds. We illustrate the difference with an example of near-earth objects data.

stat.ME↗
arXiv2013

Holonomic gradient method for the distribution function of the largest root of a Wishart matrix

We apply the holonomic gradient method introduced by Nakayama et al.(2011) to the evaluation of the exact distribution function of the largest root of a Wishart matrix, which involves a hypergeometric function 1F1 of a matrix argument. Numerical evaluation of the hypergeometric function has been one of the longstanding problems in multivariate distribution theory. The holonomic gradient method offers a totally new approach, which is complementary to the infinite series expansion around the origin in terms of zonal polynomials. It allows us to move away from the origin by the use of partial differential equations satisfied by the hypergeometric function. From numerical viewpoint we show that the method works well up to dimension 10. From theoretical viewpoint the method offers many challenging problems both to statistics and D-module theory.

math.ST↗
arXiv2013

Calculation of orthant probabilities by the holonomic gradient method

We apply the holonomic gradient method (HGM) introduced by [9] to the calculation of orthant probabilities of multivariate normal distribution. The holonomic gradient method applied to orthant probabilities is found to be a variant of Plackett's recurrence relation ([14]). However an implementation of the method yields recurrence relations more suitable for numerical computation than Plackett's recurrence relation. We derive some theoretical results on the holonomic system for the orthant probabilities. These results show that multivariate normal orthant probabilities possess some remarkable properties from the viewpoint of holonomic systems. Finally we show that numerical performance of our method is comparable or superior compared to existing methods.

math.NA↗
arXiv2012

Bayesian logistic betting strategy against probability forecasting

We propose a betting strategy based on Bayesian logistic regression modeling for the probability forecasting game in the framework of game-theoretic probability by Shafer and Vovk (2001). We prove some results concerning the strong law of large numbers in the probability forecasting game with side information based on our strategy. We also apply our strategy for assessing the quality of probability forecasting by the Japan Meteorological Agency. We find that our strategy beats the agency by exploiting its tendency of avoiding clear-cut forecasts.

math.PR↗
arXiv2012

Finite-time Regret Bound of a Bandit Algorithm for the Semi-bounded Support Model

In this paper we consider stochastic multiarmed bandit problems. Recently a policy, DMED, is proposed and proved to achieve the asymptotic bound for the model that each reward distribution is supported in a known bounded interval, e.g. [0,1]. However, the derived regret bound is described in an asymptotic form and the performance in finite time has been unknown. We inspect this policy and derive a finite-time regret bound by refining large deviation probabilities to a simple finite form. Further, this observation reveals that the assumption on the lower-boundedness of the support is not essential and can be replaced with a weaker one, the existence of the moment generating function.

math.ST↗