The statistical properties of recreational catch rate data for some fish stocks off the northeast U.S. coast

Recreational fisheries in the waters off the northeast U.S. target a variety of pelagic and demersal fish species, and catch and effort data sampled from recreational fisheries are a critical component of the information used in resource evaluation and management. Standardized indices of stock abundance developed from recreational fishery catch rates are routinely used in stock assessments. The statistical properties of both simulated and empirical recreational fishery catch-rate data such as those collected by the National Marine Fisheries Service (NMFS) Marine Recreational Fishery Statistics Survey (MRFSS) are examined, and the potential effects of different assumptions about the error structure of the catch-rate frequency distributions in computing indices of stock abundance are evaluated. Recreational fishery catch distributions sampled by the MRFSS are highly contagious and overdispersed in relation to the normal distribution and are generally best characterized by the Poisson or negative binomial distributions. The modeling of both the simulated and empirical MRFSS catch rates indicates that one may draw erroneous conclusions about stock trends by assuming the wrong error distribution in procedures used to developed standardized indices of stock abundance. The results demonstrate the importance of considering not only the overall model fit and significance of classification effects, but also the possible effects of model misspecification, when determining the most appropriate model construction.

A multivariate statistical study with a factor analysis of recent planktonic foraminiferal distribution in the Coromandel Coast of India

A study of planktonic foraminiferal assemblages from 19 stations in the neritic and oceanic regions off the Coromandel Coast, Bay of Bengal has been made using a multivariate statistical method termed as factor analysis. On the basis of abundance, 17 foraminiferal species, species were clustered into 5 groups with row normalisation and varimax rotation for Q-mode factor analysis. The 19 stations were also grouped into 5 groups with only 2 groups statistically significant using column normalisation and varimax rotation for R-mode analysis. This assemblage grouping method is suitable because groups of species/stations can explain the maximum amount of variation in them in relation to prevailing environmental conditions in the area of study.

A statistical study on weight loss in frozen prawns

This paper deals with an extensive study conducted to estimate the extent of weight loss in frozen prawns. The weight Joss varied from 7 to 12% in peeled and deveined (PD), 5 to 7% in headless (HL) and about 7% in cooked and peeled (CP) prawns from the date of processing to the date of inspection, normally within two weeks. To compensate the weight loss nearly 11% of excess material is being added with every frozen block resulting in an average annual loss of Rs. 2.68 crores in foreign exchange. The relevant data pertain to the period 1971 to 1973 and the annual average loss was estimated for the ten years ending 1973.

Active learning of model evidence using Bayesian quadrature

Numerical integration is a key component of many problems in scientific computing, statistical modelling, and machine learning. Bayesian Quadrature is a modelbased method for numerical integration which, relative to standard Monte Carlo methods, offers increased sample efficiency and a more robust estimate of the uncertainty in the estimated integral. We propose a novel Bayesian Quadrature approach for numerical integration when the integrand is non-negative, such as the case of computing the marginal likelihood, predictive distribution, or normalising constant of a probabilistic model. Our approach approximately marginalises the quadrature model's hyperparameters in closed form, and introduces an active learning scheme to optimally select function evaluations, as opposed to using Monte Carlo samples. We demonstrate our method on both a number of synthetic benchmarks and a real scientific problem from astronomy.

N-gram posterior probability confidence measures for statistical machine translation: An empirical study

We report an empirical study of n-gram posterior probability confidence measures for statistical machine translation (SMT). We first describe an efficient and practical algorithm for rapidly computing n-gram posterior probabilities from large translation word lattices. These probabilities are shown to be a good predictor of whether or not the n-gram is found in human reference translations, motivating their use as a confidence measure for SMT. Comprehensive n-gram precision and word coverage measurements are presented for a variety of different language pairs, domains and conditions. We analyze the effect on reference precision of using single or multiple references, and compare the precision of posteriors computed from k-best lists to those computed over the full evidence space of the lattice. We also demonstrate improved confidence by combining multiple lattices in a multi-source translation framework. © 2012 The Author(s).

A Grassmann-Rayleigh quotient iteration for computing invariant subspaces

The classical Rayleigh quotient iteration (RQI) allows one to compute a one-dimensional invariant subspace of a symmetric matrix A. Here we propose a generalization of the RQI which computes a p-dimensional invariant subspace of A. Cubic convergence is preserved and the cost per iteration is low compared to other methods proposed in the literature.

A Grassman-Rayleigh quotient iteration for computing invariant subspaces

The classical Rayleigh Quotient Iteration (RQI) computes a 1-dimensional invariant subspace of a symmetric matrix A with cubic convergence. We propose a generalization of the RQI which computes a p-dimensional invariant subspace of A. The geometry of the algorithm on the Grassmann manifold Gr(p,n) is developed to show cubic convergence and to draw connections with recently proposed Newton algorithms on Riemannian manifolds.