Numerical Analysis and Scientific Computing Commons^™

Reading In Binary Data And Creating An R User Interface, Malika J. Onstad, Brett Amidan, Kimberly Freeman

STAR Program Research Presentations

The Bonneville Power Administration (BPA) employs Phasor Measurement Units (PMUs) to measure variables such as Voltage, Frequency, and Phasor Angles every sixtieth of a second. These measurements result in terabytes of data which are analyzed to detect abnormalities in the power grid. Recently BPA has switched the data file format from DST to PDAT. A function does not currently exist to read in PDAT files in order to prepare the data for analysis. In order to do this the raw PMU data must be sorted and extracted to ensure its accuracy prior to analysis. This research worked to produce a …

Detecting Click Fraud In Online Advertising: A Data Mining Approach, Richard Oentaryo, Ee Peng Lim, Michael Finegold, David Lo, Feida Zhu, Clifton Phua, Eng-Yeow Cheu, Ghim-Eng Yap, Kelvin Sim, Kasun Perera, Bijay Neupane, Mustafa Faisal, Zeyar Aung, Wei Lee Woon, Wei Chen, Dhaval Patel, Daniel Berrar

Research Collection School Of Computing and Information Systems

Click fraud - the deliberate clicking on advertisements with no real interest on the product or service offered - is one of the most daunting problems in online advertising. Building an elective fraud detection method is thus pivotal for online advertising businesses. We organized a Fraud Detection in Mobile Advertising (FDMA) 2012 Competition, opening the opportunity for participants to work on real-world fraud data from BuzzCity Pte. Ltd., a global mobile advertising company based in Singapore. In particular, the task is to identify fraudulent publishers who generate illegitimate clicks, and distinguish them from normal publishers. The competition was held from …

Online Portfolio Selection: A Survey, Bin Li, Steven C. H. Hoi

Research Collection School Of Computing and Information Systems

Online portfolio selection is a fundamental problem in computational finance, which has been extensively studied across several research communities, including finance, statistics, artificial intelligence, machine learning, and data mining. This article aims to provide a comprehensive survey and a structural understanding of online portfolio selection techniques published in the literature. From an online machine learning perspective, we first formulate online portfolio selection as a sequential decision problem, and then we survey a variety of state-of-the-art approaches, which are grouped into several major categories, including benchmarks, Follow-the-Winner approaches, Follow-the-Loser approaches, Pattern-Matching--based approaches, and Meta-Learning Algorithms. In addition to the problem formulation …

The R Journal (December 2013) 5(2): Complete Issue, The R Foundation

The R Journal

Editorial, Hadley Wickham

Contributed Research Articles

factorplot: Improving Presentation of Simple Contrasts in Generalized Linear Models, David A. Armstrong II

spMC: Modelling Spatial Random Fields with Continuous Lag Markov Chains, Luca Sartore

RNetCDF: A Package for Reading and Writing NetCDF Datasets, Pavel Michna and Milton Woods

Surface Melting Curve Analysis with R, Stefan Rödiger, Alexander Böhm and Ingolf Schimke

Performance Attribution for Equity Portfolios, Yang Lu and David Kane

ExactCIdiff: An R Package for Computing Exact Confidence Intervals for the Difference of Two Proportions, Guogen Shan and Weizhen Wang

rlme: An R Package for Rank-Based Estimation and Prediction in Random …

Short-Term Inflation Forecasting Models For Nigeria, Sani I. Doguwa, Sarah O. Alade

CBN Journal of Applied Statistics (JAS)

Short-term inflation forecasting is an essential component of the monetary policy projections at the Central Bank of Nigeria. This paper proposes four short-term headline inflation forecasting models using the SARIMA and SARIMAX processes and compares their performance using the pseudo-out-of-sample forecasting procedure over July 2011 to September 2013. According to the results the best forecasting performance is demonstrated by the model based on the all items CPI estimated using the SARIMAX model. This model is, therefore, recommended for use in short-term forecasting of headline inflation in Nigeria. The forecasting performance up to eight months ahead, of the models based on …

Changes On Cran, Kurt Hornik, Achim Zeileis

The R Journal

New CRAN task views

New packages in CRAN task views

New contributed packages

Other changes

Conference Report: Deuxièmes Rencontres R, Aurelie Siberchicot, Stephane Dray

The R Journal

Following the success of the first "Rencontres R" (Bordeaux, 2-3 July 2012, http://r2012. bordeaux.inria.fr/), the second meeting was held in Lyon on 27-28 June 2013. This French speaking conference was a great success with 216 participants (110 were present at the first conference). The aim of the meeting was to provide a national forum for the exchange and sharing of ideas on the use of R in different disciplines. The number of participants and the list of sponsors (http://r2013-lyon.sciencesconf.org/resource/sponsors) demonstrate the increasing impact of R in both industry and academia in France. The program, detailed below, consisted of plenary sessions, …

News From The Bioconductor Project, Bioconductor Team

The R Journal

Bioconductor 2.13 was released on 15 October2013. It is compatible with R3.0.2, and consists of 749 software packages, 179 experiment data packages, and more than 690 up-to-date annotation packages. The release includes 84 new software packages, and enhancements to many others. Descriptions of new packages and updated NEWS files provided by current package maintainers are at http://bioconductor.org/news/bioc_2_13_release/

Complognormal: An R Package For Composite Lognormal Distributions, S. Nadarajah, S. A.A. Bakar

The R Journal

In recent years, composite models based on the lognormal distribution have become popular in actuarial sciences and related areas. In this short note, we present a new R package for computing the probability density function, cumulative density function, and quantile function, and for generating random numbersof anycomposite model based on the lognormal distribution. The use of the package is illustrated using a real data set.

Dynamic Parallelization Of R Functions, Stefan Böhringer

The R Journal

R offers several extension packages that allow it to perform parallel computations. These operate on fixed points in the program flow and make it difficult to deal with nested parallelism and to organize parallelism in complex computations in general. In this article we discuss, first, of how to detect parallelism in functions, and second, how to minimize user intervention in that process. We present a solution that requires minimal code changes and enables to flexibly and dynamically choose the degree of parallelization in the resulting computation. An implementation is provided by the R package parallelize.dynamic and practical issues are discussed …

Surface Melting Curve Analysis With R, Stefan Rödiger, Alexander Böhm, Ingolf Schimke

The R Journal

Nucleic acid Melting Curve Analysis is a powerful method to investigate the interaction of double stranded nucleic acids. Many researchers rely on closed source software which is not ubiquitously available, and gives only little control over the computation and data presentation. R in contrast, is open source, highly adaptable and provides numerous utilities for data import, sophisticated statistical analysis and presentation in publication quality. This article covers methods, implemented in the MBmca package, for DNA Melting Curve Analysis on microbead surfaces. Particularly, the use of the second derivative melting peaks is suggested as an additional parameter to characterize the melting …

Factorplot: Improving Presentation Of Simple Contrasts In Generalized Linear Models, David A. Armstrong Ii

The R Journal

Recent statistical literature has paid attention to the presentation of pairwise comparisons either from the point of view of the reference category problem in generalized linear models (GLMs) or in terms of multiple comparisons. Both schools of thought are interested in the parsimonious presentation of sufficient information to enable readers to evaluate the significance of contrasts resulting from the inclusion of qualitative variables in GLMs. These comparisons also arise when trying to interpret multinomial models where one category of the dependent variable is omitted as a reference. While considerable advances have been made, opportunities remain to improve the presentation of …

Changes In R, The R Core Team

The R Journal

CHANGES IN R 3.0.2

R Foundation News, Kurt Hornik

The R Journal

Donations and new members

• Donations
• New supporting institutions
• New supporting members

The R In Robotics, André Dietrich, Sebastian Zug, Jörg Kaiser

The R Journal

The aim of this contribution is to connect two previously separated worlds: robotic application development with the Robot Operating System (ROS) and statistical programming with R. This fruitful combination becomes apparent especially in the analysis and visualization of sensory data. We therefore introduce a new language extension for ROS that allows to implement nodes in pure R. All relevant aspects are described in a step-by-step development of a common sensor data transformation node. This includes the reception of raw sensory data via the ROS network, message interpretation, bag-file analysis, transformation and visualization, as well as the transmission of newly generated …

Spmc: Modelling Spatial Random Fields With Continuous Lag Markov Chains, Luca Sartore

The R Journal

Currently, a part of the R statistical software is developed in order to deal with spatial models. More specifically, some available packages allow the user to analyse categorical spatial random patterns. However, only the spMC package considers a viewpoint based on transition probabilities between locations. Through the use of this package it is possible to analyse the spatial variability of data, make inference, predict and simulate the categorical classes in unobserved sites. An example is presented by analysing the well-known Swiss Jura data set.

Rlme: An R Package For Rank-Based Estimation And Prediction In Random Effects Nested Models, Yusuf K. Bilgic, Herbert Susmann

The R Journal

There is a lack of robust statistical analyses for random effects linear models. In practice, statistical analyses, including estimation, prediction and inference, are not reliable when data are unbalanced, of small size, contain outliers, or not normally distributed. It is fortunate that rank-based regression analysis is a robust nonparametric alternative to likelihood and least squares analysis. We propose an R package that calculates rank-based statistical analyses for two- and three-level random effects nested designs. In this package, a new algorithm which recursively obtains robust predictions for both scale and random effects is used, along with three rank-based fitting methods.

Performance Attribution For Equity Portfolios, Yang Lu, David Kane

The R Journal

The pa package provides tools for conducting performance attribution for long-only, single currency equity portfolios. The package uses two methods: the Brinson-Hood-Beebower model (hereafter referred to as the Brinson model) and a regression-based analysis. The Brinson model takes an ANOVA-type approach and decomposes the active return of any portfolio into asset allocation, stock selection, and interaction effect. The regression-based analysis utilizes estimated coefficients, based on a regression model, to attribute active return to different factors.

Temporal Disaggregation Of Time Series, Christoph Sax, Peter Steiner

The R Journal

Temporal disaggregation methods are used to disaggregate low frequency time series to higher frequency series, where either the sum, the average, the first or the last value of the resulting high frequency series is consistent with the low frequency series. Temporal disaggregation can be performed with or without one or more high frequency indicator series. The package tempdisagg is a collection of several methods for temporal disaggregation.

On Sampling From The Multivariate T Distribution, Marius Hofert

The R Journal

The multivariate normal and the multivariate t distributions belong to the most widely used multivariate distributions in statistics, quantitative risk management, and insurance. In contrast to the multivariate normal distribution, the parameterization of the multivariate t distribution does not correspond to its moments. This, paired with a non-standard implementation in the R package mvtnorm, provides traps for working with the multivariate t distribution. In this paper, common traps are clarified and corresponding recent changes to mvtnorm are presented.