Data Structures for Computational Statistics

Klinke, Sigbert

doi:10.1007/978-3-642-59242-3

Cited by 9 publications

(5 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thus a departure from a parabolic density is also a departure from the standard normal density. For simplicity, Klinke [24] proposes to use the following uniform kernel that we will also consider in what follows.…”

Section: The Friedman-tukey Indexmentioning

confidence: 99%

“…From these pioneer papers and until the end of the nineties, the field of EPP widely expanded in the statistical literature. Let us quote Huber [21], Jones and Sibson [22], Friedman [14], Cook et al [8], Sun [42], Posse [36], Nason [34] and Klinke [24] among the most famous references of that period. In these articles, several projection indices and optimization algorithms have been proposed and studied in detail.…”

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Genetic algorithms and particle swarm optimization for exploratory projection pursuit

Berro¹,

Marie-Sainte²,

Ruiz‐Gazen³

2010

Ann Math Artif Intell

View full text Add to dashboard Cite

Exploratory Projection Pursuit (EPP) methods have been developed thirty years ago in the context of exploratory analysis of large data sets. These methods consist in looking for low-dimensional projections that reveal some interesting structure existing in the data set but not visible in high dimension. Each projection is associated with a real valued index which optima correspond to valuable projections. Several EPP indices have been proposed in the statistics literature but the main problem lies in their optimization. In the present paper, we propose to apply Genetic Algorithms (GA) and recent Particle Swarm Optimization (PSO) algorithm to the optimization of several projection pursuit indices. We explain how the EPP methods can be implemented in order to become an efficient and powerful tool for the statistician. We illustrate our proposal on several simulated and real data sets.

show abstract

Section: The Friedman-tukey Indexmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 98%

Genetic algorithms and particle swarm optimization for exploratory projection pursuit

Berro¹,

Marie-Sainte²,

Ruiz‐Gazen³

2010

Ann Math Artif Intell

View full text Add to dashboard Cite

show abstract

“…where h denotes the bandwidth. Following Klinke [2012] for the kernel choice and Silverman [1986] for the bandwidth choice, we use the uniform kernel…”

Section: Projection Indicesmentioning

confidence: 99%

REPPlab: An R package for detecting clusters and outliers using exploratory projection pursuit

Fischer

Berro

Nordhausen³

et al. 2019

Communications in Statistics - Simulation and Computation

View full text Add to dashboard Cite

The R-package REPPlab is designed to explore multivariate data sets using onedimensional unsupervised projection pursuit. It is useful in practice as a preprocessing step to find clusters or as an outlier detection tool for multivariate numerical data. Except from the package tourr that implements smooth sequences of projection matrices and rggobi that provides an interface to a dynamic graphics package called GGobi, there is no implementation of exploratory projection pursuit tools available in R especially in the context of outlier detection. REPPlab is an R interface for the Java program EPP-lab that implements four projection indices and three biologically inspired optimization algorithms. The implemented indices are either adapted to cluster or to outlier detection and the optimization algorithms have at most one parameter to tune. Following the original software EPP-lab, the exploration strategy in REPPlab is divided into two steps. Many potentially interesting projections are calculated at the first step and examined at the second step. For this second step, different tools for plotting and combining the results are proposed with specific tools for outlier detection. Compared to EPP-lab, some of these tools are new and their performance is illustrated through some simulations and using some real data sets in a clustering context. The functionalities of the package are also illustrated for outlier detection on a new data set that is provided with the package.

show abstract

“…To graphically ascertain the normal distribution of data we can use Quantile-Quantile Plot for normal distribution. Quantile-quantile plots are used to compare the distribution of random variables (Klinke, 2001 Hence for graphical analysis it can be concluded with certain confidence EPS1, EPS2 and EPS3 valuation errors are more tightly and normally distributed as compared to the rest of valuation errors from other value drivers considered here. In other words, normal distribution of valuation errors for EPS forecasted imply that expected valuation error is its 'mode', which in our case is 0.77% -2.81% (refer Table 3 Summary of Descriptive Statistics).…”

Section: Figure 12 Combined Frequency Distribution Of Graph Of All Vamentioning

confidence: 99%

Multiple Differences: Selecting Appropriate Value Driver for Multiple based Valuation in FTSEALL index

Amin

2011

JISR-MSSE

View full text Add to dashboard Cite

Multiple based valuation is a valuation technique whose use is much frequent in many situations when it comes to equity valuation. According to Damodaran (2006) though Multiple based valuation method of valuing stocks has got very little theoretical evidence, but still it is the most popular in industry. It has even been likened to "an art form" (Bhojraj, 2002). Reason of its widespread use could be attributed to its ease of use, but does this method provide a good valuation tool and if yes, what sort of value drivers are most useful in predicting equity value? To answer this question, FTSEALL index was considered as it represents more than 98% of the UK market in terms of capitalization. Valuation for each company in FTSEALL index was carried out on 8 separate value drivers, which included both historical and forward looking drivers, resulting in a set of 9 separate valuations or 'Predicted Values' for each company. These valuations were then compared with closing prices to compute the valuation errors, on which detailed statistical analysis was performed to observe which value drivers resulted in least amount of valuation error, indicating its robustness as a value driver.

show abstract

Data Structures for Computational Statistics

Cited by 9 publications

References 0 publications

Genetic algorithms and particle swarm optimization for exploratory projection pursuit

Genetic algorithms and particle swarm optimization for exploratory projection pursuit

REPPlab: An R package for detecting clusters and outliers using exploratory projection pursuit

Multiple Differences: Selecting Appropriate Value Driver for Multiple based Valuation in FTSEALL index

Contact Info

Product

Resources

About