The use of correlation and regression methods in optometry

Correlation and regression are two of the statistical procedures most widely used by optometrists. However, these tests are often misused or interpreted incorrectly, leading to erroneous conclusions from clinical experiments. This review examines the major statistical tests concerned with correlation and regression that are most likely to arise in clinical investigations in optometry. First, the use, interpretation and limitations of Pearson's product moment correlation coefficient are described. Second, the least squares method of fitting a linear regression to data and for testing how well a regression line fits the data are described. Third, the problems of using linear regression methods in observational studies, if there are errors associated in measuring the independent variable and for predicting a new value of Y for a given X, are discussed. Finally, methods for testing whether a non-linear relationship provides a better fit to the data and for comparing two or more regression lines are considered.

Higher-order co-occurrences for exploratory point pattern analysis and decision tree clustering on spatial data

Analyzing geographical patterns by collocating events, objects or their attributes has a long history in surveillance and monitoring, and is particularly applied in environmental contexts, such as ecology or epidemiology. The identification of patterns or structures at some scales can be addressed using spatial statistics, particularly marked point processes methodologies. Classification and regression trees are also related to this goal of finding "patterns" by deducing the hierarchy of influence of variables on a dependent outcome. Such variable selection methods have been applied to spatial data, but, often without explicitly acknowledging the spatial dependence. Many methods routinely used in exploratory point pattern analysis are2nd-order statistics, used in a univariate context, though there is also a wide literature on modelling methods for multivariate point pattern processes. This paper proposes an exploratory approach for multivariate spatial data using higher-order statistics built from co-occurrences of events or marks given by the point processes. A spatial entropy measure, derived from these multinomial distributions of co-occurrences at a given order, constitutes the basis of the proposed exploratory methods. © 2010 Elsevier Ltd.

Higher-order co-occurrences for exploratory point pattern analysis and decision tree clustering on spatial data

Analyzing geographical patterns by collocating events, objects or their attributes has a long history in surveillance and monitoring, and is particularly applied in environmental contexts, such as ecology or epidemiology. The identification of patterns or structures at some scales can be addressed using spatial statistics, particularly marked point processes methodologies. Classification and regression trees are also related to this goal of finding "patterns" by deducing the hierarchy of influence of variables on a dependent outcome. Such variable selection methods have been applied to spatial data, but, often without explicitly acknowledging the spatial dependence. Many methods routinely used in exploratory point pattern analysis are2nd-order statistics, used in a univariate context, though there is also a wide literature on modelling methods for multivariate point pattern processes. This paper proposes an exploratory approach for multivariate spatial data using higher-order statistics built from co-occurrences of events or marks given by the point processes. A spatial entropy measure, derived from these multinomial distributions of co-occurrences at a given order, constitutes the basis of the proposed exploratory methods. © 2010 Elsevier Ltd.

Elemental partitioning and microstructural development in duplex stainless steel weld metal

A thermodynamic analysis which is capable of estimating the austenite/ferrite equilibria in duplex stainless steels has been carried out using the sublattice thermodynamic model. The partitioning of alloying elements between the austenite and ferrite phases has been calculated as a function of temperature. The results showed that chromium partitioning was not influenced significantly by the temperature. The molybdenum, on the other hand, was found to partition preferentially into ferrite phase as the temperature decreases. A strong partitioning of nickel into the austenite was observed to decrease gradually with increasing temperature. Among the alloying elements, average nitrogen concentration was found to have the most profound effect on the phase balance and the partitioning of nitrogen into the austenite. The partitioning coefficient of nitrogen (the ratio of the mole fraction of nitrogen in the austenite to that in the ferrite) was found to be as high as 7.0 around 1300 K. Consequently, the volume fraction of austenite was influenced by relatively small additions of nitrogen. The results are compared with the experimentally observed data in a duplex stainless steel weld metal in conjunction with the solid state δ → δ + γ phase transformation. Particular attention was given to the morphological instability of grain boundary austenite allotriomorphs. A compariso between the experimental results and calculations indicated that the instability associated with irregular austenite perturbations results from the high degree of undercooling. The results suggest that the model can be used successfully to understand the development of the microstructure in duplex stainless steel weld metals.

Using genetic programming and decision trees for generating structural descriptions of four bar mechanisms

Four bar mechanisms are basic components of many important mechanical devices. The kinematic synthesis of four bar mechanisms is a difficult design problem. A novel method that combines the genetic programming and decision tree learning methods is presented. We give a structural description for the class of mechanisms that produce desired coupler curves. Constructive induction is used to find and characterize feasible regions of the design space. Decision trees constitute the learning engine, and the new features are created by genetic programming.

Robust H.264/AVC video transmission using data partitioning and unequal loss protection

In this work, we present an adaptive unequal loss protection (ULP) scheme for H264/AVC video transmission over lossy networks. This scheme combines erasure coding, H.264/AVC error resilience techniques and importance measures in video coding. The unequal importance of the video packets is identified in the group of pictures (GOP) and the H.264/AVC data partitioning levels. The presented method can adaptively assign unequal amount of forward error correction (FEC) parity across the video packets according to the network conditions, such as the available network bandwidth, packet loss rate and average packet burst loss length. A near optimal algorithm is developed to deal with the FEC assignment for optimization. The simulation results show that our scheme can effectively utilize network resources such as bandwidth, while improving the quality of the video transmission. In addition, the proposed ULP strategy ensures graceful degradation of the received video quality as the packet loss rate increases. © 2010 IEEE.

Minimum description length, regularisation and multi-modal data

Conventional feed forward Neural Networks have used the sum-of-squares cost function for training. A new cost function is presented here with a description length interpretation based on Rissanen's Minimum Description Length principle. It is a heuristic that has a rough interpretation as the number of data points fit by the model. Not concerned with finding optimal descriptions, the cost function prefers to form minimum descriptions in a naive way for computational convenience. The cost function is called the Naive Description Length cost function. Finding minimum description models will be shown to be closely related to the identification of clusters in the data. As a consequence the minimum of this cost function approximates the most probable mode of the data rather than the sum-of-squares cost function that approximates the mean. The new cost function is shown to provide information about the structure of the data. This is done by inspecting the dependence of the error to the amount of regularisation. This structure provides a method of selecting regularisation parameters as an alternative or supplement to Bayesian methods. The new cost function is tested on a number of multi-valued problems such as a simple inverse kinematics problem. It is also tested on a number of classification and regression problems. The mode-seeking property of this cost function is shown to improve prediction in time series problems. Description length principles are used in a similar fashion to derive a regulariser to control network complexity.

The sedimentology and diagenesis of Lower Permian (Rotliegend) sediments: (onshore U.K. and southern North Sea). Available in 2 volumes.

The thesis provides a comparative study of both sedimentology and diagenesis of Lower Permian (Rotliegend) strata, onshore and offshore U.K. (Southern North Sea). Onshore formations studied include the Bridgnorth, Penrith and Hopeman Sandstone, and are dominated by aeolian facies, with lesser amounts of interbedded fluvial sediments. Aeolian and fluvial strata in onshore basins typically grade laterally into alluvial fan breccias at basin margins. Onshore basins represent proximal examples of Rotliegend desert sediments. The Leman Sandstone Formation of the Ravenspurn area in the Southern North Sea displays a variety of facies indicative of a distal sedimentological setting; Aeolian, fluvial, sabkha, and playa lake sediments all being present. "Sheet-like" geometry of stratigraphical units within the Leman Sandstone, and alternation of fluvial and aeolian deposition was climatically controlled. Major first order bounding surfaces are laterally extensive and were produced by lacustrine transgression and regression from the north-west. Diagenesis within Permian strata was studied using standard petrographic microscopy, scanning electron microscopy, cold cathodo-Iuminescence, X-ray diffraction clay analysis, X-ray fluorescence spectroscopy, fluid inclusion microthermometry, and K-Ar dating of illites. The diagenesis of Permian sediments within onshore basins is remarkably similar, and a paragenetic sequence of early haematite, illitic clays, feldspar, kaolinite, quartz and late calcite is observed. In the Leman Sandstone formation, authigenic mineralogy is complex and includes early quartz, sulphates and dolomite, chlorite, kaolinite, late quartz, illite and siderite. Primary lithological variation, facies type, and the interdigitation and location of facies within a basin are important initial controls upon diagenesis. Subsequently, burial history, structure, the timing of gas emplacement, and the nature of sediments within underlying formations may also exersize significant controls upon diagenesis within Rotliegend strata.

Statnote 24: Multiple regression

In previous statnotes, the application of correlation and regression methods to the analysis of two variables (X,Y) was described. These methods can be used to determine whether there is a linear relationship between the two variables, whether the relationship is positive or negative, to test the degree of significance of the linear relationship, and to obtain an equation relating Y to X. This Statnote extends the methods of linear correlation and regression to situations where there are two or more X variables, i.e., 'multiple linear regression’.

Does money matter in inflation forecasting

This paper provides the most fully comprehensive evidence to date on whether or not monetary aggregates are valuable for forecasting US inflation in the early to mid 2000s. We explore a wide range of different definitions of money, including different methods of aggregation and different collections of included monetary assets. In our forecasting experiment we use two non-linear techniques, namely, recurrent neural networks and kernel recursive least squares regression - techniques that are new to macroeconomics. Recurrent neural networks operate with potentially unbounded input memory, while the kernel regression technique is a finite memory predictor. The two methodologies compete to find the best fitting US inflation forecasting models and are then compared to forecasts from a naive random walk model. The best models were non-linear autoregressive models based on kernel methods. Our findings do not provide much support for the usefulness of monetary aggregates in forecasting inflation.

Quality management and performance:a comparison between the UK and Turkey

Purpose: The aim of this article is to detail the correlation between quality management, specifically its tools and critical success factors, and performance in terms of primary operational and secondary organisational performances. Design/methodology/approach: Survey data from the UK and Turkey were analysed using exploratory factor analyses, structural equation modelling and regression analysis. Findings: The results show that quality management has a significant and positive impact on both primary and secondary performances; that Turkish and UK attitudes to quality management are similar; and that quality management is widely practised in manufacturing and service industries but has more statistical emphasis in the manufacturing sector. The main challenge for making quality management practice more effective lies in an appropriate balanced use of the different sorts of the tools and critical success factors. Originality/value: This study takes a novel approach by: (i) exploring the relationship between primary operational and secondary organisational performances, (ii) using service and manufacturing data and (iii) making a cross-country comparison between the UK (a developed economy) and Turkey (a developing economy). Limitations: Detailed contrast provided between only two countries. © 2013 Copyright Taylor and Francis Group, LLC.

Work-life balance benefits:employee attitudes and behaviors through the lens of social exchange theory

The question of what to provide employees in order that they reciprocate with desirable behaviors in the work place has resulted in a great amount of work in the area of social exchange. Although offering fair compensation, including salary or wages and employee benefits, has been extensively studied, the effects of offering specific types of benefits, such as work-life balance benefits, and the intangible rewards that such an offering inadvertently offers, has only been minimally explored. Utilizing past literature, this current research examined the offering of work-life balance benefits, the value employees place on those benefits, the communication of the benefits by the organization to employees, and their effect on employee attitudes and behaviors. The goal was to identify the effect on desirable outcomes when work-life balance benefits are offered to determine the usefulness to the organization of offering such benefits. To test these effects, a study of an organization known to offer a strong work-life balance benefits package was undertaken. This was accomplished through the distribution of questionnaires to identify the possible relationships involving 408 employee respondents and their 79 supervisors. This was followed with interviews of 12 individuals to ascertain the true reasons for links observed through analysis. Analysis of the data was accomplished through correlation analysis, multilevel analysis and regression analysis generated by SPSS. The results of the quantitative analysis showed support for a relationship between the offering of work-life balance benefits and perceived organizational support, perceived distributive justice, job satisfaction and OCBO. The analysis also showed a lack of support for a relationship between the offering of work-life balance benefits and organizational commitment, OCBI and IRB. The interviews offered possible reasons for the lack of support regarding the relationship between the offering of work-life balance benefits and organizational commitment as well as organizational citizenship behaviors (OCBI and IRB). The implications of these findings on future research, theory and practice in the offering of work-life balance benefits are discussed.

The effect of timeliness and credit ratings on the information content of earnings announcements

Abstract: This paper investigates the impact of timeliness and credit ratings on the information content of the earnings announcements of Greek listed firms from 2001 to 2008. Using the classical event study methodology and regression analysis, we find that firms tend to release good news on time and are inclined to delay the release of bad news. We also provide evidence that the level of corporate risk differentiates the information content of earnings according to the credit rating category. Specifically, firms displaying high creditworthiness enjoy positive excess returns on earnings announcement dates. In contrast, firms with low creditworthiness undergo significant share price erosions on earnings announcement days. We also observe a substitution effect between timeliness and credit ratings in relation to the information content of earnings announcements. Specifically, we find that as the credit category of earnings-announcing firms improves, the informational role of timeliness is mitigated.

Networks, space and organisational performance:a study of the determinants of industrial research income generation by universities

This paper examines the extent to which both network structure and spatial factors impact on the organizational performance of universities as measured by the generation of industrial research income. Drawing on data concerning the interactions of universities in the UK with large research and development (R&D)-intensive firms, the paper employs both social network analysis and regression analysis. It is found that the structural position of a university within networks with large R&D-intensive firms is significantly associated with the level of research income gained from industry. Spatial factors, on the other hand, are not found to be clearly associated with performance, suggesting that universities operate on a level playing field across regional environments once other factors are controlled for.