PhD in Statistics and Operations Research (UPC) program, with a work framed in factorial data analyzes. Degree in Mathematical Sciences (UAB). Lecture at the University of Girona, Department of Computer Science, Applied Mathematics and Statistics..My field of work in teaching focuses on statistics both at the basic level and at the most advanced level in methods of multivariate statistical analysis: factorial methods and classification.The research focuses on three areas: 1- The analysis of compositional data (CoDA), where I have published theoretical contributions of development of a specific methodology for compositional data and applications in different fields: geology, archeology, animal welfare ... 2- The methods of data fusion, which seeks the integration of information from large databases with applications to health data, merging health surveys and exams, and data consumption and audience among others. 3-the applications of statistics in medical research and, more specifically, in the fields of radiology and the medical image of infarcts, strokes and cardiovascular diseases.I have published in several specialized magazines: Computers
In this chapter, we will use compositional data analysis (CODA) to document the geometric variati... more In this chapter, we will use compositional data analysis (CODA) to document the geometric variation of limb proportions in ternary morphospaces and in linear bivariate spaces. This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.
In this chapter, we will use compositional data analysis (CODA) to document the geometric variati... more In this chapter, we will use compositional data analysis (CODA) to document the geometric variation of limb proportions in ternary morphospaces and in linear bivariate spaces. This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.
Estimation of health prevalences is usually performed with a single survey. Some attempts have be... more Estimation of health prevalences is usually performed with a single survey. Some attempts have been made to integrate more than one source of data. We propose here to validate this approach through data fusion. Data Fusion is the process of integrating two sources of data into one combined file. It allows us to take even greater advantage of existing information collected in databases. Here, we use data fusion to improve the estimation of health prevalences for two primary health factors: cardiovascular diseases and diabetes. We use a real data fusion operation on population health, where the imputation of basic health risk factors is used to enrich a large-scale survey on self-reported health status. We propose choosing the imputation methodology for this problem through a suite of validation statistics that assess the quality of the fused data. The compared imputation techniques have been chosen from among the main imputation methodologies: k-nearest neighbor, probabilistic modeling and regression. We use the 2006 Health Survey of Catalonia, which provides a complete report of the perceived health status. In order to deal with the uncertainty problem, we compare these methodologies under the single and multiple imputation frames. A suite of validation statistics allows us to discern the strengths and weaknesses of studied imputation methods. Multiple outperforms single imputation by providing better and much more stable estimates, according to the computed validation statistics. The summarized results indicate that the probabilistic methods preserve the multivariate structure better; sequential regression methods deliver greater accuracy of imputed data; and nearest neighbor methods end up with a more realistic distribution of imputed data. Data fusion allows us to integrate two sources of information in order to take grater advantage of the available data. Multiple imputed sequential regression models have the advantage of grater interpretability and can be used for health policy. Under certain conditions, more accurate estimates of the prevalences can be obtained using fused data (the original data plus the imputed data) than just by using only the observed data.
Xxxi Congreso Nacional De Estadistica E Investigacion Operativa V Jornadas De Estadistica Publica Murcia 10 13 De Febrero De 2009 Libro De Actas 2009 Isbn 978 84 691 8159 1, 2009
El Aprendizaje Basado en Problemas (ABP) tiene sus or genes en la educaci on m edica, por esta ra... more El Aprendizaje Basado en Problemas (ABP) tiene sus or genes en la educaci on m edica, por esta raz on se equipara a menudo problemas con casos. La clase de \casos" o \problemas" que se presentan en ambientes de ABP son situaciones en las que los estudiantes ...
La memòria d'aquesta tesi doctoral s'estructura en un primer capítol on es descriuen el... more La memòria d'aquesta tesi doctoral s'estructura en un primer capítol on es descriuen els objectius de la tesi i l'organització del treball de recerca. Després, el Capítol 2: Anàlisis factorials de dades es destina a presentar les eines utilitzades en les anàlisis factorials de dades. ...
Xxv Congreso Nacional De Estadistica E Investigacion Operativa Vigo 4 7 De Abril De 2000 2000 Isbn 84 8158 152 6 Pags 61 62, 2000
Información del artículo Análisis de la movilidad obligada por el trabajo entre las 41 comarcas c... more Información del artículo Análisis de la movilidad obligada por el trabajo entre las 41 comarcas catalanas.
Resumen Esta comunicación aborda la especificidad de la actividad investigadora del proyecto arqu... more Resumen Esta comunicación aborda la especificidad de la actividad investigadora del proyecto arquitectónico y su aplicación en la docencia de la asignatura de proyectos arquitectónicos. Frente a las investigaciones de otras disciplinas científicas, la arquitectura como ...
Actas Del Xxx Congreso Nacional De Estadistica E Investigacion Operativa Y De Las Iv Jornadas De Estadistica Publica 2007 Isbn 978 84 690 7249 3, 2007
La metodolog a log-cociente pone a disposici on de los investigadores pode-rosas herramientas par... more La metodolog a log-cociente pone a disposici on de los investigadores pode-rosas herramientas para el an alisis de datos composicionales, sin embargo, el uso de esta metodolog as olo es posible para aquellos conjuntos de datos que no contengan valores nulos o ceros. En ...
A new integrated approach to the analysis of square nonsymmetric tables is introduced by means of... more A new integrated approach to the analysis of square nonsymmetric tables is introduced by means of correspondence analysis. The application of correspondence analysis to such tables is not successful due to the strong role played by the diagonal values: overloaded diagonals and structural zeros. Two main families of methods of resolution are integrated in this paper. The resulting method is applied to a study of commuting between the 41 Catalan counties.
The log-ratio methodology makes available powerful tools for analyzing compositionaldata. Neverth... more The log-ratio methodology makes available powerful tools for analyzing compositionaldata. Nevertheless, the use of this methodology is only possible for those data setswithout null values. Consequently, in those data sets where the zeros are present, aprevious treatment becomes necessary. Last advances in the treatment of compositionalzeros have been centered especially in the zeros of structural nature and in the roundedzeros. These tools do not contemplate the particular case of count compositional datasets with null values. In this work we deal with \count zeros" and we introduce atreatment based on a mixed Bayesian-multiplicative estimation. We use the Dirichletprobability distribution as a prior and we estimate the posterior probabilities. Then weapply a multiplicative modi¯cation for the non-zero values. We present a case studywhere this new methodology is applied.Key words: count data, multiplicative replacement, composition, log-ratio analysis Geologische Vereinigung; I...
PURPOSE Computed tomography (CT) can detect thrombi in arteries after acute stroke and can provid... more PURPOSE Computed tomography (CT) can detect thrombi in arteries after acute stroke and can provide a measure of their composition based on Hounsfield units (HU). Thromboembolic stroke can be caused by platelet-rich, erythrocyte-rich, or mixed clots. Platelet-rich thrombi are resistant to thrombolytic drugs, but erythrocyte-rich thrombi are not. Experimental studies have shown lower HU in platelet-rich thrombi than in erythrocyte-rich thrombi. We investigated whether HU values can predict the lysability of thrombi. METHOD AND MATERIALS We performed thin-section noncontrast CT and CT angiographic studies before IV rt-PA in 30 patients (18 men; mean age,67±10years) with thrombi in the middle cerebral artery within 4.5 hours of stroke onset. HU values were measured in the thrombosed and contralateral normal segments and calculated the ratios (rHU) between them. Recanalization was determined by MR angiography after IV rt-PA. We used linear discriminant analysis to find the cutoff value o...
The linkage among the tissue iron stores, insulin resistance (IR), and cognition remains unclear ... more The linkage among the tissue iron stores, insulin resistance (IR), and cognition remains unclear in the obese population. We aimed to identify the factors that contribute to increased hepatic iron concentration (HIC) and brain iron overload (BIO), as evaluated by MRI, and to evaluate their impact on cognitive performance in obese and nonobese subjects. We prospectively recruited 23 middle-aged obese subjects without diabetes (13 women; age 50.4 ± 7.7 years; BMI 43.7 ± 4.48 kg/m2) and 20 healthy nonobese volunteers (10 women; age 48.8 ± 9.5 years; BMI 24.3 ± 3.54 kg/m2) in whom iron load was assessed in white and gray matter and the liver by MRI. IR was measured from HOMA-IR and an oral glucose tolerance test. A battery of neuropsychological tests was used to evaluate the cognitive performance. Multivariate regression analysis was used to identify the independent associations of BIO and cognitive performance. A significant increase in iron load was detected at the caudate nucleus (P < 0.001), lenticular nucleus (P = 0.004), hypothalamus (P = 0.002), hippocampus (P < 0.001), and liver (P < 0.001) in obese subjects. There was a positive correlation between HIC and BIO at caudate (r = 0.517, P < 0.001), hypothalamus (r = 0.396, P = 0.009), and hippocampus (r = 0.347, P < 0.023). The area under the curve of insulin was independently associated with BIO at the caudate (P = 0.001), hippocampus (P = 0.028), and HIC (P = 0.025). BIOs at the caudate (P = 0.028), hypothalamus (P = 0.006), and lenticular nucleus (P = 0.012) were independently associated with worse cognitive performance. Obesity and IR may contribute to increased HIC and BIO being associated with worse cognitive performance. BIO could be a potentially useful MRI biomarker for IR and obesity-associated cognitive dysfunction.
In this chapter, we will use compositional data analysis (CODA) to document the geometric variati... more In this chapter, we will use compositional data analysis (CODA) to document the geometric variation of limb proportions in ternary morphospaces and in linear bivariate spaces. This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.
In this chapter, we will use compositional data analysis (CODA) to document the geometric variati... more In this chapter, we will use compositional data analysis (CODA) to document the geometric variation of limb proportions in ternary morphospaces and in linear bivariate spaces. This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.
In this chapter, we will use compositional data analysis (CODA) to document the geometric variati... more In this chapter, we will use compositional data analysis (CODA) to document the geometric variation of limb proportions in ternary morphospaces and in linear bivariate spaces. This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.
Estimation of health prevalences is usually performed with a single survey. Some attempts have be... more Estimation of health prevalences is usually performed with a single survey. Some attempts have been made to integrate more than one source of data. We propose here to validate this approach through data fusion. Data Fusion is the process of integrating two sources of data into one combined file. It allows us to take even greater advantage of existing information collected in databases. Here, we use data fusion to improve the estimation of health prevalences for two primary health factors: cardiovascular diseases and diabetes. We use a real data fusion operation on population health, where the imputation of basic health risk factors is used to enrich a large-scale survey on self-reported health status. We propose choosing the imputation methodology for this problem through a suite of validation statistics that assess the quality of the fused data. The compared imputation techniques have been chosen from among the main imputation methodologies: k-nearest neighbor, probabilistic modeling and regression. We use the 2006 Health Survey of Catalonia, which provides a complete report of the perceived health status. In order to deal with the uncertainty problem, we compare these methodologies under the single and multiple imputation frames. A suite of validation statistics allows us to discern the strengths and weaknesses of studied imputation methods. Multiple outperforms single imputation by providing better and much more stable estimates, according to the computed validation statistics. The summarized results indicate that the probabilistic methods preserve the multivariate structure better; sequential regression methods deliver greater accuracy of imputed data; and nearest neighbor methods end up with a more realistic distribution of imputed data. Data fusion allows us to integrate two sources of information in order to take grater advantage of the available data. Multiple imputed sequential regression models have the advantage of grater interpretability and can be used for health policy. Under certain conditions, more accurate estimates of the prevalences can be obtained using fused data (the original data plus the imputed data) than just by using only the observed data.
Xxxi Congreso Nacional De Estadistica E Investigacion Operativa V Jornadas De Estadistica Publica Murcia 10 13 De Febrero De 2009 Libro De Actas 2009 Isbn 978 84 691 8159 1, 2009
El Aprendizaje Basado en Problemas (ABP) tiene sus or genes en la educaci on m edica, por esta ra... more El Aprendizaje Basado en Problemas (ABP) tiene sus or genes en la educaci on m edica, por esta raz on se equipara a menudo problemas con casos. La clase de \casos" o \problemas" que se presentan en ambientes de ABP son situaciones en las que los estudiantes ...
La memòria d'aquesta tesi doctoral s'estructura en un primer capítol on es descriuen el... more La memòria d'aquesta tesi doctoral s'estructura en un primer capítol on es descriuen els objectius de la tesi i l'organització del treball de recerca. Després, el Capítol 2: Anàlisis factorials de dades es destina a presentar les eines utilitzades en les anàlisis factorials de dades. ...
Xxv Congreso Nacional De Estadistica E Investigacion Operativa Vigo 4 7 De Abril De 2000 2000 Isbn 84 8158 152 6 Pags 61 62, 2000
Información del artículo Análisis de la movilidad obligada por el trabajo entre las 41 comarcas c... more Información del artículo Análisis de la movilidad obligada por el trabajo entre las 41 comarcas catalanas.
Resumen Esta comunicación aborda la especificidad de la actividad investigadora del proyecto arqu... more Resumen Esta comunicación aborda la especificidad de la actividad investigadora del proyecto arquitectónico y su aplicación en la docencia de la asignatura de proyectos arquitectónicos. Frente a las investigaciones de otras disciplinas científicas, la arquitectura como ...
Actas Del Xxx Congreso Nacional De Estadistica E Investigacion Operativa Y De Las Iv Jornadas De Estadistica Publica 2007 Isbn 978 84 690 7249 3, 2007
La metodolog a log-cociente pone a disposici on de los investigadores pode-rosas herramientas par... more La metodolog a log-cociente pone a disposici on de los investigadores pode-rosas herramientas para el an alisis de datos composicionales, sin embargo, el uso de esta metodolog as olo es posible para aquellos conjuntos de datos que no contengan valores nulos o ceros. En ...
A new integrated approach to the analysis of square nonsymmetric tables is introduced by means of... more A new integrated approach to the analysis of square nonsymmetric tables is introduced by means of correspondence analysis. The application of correspondence analysis to such tables is not successful due to the strong role played by the diagonal values: overloaded diagonals and structural zeros. Two main families of methods of resolution are integrated in this paper. The resulting method is applied to a study of commuting between the 41 Catalan counties.
The log-ratio methodology makes available powerful tools for analyzing compositionaldata. Neverth... more The log-ratio methodology makes available powerful tools for analyzing compositionaldata. Nevertheless, the use of this methodology is only possible for those data setswithout null values. Consequently, in those data sets where the zeros are present, aprevious treatment becomes necessary. Last advances in the treatment of compositionalzeros have been centered especially in the zeros of structural nature and in the roundedzeros. These tools do not contemplate the particular case of count compositional datasets with null values. In this work we deal with \count zeros" and we introduce atreatment based on a mixed Bayesian-multiplicative estimation. We use the Dirichletprobability distribution as a prior and we estimate the posterior probabilities. Then weapply a multiplicative modi¯cation for the non-zero values. We present a case studywhere this new methodology is applied.Key words: count data, multiplicative replacement, composition, log-ratio analysis Geologische Vereinigung; I...
PURPOSE Computed tomography (CT) can detect thrombi in arteries after acute stroke and can provid... more PURPOSE Computed tomography (CT) can detect thrombi in arteries after acute stroke and can provide a measure of their composition based on Hounsfield units (HU). Thromboembolic stroke can be caused by platelet-rich, erythrocyte-rich, or mixed clots. Platelet-rich thrombi are resistant to thrombolytic drugs, but erythrocyte-rich thrombi are not. Experimental studies have shown lower HU in platelet-rich thrombi than in erythrocyte-rich thrombi. We investigated whether HU values can predict the lysability of thrombi. METHOD AND MATERIALS We performed thin-section noncontrast CT and CT angiographic studies before IV rt-PA in 30 patients (18 men; mean age,67±10years) with thrombi in the middle cerebral artery within 4.5 hours of stroke onset. HU values were measured in the thrombosed and contralateral normal segments and calculated the ratios (rHU) between them. Recanalization was determined by MR angiography after IV rt-PA. We used linear discriminant analysis to find the cutoff value o...
The linkage among the tissue iron stores, insulin resistance (IR), and cognition remains unclear ... more The linkage among the tissue iron stores, insulin resistance (IR), and cognition remains unclear in the obese population. We aimed to identify the factors that contribute to increased hepatic iron concentration (HIC) and brain iron overload (BIO), as evaluated by MRI, and to evaluate their impact on cognitive performance in obese and nonobese subjects. We prospectively recruited 23 middle-aged obese subjects without diabetes (13 women; age 50.4 ± 7.7 years; BMI 43.7 ± 4.48 kg/m2) and 20 healthy nonobese volunteers (10 women; age 48.8 ± 9.5 years; BMI 24.3 ± 3.54 kg/m2) in whom iron load was assessed in white and gray matter and the liver by MRI. IR was measured from HOMA-IR and an oral glucose tolerance test. A battery of neuropsychological tests was used to evaluate the cognitive performance. Multivariate regression analysis was used to identify the independent associations of BIO and cognitive performance. A significant increase in iron load was detected at the caudate nucleus (P < 0.001), lenticular nucleus (P = 0.004), hypothalamus (P = 0.002), hippocampus (P < 0.001), and liver (P < 0.001) in obese subjects. There was a positive correlation between HIC and BIO at caudate (r = 0.517, P < 0.001), hypothalamus (r = 0.396, P = 0.009), and hippocampus (r = 0.347, P < 0.023). The area under the curve of insulin was independently associated with BIO at the caudate (P = 0.001), hippocampus (P = 0.028), and HIC (P = 0.025). BIOs at the caudate (P = 0.028), hypothalamus (P = 0.006), and lenticular nucleus (P = 0.012) were independently associated with worse cognitive performance. Obesity and IR may contribute to increased HIC and BIO being associated with worse cognitive performance. BIO could be a potentially useful MRI biomarker for IR and obesity-associated cognitive dysfunction.
In this chapter, we will use compositional data analysis (CODA) to document the geometric variati... more In this chapter, we will use compositional data analysis (CODA) to document the geometric variation of limb proportions in ternary morphospaces and in linear bivariate spaces. This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.
Uploads
Papers by Pepus Daunis-I-Estadella
This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of
proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb
segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences
among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.
This chapter will reanalyse the data of Dyke et al. (2006) and McGowan and Dyke (2007) using CODA (Aitchison 1986), specifically designed to deal with the statistical properties of
proportions. CODA is appropriate for studying the evolution of flight mechanics because the functional properties of wings and hindlimbs can be expressed as the proportion of one limb
segment to another. The limb element lengths of the specimens used by Dyke et al. (2006) and McGowan and Dyke (2007) have been used to infer biomechanic similarities and differences
among three flying vertebrate groups, namely birds (Aves), pterosaurs and bats. McGowan and Dyke (2007) proposed there was competitive exclusion between extinct and living flying vertebrates. Dyke et al. (2006) attempted to determine whether the extinct pterosaur flew in a ‘bird-like’ mode (only forelimb involved) or in a ‘bat-like’ mode (both fore- and hindlimbs). Dyke et al. (2006) used an individual of Sordes pilosus – one of the few pterosaur specimens with a preserved flight membrane – as a model, in order to contrast each pterosaur flight paradigm.