DSpace Collection:
http://hdl.handle.net/2117/3406
Fri, 18 Apr 2014 03:25:10 GMT2014-04-18T03:25:10Zwebmaster.bupc@upc.eduUniversitat Politècnica de Catalunya. Servei de Biblioteques i DocumentaciónoOn left-truncating and mixing poisson distributions
http://hdl.handle.net/2117/22548
Title: On left-truncating and mixing poisson distributions
Authors: Valero Baya, Jordi; Pérez Casany, Marta; Ginebra Molins, Josep
Abstract: The distributions obtained by left-truncating at k a mixed Poisson distribution, kT-MP, and those obtained by mixing previously left-truncated Poisson distributions, M-kTP, are characterized by means of their probability generating function. The main consequence is that every kT-MP distribution is a M-kTP distribution, but not the other way around.Mon, 07 Apr 2014 17:26:00 GMThttp://hdl.handle.net/2117/225482014-04-07T17:26:00ZValero Baya, Jordi; Pérez Casany, Marta; Ginebra Molins, JosepnoCount variable: Left-truncated distribution, Mixed distribution, Model characterization, Probability generating functionThe distributions obtained by left-truncating at k a mixed Poisson distribution, kT-MP, and those obtained by mixing previously left-truncated Poisson distributions, M-kTP, are characterized by means of their probability generating function. The main consequence is that every kT-MP distribution is a M-kTP distribution, but not the other way around.A competitive variable neighbourhood search algorithm for the blocking flow shop problem
http://hdl.handle.net/2117/22388
Title: A competitive variable neighbourhood search algorithm for the blocking flow shop problem
Authors: Ribas Vila, Immaculada; Companys Pascual, Ramón; Tort-Martorell Llabrés, XavierWed, 26 Mar 2014 12:46:44 GMThttp://hdl.handle.net/2117/223882014-03-26T12:46:44ZRibas Vila, Immaculada; Companys Pascual, Ramón; Tort-Martorell Llabrés, XaviernoLight consulting after design of experiments training: Lessons from a case study
http://hdl.handle.net/2117/22376
Title: Light consulting after design of experiments training: Lessons from a case study
Authors: Marco Almagro, Lluís; Grima Cintas, Pedro; Tort-Martorell Llabrés, Xavier
Abstract: We present a case study to show some positive and some problematic points regarding light consulting—when none of the parts is fully committed—on Design of Experiments (DOE). The situation is not unusual and frequently occurs after a DOE training course. In fact, the case used as an illustration occurred when advising a team of engineers from an aeronautical industrial company some months after giving them a 32-h DOE seminar. A key point is that, in spite of all the inconveniences and pitfalls, it may be a good alternative for the two parts, especially nowadays with all the internet facilities. The case study included in the paper serves as an excuse to give a set of recommendations on how to develop this light consulting approach the best possible way.Tue, 25 Mar 2014 12:22:54 GMThttp://hdl.handle.net/2117/223762014-03-25T12:22:54ZMarco Almagro, Lluís; Grima Cintas, Pedro; Tort-Martorell Llabrés, Xaviernodesign of experiments, variation reduction, statistical consultingWe present a case study to show some positive and some problematic points regarding light consulting—when none of the parts is fully committed—on Design of Experiments (DOE). The situation is not unusual and frequently occurs after a DOE training course. In fact, the case used as an illustration occurred when advising a team of engineers from an aeronautical industrial company some months after giving them a 32-h DOE seminar. A key point is that, in spite of all the inconveniences and pitfalls, it may be a good alternative for the two parts, especially nowadays with all the internet facilities. The case study included in the paper serves as an excuse to give a set of recommendations on how to develop this light consulting approach the best possible way.¿Es pi un impostor?
http://hdl.handle.net/2117/22351
Title: ¿Es pi un impostor?
Authors: Grima Cintas, Pedro
Abstract: Parece que sobre p está todo dicho y que su papel protagonista en el reino de los números es indiscutible pero… ¿son posibles unas matemáticas en las que el número pi no tenga un papel protagonista, en las que 3,14159… suene a número raro que no sugiere nada? Este artículo pretende poner de manifiesto que sí lo son, que el número pi no tiene la genuina especificidad de otros números singulares. Pero tranquilos, son las matemáticas de siempre, solo se trata repasar los terrenos en que aparece pi y descubrir que seguramente sería más justo que el número singular fuera otro.Mon, 24 Mar 2014 11:48:36 GMThttp://hdl.handle.net/2117/223512014-03-24T11:48:36ZGrima Cintas, PedronoBachillerato, divulgación, fórmulas, Números, piParece que sobre p está todo dicho y que su papel protagonista en el reino de los números es indiscutible pero… ¿son posibles unas matemáticas en las que el número pi no tenga un papel protagonista, en las que 3,14159… suene a número raro que no sugiere nada? Este artículo pretende poner de manifiesto que sí lo son, que el número pi no tiene la genuina especificidad de otros números singulares. Pero tranquilos, son las matemáticas de siempre, solo se trata repasar los terrenos en que aparece pi y descubrir que seguramente sería más justo que el número singular fuera otro.¿Es pi un impostor?
http://hdl.handle.net/2117/22351
Title: ¿Es pi un impostor?
Authors: Grima Cintas, Pedro
Abstract: Parece que sobre p está todo dicho y que su papel protagonista en el reino de los números es indiscutible pero… ¿son posibles unas matemáticas en las que el número pi no tenga un papel protagonista, en las que 3,14159… suene a número raro que no sugiere nada? Este artículo pretende poner de manifiesto que sí lo son, que el número pi no tiene la genuina especificidad de otros números singulares. Pero tranquilos, son las matemáticas de siempre, solo se trata repasar los terrenos en que aparece pi y descubrir que seguramente sería más justo que el número singular fuera otro.Mon, 24 Mar 2014 11:48:36 GMThttp://hdl.handle.net/2117/223512014-03-24T11:48:36ZGrima Cintas, PedronoBachillerato, divulgación, fórmulas, Números, piParece que sobre p está todo dicho y que su papel protagonista en el reino de los números es indiscutible pero… ¿son posibles unas matemáticas en las que el número pi no tenga un papel protagonista, en las que 3,14159… suene a número raro que no sugiere nada? Este artículo pretende poner de manifiesto que sí lo son, que el número pi no tiene la genuina especificidad de otros números singulares. Pero tranquilos, son las matemáticas de siempre, solo se trata repasar los terrenos en que aparece pi y descubrir que seguramente sería más justo que el número singular fuera otro.The time has come: Statistics in bestselling books
http://hdl.handle.net/2117/21021
Title: The time has come: Statistics in bestselling books
Authors: Grima Cintas, Pedro; Marco Almagro, Lluís
Abstract: Beyond textbooks, statistics is also present in bestselling books, those
that appear on the top 10 lists of bookshops and online bookstores. This
paper discusses five of those books, highlighting the role of statistics in
each one. Besides describing the general topics of the books, we want to
show that the knowledge of the world around us – and also the knowledge
on ourselves – advances thanks to the application of the scientific method
of which statistics is a key element. The paper finishes with some thoughts
on the desirability of a practical approach to teaching statisticsTue, 17 Dec 2013 11:40:27 GMThttp://hdl.handle.net/2117/210212013-12-17T11:40:27ZGrima Cintas, Pedro; Marco Almagro, LluísnoBeyond textbooks, statistics is also present in bestselling books, those
that appear on the top 10 lists of bookshops and online bookstores. This
paper discusses five of those books, highlighting the role of statistics in
each one. Besides describing the general topics of the books, we want to
show that the knowledge of the world around us – and also the knowledge
on ourselves – advances thanks to the application of the scientific method
of which statistics is a key element. The paper finishes with some thoughts
on the desirability of a practical approach to teaching statisticsA Cluster analysis of vote transitions
http://hdl.handle.net/2117/20843
Title: A Cluster analysis of vote transitions
Authors: Puig Oriol, Xavier; Ginebra Molins, Josep
Abstract: To help settle the debate triggered the day after any election around the origin and destination of the vote of winners and losers, a Bayesian analysis of the results in a pair of consecutive elections is proposed. It is based on a model that simultaneously carries out a cluster analysis of the areas in which the results are broken into and links the results in the two elections of areas in a given cluster through a vote switch matrix. The number of clusters is chosen both through predictive checks as well as by testing whether the residuals are spatially correlated or not. The analysis is tried on the results in Barcelona of a pair of consecutive elections held just four months apart, in 2003 for the Catalan parliament and in 2004 for the Spanish parliament. The proposed approach, which reconstructs individual behavior from aggregated data, can be exported to be a solution for any ecological inference problem where one cannot assume that all the areas are exchangeable the way typically assumed by other ecological inference methods.
Description: Bayesian model checking;
Bayesian hierarchical model;
Ecological inference;
Election data;
Spatial dataFri, 29 Nov 2013 08:44:33 GMThttp://hdl.handle.net/2117/208432013-11-29T08:44:33ZPuig Oriol, Xavier; Ginebra Molins, JosepnoBayesian model checking, Bayesian hierarchical model, Ecological inference, Election data, Spatial dataTo help settle the debate triggered the day after any election around the origin and destination of the vote of winners and losers, a Bayesian analysis of the results in a pair of consecutive elections is proposed. It is based on a model that simultaneously carries out a cluster analysis of the areas in which the results are broken into and links the results in the two elections of areas in a given cluster through a vote switch matrix. The number of clusters is chosen both through predictive checks as well as by testing whether the residuals are spatially correlated or not. The analysis is tried on the results in Barcelona of a pair of consecutive elections held just four months apart, in 2003 for the Catalan parliament and in 2004 for the Spanish parliament. The proposed approach, which reconstructs individual behavior from aggregated data, can be exported to be a solution for any ecological inference problem where one cannot assume that all the areas are exchangeable the way typically assumed by other ecological inference methods.On poisson-stopped-sums that are mixed poisson
http://hdl.handle.net/2117/19966
Title: On poisson-stopped-sums that are mixed poisson
Authors: Valero Baya, Jordi; Pérez Casany, Marta; Ginebra Molins, Josep
Abstract: Maceda (1948) characterized the mixed Poisson distributions that are Poisson-stopped-sum distributions based on the mixing distribution. In an alternative characterization of the same set of distributions here the Poisson-stopped-sum distributions that are mixed Poisson distributions is proved to be the set of Poisson-stopped-sums of either a mixture of zero-truncated Poisson distributions or a zero-modification of it.Wed, 17 Jul 2013 10:39:07 GMThttp://hdl.handle.net/2117/199662013-07-17T10:39:07ZValero Baya, Jordi; Pérez Casany, Marta; Ginebra Molins, JosepnoPoisson-stopped-sum distribution, Poisson mixture, Zero-truncated distribution, Zero-modified distributionMaceda (1948) characterized the mixed Poisson distributions that are Poisson-stopped-sum distributions based on the mixing distribution. In an alternative characterization of the same set of distributions here the Poisson-stopped-sum distributions that are mixed Poisson distributions is proved to be the set of Poisson-stopped-sums of either a mixture of zero-truncated Poisson distributions or a zero-modification of it.An efficient iterated local search algorithm for the total tardiness blocking flow shop problem
http://hdl.handle.net/2117/19622
Title: An efficient iterated local search algorithm for the total tardiness blocking flow shop problem
Authors: Ribas Vila, Immaculada; Companys Pascual, ramon; Tort-Martorell Llabrés, Xavier
Abstract: This paper deals with the blocking flow shop problem and proposes an Iterated Local Search (ILS) procedure combined with a variable neighbourhood search (VNS) for the total tardiness minimisation. The proposed ILS makes use of a NEH-based procedure to generate the initial solution, and uses a local search to intensify the exploration that combines the insertion and swap neighbourhood and uses a perturbation mechanism consisting of three neighbourhood operators to diversify the search. The computational evaluation has shown the effectiveness of combining the insertion and swap neighbourhood during the search despite the insertion neighbourhood being more effective than the swap neighbourhood for this problem. Finally, the computation of this algorithm when evaluated against two other algorithms from the literature shows good performance.Tue, 25 Jun 2013 07:30:25 GMThttp://hdl.handle.net/2117/196222013-06-25T07:30:25ZRibas Vila, Immaculada; Companys Pascual, ramon; Tort-Martorell Llabrés, Xaviernoblocking flow shop, scheduling, tardiness, iterated local search, variable local search, heuristicsThis paper deals with the blocking flow shop problem and proposes an Iterated Local Search (ILS) procedure combined with a variable neighbourhood search (VNS) for the total tardiness minimisation. The proposed ILS makes use of a NEH-based procedure to generate the initial solution, and uses a local search to intensify the exploration that combines the insertion and swap neighbourhood and uses a perturbation mechanism consisting of three neighbourhood operators to diversify the search. The computational evaluation has shown the effectiveness of combining the insertion and swap neighbourhood during the search despite the insertion neighbourhood being more effective than the swap neighbourhood for this problem. Finally, the computation of this algorithm when evaluated against two other algorithms from the literature shows good performance.Twenty-five analogies for explaining statistical concepts
http://hdl.handle.net/2117/18719
Title: Twenty-five analogies for explaining statistical concepts
Authors: Behar, Roberto; Grima Cintas, Pedro; Marco Almagro, Lluís
Abstract: The use of analogies is a resource that can be used for transmitting concepts and making classes more enjoyable. This article presents 25 analogies that we use in our introductory statistical courses for introducing concepts and clarifying possible doubts. We have found that these analogies draw students’ attention and reinforce the ideas that we want to transmit.Tue, 09 Apr 2013 09:54:11 GMThttp://hdl.handle.net/2117/187192013-04-09T09:54:11ZBehar, Roberto; Grima Cintas, Pedro; Marco Almagro, LluísnoThe use of analogies is a resource that can be used for transmitting concepts and making classes more enjoyable. This article presents 25 analogies that we use in our introductory statistical courses for introducing concepts and clarifying possible doubts. We have found that these analogies draw students’ attention and reinforce the ideas that we want to transmit.An approach to disaggregating total household water consumption into major end-uses
http://hdl.handle.net/2117/17873
Title: An approach to disaggregating total household water consumption into major end-uses
Authors: Fontdecaba Rigat, Sara; Sánchez Espigares, Josep Anton; Marco Almagro, Lluís; Tort-Martorell Llabrés, Xavier; Cabrespina, Francesc; Zubelzu, Jordi
Abstract: The aim of this project is to assign domestic water consumption to different
devices based on the information provided by the water meter. We monitored a sample of
Barcelona and Murcia with flow switches that recorded when a particular device was in use.
In addition, the water meter readings were recorded every 5 and 1 s, respectively, in
Barcelona and Murcia. The initial work used Barcelona data, and the method was later
verified and adjusted with the Murcia data. The proposed method employs an algorithm that
characterizes the water consumption of each device, using Barcelona to establish the initial parameters which, afterwards, provide information for adjusting the parameters of each household studied. Once the parameters have been adjusted, the algorithm assigns the consumption to each device. The efficacy of the assignation process is summarized in terms of: sensitivity and specificity. The algorithm provides a correct identification rate of between 70 % and 80 %; sometimes even higher, depending on how well the chosen parameters reflect household consumption patterns. Considering the high variability of the patterns and the fact that use is characterized by only the aggregate consumption that the water meter provides, the results are quite satisfactory.Tue, 19 Feb 2013 11:37:54 GMThttp://hdl.handle.net/2117/178732013-02-19T11:37:54ZFontdecaba Rigat, Sara; Sánchez Espigares, Josep Anton; Marco Almagro, Lluís; Tort-Martorell Llabrés, Xavier; Cabrespina, Francesc; Zubelzu, JordinoThe aim of this project is to assign domestic water consumption to different
devices based on the information provided by the water meter. We monitored a sample of
Barcelona and Murcia with flow switches that recorded when a particular device was in use.
In addition, the water meter readings were recorded every 5 and 1 s, respectively, in
Barcelona and Murcia. The initial work used Barcelona data, and the method was later
verified and adjusted with the Murcia data. The proposed method employs an algorithm that
characterizes the water consumption of each device, using Barcelona to establish the initial parameters which, afterwards, provide information for adjusting the parameters of each household studied. Once the parameters have been adjusted, the algorithm assigns the consumption to each device. The efficacy of the assignation process is summarized in terms of: sensitivity and specificity. The algorithm provides a correct identification rate of between 70 % and 80 %; sometimes even higher, depending on how well the chosen parameters reflect household consumption patterns. Considering the high variability of the patterns and the fact that use is characterized by only the aggregate consumption that the water meter provides, the results are quite satisfactory.Statistical methods in Kansei engineering: a case of statistical engineering
http://hdl.handle.net/2117/16784
Title: Statistical methods in Kansei engineering: a case of statistical engineering
Authors: Marco Almagro, Lluís; Tort-Martorell Llabrés, Xavier
Abstract: Kansei engineering (KE) is a methodology used to incorporate emotions in products and services design. Its basic purpose is
discovering in which way some properties of a product or a service convey certain emotions in its users. Data are typically
collected using questionnaires. KE studies follow a model with three main steps: (i) defining the elicited emotions (semantic
space); (ii) deciding on the factors that might affect the responses (space of properties); and (iii) modeling how each factor
affects each response (synthesis phase). The procedure resembles that of an experimental design in an industrial context.
However, practitioners of KE are hardly ever statisticians. Statistical techniques in KE are sometimes misused, and the
discipline could benefit from a more extensive use of statistical methods. KE is thus a good area of application of statistical
engineering: focusing not in advancement of statistics but on how current techniques can be best used in a new area.
The aim of this paper is twofold: (i) to present the fundamentals of KE while giving an easy to understand example to
illustrate the procedure; and (ii) to explain why KE is a good example of statistical engineering by proposing improvements
that emanate from the adequate use of statistical techniques.Tue, 23 Oct 2012 11:35:05 GMThttp://hdl.handle.net/2117/167842012-10-23T11:35:05ZMarco Almagro, Lluís; Tort-Martorell Llabrés, XaviernoKansei engineering (KE) is a methodology used to incorporate emotions in products and services design. Its basic purpose is
discovering in which way some properties of a product or a service convey certain emotions in its users. Data are typically
collected using questionnaires. KE studies follow a model with three main steps: (i) defining the elicited emotions (semantic
space); (ii) deciding on the factors that might affect the responses (space of properties); and (iii) modeling how each factor
affects each response (synthesis phase). The procedure resembles that of an experimental design in an industrial context.
However, practitioners of KE are hardly ever statisticians. Statistical techniques in KE are sometimes misused, and the
discipline could benefit from a more extensive use of statistical methods. KE is thus a good area of application of statistical
engineering: focusing not in advancement of statistics but on how current techniques can be best used in a new area.
The aim of this paper is twofold: (i) to present the fundamentals of KE while giving an easy to understand example to
illustrate the procedure; and (ii) to explain why KE is a good example of statistical engineering by proposing improvements
that emanate from the adequate use of statistical techniques.Time series outlier and intervention analysis: Irrigation management influences on soil water content in silty loam soil
http://hdl.handle.net/2117/16683
Title: Time series outlier and intervention analysis: Irrigation management influences on soil water content in silty loam soil
Authors: Basem, Aljoumani; Sánchez Espigares, Josep Anton; Cañameras Riba, Núria; Josa March, Ramon; Monserrat, Joaquim
Abstract: Understanding the field soil water regime is fundamental in scheduling irrigation as well as for monitoring water flow and solute transport. This study was carried out on variable interval irrigation and used time series analysis techniques to predict the soil water content at the interested depth by measuring one single depth in order to precisely determine the next irrigation time and its effect on soil water content at the interested depth. Volumetric water content of silty loam soil in Barcelona was measured in situ with capacitance soil moisture sensors at five depths within the root zone for a horticultural crop during its life cycle in 2010. The time series consisted of hourly measurements of soil water content and was transformed to a stationary situation. Subsequently, the transformed data were used to conduct analyses in the time domain in order to obtain the parameters of a seasonal autoregressive integrated moving average (ARIMA) model. In the case of variable interval irrigation, predicting the soil water content time series cannot be properly explained by the ARIMA model and its underlying normality assumption. By completing the ARIMA model with intervention analysis and outlier detection, the prediction of soil water content in variable interval irrigation can be made. The transfer function models were then used to predict water contents at depths of interest (0.20, 0.35, 0.50 and 0.60 m depths) as well as the average water content WAVG in the top 0.60 m soil profile by measuring water content at 0.10 m depth. As a result, the predictions were logical. Also, the next irrigation time and its effect on soil water content at the depth of interest were correctly estimated. To confirm results of the models, the experiment was repeated in 2011, and the predicted and observed values agree reasonably well.
Description: Soil volumetric water content;
Autoregressive integrated moving average (ARIMA);
Outlier detection;
Transfer function modelTue, 09 Oct 2012 11:01:57 GMThttp://hdl.handle.net/2117/166832012-10-09T11:01:57ZBasem, Aljoumani; Sánchez Espigares, Josep Anton; Cañameras Riba, Núria; Josa March, Ramon; Monserrat, JoaquimnoAutoregressive integrated moving average (ARIMA), Outlier detection, Soil volumetric water content, Transfer function modelUnderstanding the field soil water regime is fundamental in scheduling irrigation as well as for monitoring water flow and solute transport. This study was carried out on variable interval irrigation and used time series analysis techniques to predict the soil water content at the interested depth by measuring one single depth in order to precisely determine the next irrigation time and its effect on soil water content at the interested depth. Volumetric water content of silty loam soil in Barcelona was measured in situ with capacitance soil moisture sensors at five depths within the root zone for a horticultural crop during its life cycle in 2010. The time series consisted of hourly measurements of soil water content and was transformed to a stationary situation. Subsequently, the transformed data were used to conduct analyses in the time domain in order to obtain the parameters of a seasonal autoregressive integrated moving average (ARIMA) model. In the case of variable interval irrigation, predicting the soil water content time series cannot be properly explained by the ARIMA model and its underlying normality assumption. By completing the ARIMA model with intervention analysis and outlier detection, the prediction of soil water content in variable interval irrigation can be made. The transfer function models were then used to predict water contents at depths of interest (0.20, 0.35, 0.50 and 0.60 m depths) as well as the average water content WAVG in the top 0.60 m soil profile by measuring water content at 0.10 m depth. As a result, the predictions were logical. Also, the next irrigation time and its effect on soil water content at the depth of interest were correctly estimated. To confirm results of the models, the experiment was repeated in 2011, and the predicted and observed values agree reasonably well.Weeds, aphids, and specialist parasitoids and predators benefeit differently from organic and conventional cropping of winter cereals
http://hdl.handle.net/2117/14953
Title: Weeds, aphids, and specialist parasitoids and predators benefeit differently from organic and conventional cropping of winter cereals
Authors: Caballero-López, Berta; Blanco-Moreno, José M.; Pérez-Hidalgo, Nicolás; Michelena-Saval, José M.; Pujade-Villar, Juli; Guerrieri, Emilio; Sánchez Espigares, Josep Anton; Sans, F. Xavier
Abstract: The aphid–natural enemy interaction in winter
wheat fields constitutes a complex system that has been
frequently studied because of its implication for biological
control. However, not all of the aphids living in cereal
fields are crop pests, as there are also aphids living on
weeds that may serve as alternative hosts or prey for aphid
parasitoids or predators. In this context, a concomitant
survey of the plant and insect communities was conducted
to understand how different plant communities affect the
abundance and richness of aphids and the interactions with
their natural enemies. The plant community was split into
functional groups (grasses, legumes and forbs), and the
aphid community was divided into feeding groups
according to their host preferences (specialists in grasses or
forbs). The grass aphids, which dominated the total aphid
catches, responded positively to grass cover, which was
particularly enhanced in the conventional fields. Conversely,
the forb aphids, which mainly conditioned the total
species richness of the aphids, were closely correlated with
the local abundance of legumes. The system of cereal
aphid-parasitoids was enhanced in the conventional fields,
where the abundance of grasses was higher, whereas the
legumes of the organic fields indirectly played a key role in
enhancing the richness of the parasitoids and the abundance
of predators. Our findings indicate that a bottom-up
effect exists throughout the plant community, aphids, and
aphidophagous insects and that plant community characteristics
shouldMon, 06 Feb 2012 11:58:17 GMThttp://hdl.handle.net/2117/149532012-02-06T11:58:17ZCaballero-López, Berta; Blanco-Moreno, José M.; Pérez-Hidalgo, Nicolás; Michelena-Saval, José M.; Pujade-Villar, Juli; Guerrieri, Emilio; Sánchez Espigares, Josep Anton; Sans, F. XaviernoThe aphid–natural enemy interaction in winter
wheat fields constitutes a complex system that has been
frequently studied because of its implication for biological
control. However, not all of the aphids living in cereal
fields are crop pests, as there are also aphids living on
weeds that may serve as alternative hosts or prey for aphid
parasitoids or predators. In this context, a concomitant
survey of the plant and insect communities was conducted
to understand how different plant communities affect the
abundance and richness of aphids and the interactions with
their natural enemies. The plant community was split into
functional groups (grasses, legumes and forbs), and the
aphid community was divided into feeding groups
according to their host preferences (specialists in grasses or
forbs). The grass aphids, which dominated the total aphid
catches, responded positively to grass cover, which was
particularly enhanced in the conventional fields. Conversely,
the forb aphids, which mainly conditioned the total
species richness of the aphids, were closely correlated with
the local abundance of legumes. The system of cereal
aphid-parasitoids was enhanced in the conventional fields,
where the abundance of grasses was higher, whereas the
legumes of the organic fields indirectly played a key role in
enhancing the richness of the parasitoids and the abundance
of predators. Our findings indicate that a bottom-up
effect exists throughout the plant community, aphids, and
aphidophagous insects and that plant community characteristics
shouldBayesian analysis of frequency count data
http://hdl.handle.net/2117/14798
Title: Bayesian analysis of frequency count data
Authors: Font Valverde, Martí; Puig Oriol, Xavier; Ginebra Molins, Josep
Abstract: The zero truncated inverse Gaussian–Poisson model, obtained by first mixing the Poisson model assuming its expected value has an inverse Gaussian distribution and then truncating the model at zero, is very useful when modelling frequency count data. A Bayesian analysis based on this statistical model is implemented on the word frequency counts of various texts, and its validity is checked by exploring the posterior distribution of the Pearson errors and by implementing posterior predictive consistency checks. The analysis based on this model is useful because it allows one to use the posterior distribution of the model mixing density as an approximation of the posterior distribution of the density of the word frequencies of the vocabulary of the author, which is useful to characterize the style of that author. The posterior distribution of the expectation and of measures of the variability of that mixing distribution can be used to assess the size and diversity of his vocabulary. An alternative analysis is proposed based on the inverse Gaussian-zero truncated Poisson mixture model, which is obtained by switching the order of the mixing and the truncation stages. Even though this second model fits some of the word frequency data sets more accurately than the first model, in practice the analysis based on it is not as useful because it does not allow one to estimate the word frequency distribution of the vocabulary.Wed, 25 Jan 2012 12:11:32 GMThttp://hdl.handle.net/2117/147982012-01-25T12:11:32ZFont Valverde, Martí; Puig Oriol, Xavier; Ginebra Molins, JosepnoThe zero truncated inverse Gaussian–Poisson model, obtained by first mixing the Poisson model assuming its expected value has an inverse Gaussian distribution and then truncating the model at zero, is very useful when modelling frequency count data. A Bayesian analysis based on this statistical model is implemented on the word frequency counts of various texts, and its validity is checked by exploring the posterior distribution of the Pearson errors and by implementing posterior predictive consistency checks. The analysis based on this model is useful because it allows one to use the posterior distribution of the model mixing density as an approximation of the posterior distribution of the density of the word frequencies of the vocabulary of the author, which is useful to characterize the style of that author. The posterior distribution of the expectation and of measures of the variability of that mixing distribution can be used to assess the size and diversity of his vocabulary. An alternative analysis is proposed based on the inverse Gaussian-zero truncated Poisson mixture model, which is obtained by switching the order of the mixing and the truncation stages. Even though this second model fits some of the word frequency data sets more accurately than the first model, in practice the analysis based on it is not as useful because it does not allow one to estimate the word frequency distribution of the vocabulary.