There is a large number of data archives and web services offering free access to multispectral s... more There is a large number of data archives and web services offering free access to multispectral satellite imagery. Images from multiple sources are increasingly combined to improve the spatio-temporal coverage of measurements while achieving more accurate results. Archives and web services differ in their protocols, formats, and data standards, which are barriers to combine datasets. Here, we present RGISTools, an R package to create time-series of multispectral satellite images from multiple platforms in a harmonized and standardized way. We first provide an overview of the package functionalities, namely downloading, customizing, and processing multispectral satellite imagery for a region and time period of interest as well as a recent statistical method for gap-filling and smoothing series of images, called interpolation of the mean anomalies. We further show the capabilities of the package through a case study that combines Landsat-8 and Sentinel-2 satellite optical imagery to estimate the level of a water reservoir in Northern Spain. We expect RGISTools to foster research on data fusion and spatio-temporal modelling using satellite images from multiple programs.
• Extrapolating cancer mortality trends can be very valuable as a tool to predict cancer burden. ... more • Extrapolating cancer mortality trends can be very valuable as a tool to predict cancer burden. National Health Agencies use different models to figure out future evolution of cancer, but they mainly work at national level. However, developed countries are divided into different regions with their own governments and health care systems, and this should be taken into account. In this paper, an ANOVA-type P-spline model is considered to predict the number of mortality cases in forthcoming years in regions within a country. The model is very interesting as it allows to split the predictions into components representing region-specific features and characteristics common to the whole country. Prediction variability is also calculated to provide prediction intervals. Real data on cancer mortality are used for illustration.
In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in de... more In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in developed countries is breast cancer, where mortality increased sharply until the nineties and then decreased. This clear unimodal pattern is not necessarily generalizable to all regions within a country. In this work, we develop statistical tests to check if this unimodality persists within regions using order restricted inference. The same methodology will be also used to provide change-points within regions as well as confidence intervals. Results will be illustrated using age-specific breast cancer mortality data from Spain in the period 1975-2005.
The combination of freely accessible satellite imagery from multiple programs improves the spatio... more The combination of freely accessible satellite imagery from multiple programs improves the spatio-temporal coverage of remote sensing data, but it exhibits barriers regarding the variety of web services, file formats, and data standards. Ris an open-source software environment with state-of-the-art statistical packages for the analysis of optical imagery. However, it lacks the tools for providing unified access to multi-program archives to customize and process the time series of images. This manuscript introduces RGISTools, a new software that solves these issues, and provides a working example on water mapping, which is a socially and environmentally relevant research field. The case study uses a digital elevation model and a rarely assessed combination of Landsat-8 and Sentinel-2 imagery to determine the water level of a reservoir in Northern Spain. The case study demonstrates how to acquire and process time series of surface reflectance data in an efficient manner. Our method ac...
Satellite remote sensing data have become available in meteorology, agriculture, forestry, geolog... more Satellite remote sensing data have become available in meteorology, agriculture, forestry, geology, regional planning, hydrology or natural environment sciences since several decades ago, because satellites provide routinely high quality images with different temporal and spatial resolutions. Joining, combining or smoothing these images for a better quality of information is a challenge not always properly solved. In this regard, geostatistics, as the spatio-temporal stochastic techniques of geo-referenced data, is a very helpful and powerful tool not enough explored in this area yet. Here, we analyze the current use of some of the geostatistical tools in satellite image analysis, and provide an introduction to this subject for potential researchers.
Detecting change-points and trends are common tasks in the analysis of remote sensing data. Over ... more Detecting change-points and trends are common tasks in the analysis of remote sensing data. Over the years, many different methods have been proposed for those purposes, including (modified) Mann–Kendall and Cox–Stuart tests for detecting trends; and Pettitt, Buishand range, Buishand U, standard normal homogeneity (Snh), Meanvar, structure change (Strucchange), breaks for additive season and trend (BFAST), and hierarchical divisive (E.divisive) for detecting change-points. In this paper, we describe a simulation study based on including different artificial, abrupt changes at different time-periods of image time series to assess the performances of such methods. The power of the test, type I error probability, and mean absolute error (MAE) were used as performance criteria, although MAE was only calculated for change-point detection methods. The study reveals that if the magnitude of change (or trend slope) is high, and/or the change does not occur in the first or last time-periods,...
IEEE Transactions on Geoscience and Remote Sensing, 2019
When monitoring time series of remote sensing data, it is advisable to fill gaps, i.e., missing o... more When monitoring time series of remote sensing data, it is advisable to fill gaps, i.e., missing or distorted data, caused by atmospheric effects or technical failures. In this paper, a new method for filling these gaps called interpolation of the mean anomalies (IMA) is proposed and compared with some competitors. The method consists of: 1) defining a neighborhood for the target image from previous and subsequent images across previous and subsequent years; 2) computing the mean target image of the neighborhood; 3) estimating the anomalies in the target image by subtracting the mean image from the target image; 4) filtering the anomalies; 5) averaging the anomalies over a predefined window; 6) interpolating the averaged anomalies; and 7) adding the interpolated anomalies to the mean image. To assess the performance of the IMA method, both a real example and a simulation study are conducted with a time series of Moderate Resolution Imaging Spectroradiometer (MODIS) TERRA and MODIS AQUA images captured over the region of Navarre (Spain) from 2011 to 2013. We analyze the land surface temperature (LST) day and night, and the normalized difference vegetation index (NDVI). In the simulation study, seven sizes of artificial clouds are randomly introduced to each image in the studied time series. The square root of the mean-squared prediction error (RMSE) between the original and the filled data is chosen as an indicator of the goodness of fit. The results show that the IMA method outperforms Timesat, Hants, and Gapfill (GF) in filling small, moderate, and big cloud gaps in both the day and night LST and NDVI data, reaching RMSE reductions of up to 23%.
To predict accumulated daily rainfall in a particular location where no rain gauges are available... more To predict accumulated daily rainfall in a particular location where no rain gauges are available is important for agriculture, meteorology, traffic networks, environment and many other areas. However, statistical modelling of rain is not trivial because a high variability is presented within and between days. In this work we analyze the performance of alternative spatio-temporal models. The sampled data consist of daily observations taken in 87 manual rainfall gauges during the 1990-2010 period in Navarre, Spain. The accuracy and precision of the interpolated data is checked with data of 28 automated rainfall non-sampled gauges of the same region but placed in different locations than the manual rainfall gauges. Interpolations will be mapped on a squared grid of 1km 2 grid over the whole study region and to assess the prediction performance of the models a continuous ranked probability score (CRPS) has also been calculated.
In this paper, space-time patterns of colorectal cancer (CRC) mortality risks are studied by sex ... more In this paper, space-time patterns of colorectal cancer (CRC) mortality risks are studied by sex and age group (50-69, ≥70) in Spanish provinces during the period 1975-2008. Space-time conditional autoregressive models are used to perform the statistical analyses. A pronounced increase in mortality risk has been observed in males for both age-groups. For males between 50 and 69 years of age, trends seem to stabilize from 2001 onward. In females, trends reflect a more stable pattern during the period in both age groups. However, for the 50-69 years group, risks take an upward trend in the period 2006-2008 after the slight decline observed in the second half of the period. This study offers interesting information regarding CRC mortality distribution among different Spanish provinces that could be used to improve prevention policies and resource allocation in different regions.
Spatio-temporal disease mapping comprises a wide range of models used to describe the distributio... more Spatio-temporal disease mapping comprises a wide range of models used to describe the distribution of a disease in space and its evolution in time. These models have been commonly formulated within a hierarchical Bayesian framework with two main approaches: an empirical Bayes (EB) and a fully Bayes (FB) approach. The EB approach provides point estimates of the parameters relying on the well-known penalized quasi-likelihood (PQL) technique. The FB approach provides the posterior distribution of the target parameters. These marginal distributions are not usually available in closed form and common estimation procedures are based on Markov chain Monte Carlo (MCMC) methods. However, the spatio-temporal models used in disease mapping are often very complex and MCMC methods may lead to large Monte Carlo errors and a huge computation time if the dimension of the data at hand is large. To circumvent these potential inconveniences, a new technique called integrated nested Laplace approximati...
The results of the four trials to determine the effects of different defoliation treatments on ga... more The results of the four trials to determine the effects of different defoliation treatments on garlic yield, carried out in the Central Ebro Valley (Spain), are presented. Four defoliation levels of 0 (control), 33, 66 and 100% were applied at seven different developmental stages. The results demonstrate a close relationship between yield reduction and the defoliation treatment in¯icted. The higher the defoliation level, the higher the yield reduction. Defoliation imposed at the onset of bulb formation resulted in maximum yield reduction.
Hemos elaborado un estudio descriptivo de algunas de las afecciones respiratorias con mayor trasc... more Hemos elaborado un estudio descriptivo de algunas de las afecciones respiratorias con mayor trascendencia clínica en el niño, como son neumonías, bronquitis y bronquiolitis.
Background: Few studies on occupational mortality have been conducted in Spain. The objective of ... more Background: Few studies on occupational mortality have been conducted in Spain. The objective of this work was to analyse inequalities on global mortality and on mortality due to specific causes according to occupation in a historical cohort of males from the province of Navarra, Spain. Methods: The base population for this historical cohort comprised all employed men over age 34 from Navarra in the 1986 population register. Age-standardised point estimates and confidence intervals for occupational-specific mortality risks were computed. Results: There exist differences in mortality risks with respect to the overall risk of Navarra in certain occupational activities for several major causes of mortality. Some of the results corroborate previous findings in other works, such as the significant high risk that presents in leather, clothing workers and shoemakers when analysing kidney, bladder and other urinary malignant tumours, while others present a certain degree of novelty. Conclusion: This work contributes to filling the gap in the lack of works on occupational mortality in Spain. It also complements the information that other monitoring systems may provide on occupational health.
Stochastic Environmental Research and Risk Assessment, 2015
In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in de... more In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in developed countries is breast cancer, where mortality increased sharply until the nineties and then decreased. This clear unimodal pattern is not necessarily applicable to all regions within a country. In this paper, we develop statistical tools to check if the unimodality pattern persists within regions using order restricted inference. Break points as well as confidence intervals are also provided. In addition, a new test for checking monotonicity against unimodality is derived allowing to discriminate between a simple increasing pattern and an up-then-down response pattern. A comparison with the widely used joinpoint regression technique under unimodality is provided. We show that the joinpoint technique could fail when the underlying function is not piecewise linear. Results will be illustrated using age-specific breast cancer mortality data from Spain in the period 1975-2005.
The main goal of spatio-temporal disease mapping is describing the evolution of geographical patt... more The main goal of spatio-temporal disease mapping is describing the evolution of geographical patterns of mortality or incidence risks (rates). This could give clues to epidemiologists and public health researchers to formulate etiologic hypothesis of the disease. However, the ability of disease mapping models to make predictions about future mortality or incidence risks has not been widely explored. In this work, a flexible spatio-temporal model is considered for risk estimation and forecasting. The prediction MSE of both fitted and forecast values, as well as estimators of those quantities, will be derived. Spanish cancer mortality data will be used for illustration.
Accurate and precise knowledge about the distribution and evolution of a disease in space and tim... more Accurate and precise knowledge about the distribution and evolution of a disease in space and time is crucial to develop health policies and to help researchers to look into risk factors related to the disease. During the last years, the availability of modern computers has made it possible the development of statistical models and estimation techniques to analyze spatio-temporal data. Such spatio-temporal models have been used in disease mapping to study how a disease evolves in space throughout the years. It is common in practice to study age-standardized mortality or incidence risks or rates such that a single measure is provided for the whole region and all age groups. However, if the evolution of the disease is not the same among the age groups, age-specific rates within each region should be provided. In this work, several age-space-time models are considered and fitted to study the evolution of age-specific rates along time in different small areas. Spanish prostate cancer mo...
Stochastic Environmental Research and Risk Assessment, 2021
We propose an adaptive-sliding-window approach (LACPD) for the problem of change-point detection ... more We propose an adaptive-sliding-window approach (LACPD) for the problem of change-point detection in a set of time-ordered observations. The proposed method is combined with sub-sampling techniques to compensate for the lack of enough data near the time series’ tails. Through a simulation study, we analyse its behaviour in the presence of an early/middle/late change-point in the mean, and compare its performance with some of the frequently used and recently developed change-point detection methods in terms of power, type I error probability, area under the ROC curves (AUC), absolute bias, variance, and root-mean-square error (RMSE). We conclude that LACPD outperforms other methods by maintaining a low type I error probability. Unlike some other methods, the performance of LACPD does not depend on the time index of change-points, and it generally has lower bias than other alternative methods. Moreover, in terms of variance and RMSE, it outperforms other methods when change-points are ...
Stochastic Environmental Research and Risk Assessment, 2019
Outliers and missing data are commonly found in satellite imagery. These are usually caused by at... more Outliers and missing data are commonly found in satellite imagery. These are usually caused by atmospheric or electronic failures, hampering the correct monitoring of remote-sensing data. To avoid distorted data, we propose a procedure called "spatial functional prediction" (SFP). The SFP procedure consists of the following: 1) aggregating remote-sensing data for reducing the number of missing data and/or outliers; 2) additively decomposing the time series of images into a trend, a seasonal, and an error component; 3) defining the spatial functional data and predicting the trend component using an ordinary kriging; and 4) adding back the seasonal and error components to the predicted trend. The benefits of the SFP procedure are illustrated in the following scenarios: introducing random outliers, random missing data, mixtures of both, and artificial clouds in an extensive simulation study of composite images, and using daily images with real clouds. The following two derived variables are considered: land surface temperature (LST day) and normalized vegetation index (NDVI), which are obtained as remote-sensing data in a region in northern Spain during 2003-2016. The performance of SFP was checked using the root mean square error (RMSE). A comparison with a procedure based on predicting with thin-plate splines (TpsP) is also made. We conclude that SFP is simpler and faster than TpsP, and provides smaller values of RMSE.
There is a large number of data archives and web services offering free access to multispectral s... more There is a large number of data archives and web services offering free access to multispectral satellite imagery. Images from multiple sources are increasingly combined to improve the spatio-temporal coverage of measurements while achieving more accurate results. Archives and web services differ in their protocols, formats, and data standards, which are barriers to combine datasets. Here, we present RGISTools, an R package to create time-series of multispectral satellite images from multiple platforms in a harmonized and standardized way. We first provide an overview of the package functionalities, namely downloading, customizing, and processing multispectral satellite imagery for a region and time period of interest as well as a recent statistical method for gap-filling and smoothing series of images, called interpolation of the mean anomalies. We further show the capabilities of the package through a case study that combines Landsat-8 and Sentinel-2 satellite optical imagery to estimate the level of a water reservoir in Northern Spain. We expect RGISTools to foster research on data fusion and spatio-temporal modelling using satellite images from multiple programs.
• Extrapolating cancer mortality trends can be very valuable as a tool to predict cancer burden. ... more • Extrapolating cancer mortality trends can be very valuable as a tool to predict cancer burden. National Health Agencies use different models to figure out future evolution of cancer, but they mainly work at national level. However, developed countries are divided into different regions with their own governments and health care systems, and this should be taken into account. In this paper, an ANOVA-type P-spline model is considered to predict the number of mortality cases in forthcoming years in regions within a country. The model is very interesting as it allows to split the predictions into components representing region-specific features and characteristics common to the whole country. Prediction variability is also calculated to provide prediction intervals. Real data on cancer mortality are used for illustration.
In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in de... more In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in developed countries is breast cancer, where mortality increased sharply until the nineties and then decreased. This clear unimodal pattern is not necessarily generalizable to all regions within a country. In this work, we develop statistical tests to check if this unimodality persists within regions using order restricted inference. The same methodology will be also used to provide change-points within regions as well as confidence intervals. Results will be illustrated using age-specific breast cancer mortality data from Spain in the period 1975-2005.
The combination of freely accessible satellite imagery from multiple programs improves the spatio... more The combination of freely accessible satellite imagery from multiple programs improves the spatio-temporal coverage of remote sensing data, but it exhibits barriers regarding the variety of web services, file formats, and data standards. Ris an open-source software environment with state-of-the-art statistical packages for the analysis of optical imagery. However, it lacks the tools for providing unified access to multi-program archives to customize and process the time series of images. This manuscript introduces RGISTools, a new software that solves these issues, and provides a working example on water mapping, which is a socially and environmentally relevant research field. The case study uses a digital elevation model and a rarely assessed combination of Landsat-8 and Sentinel-2 imagery to determine the water level of a reservoir in Northern Spain. The case study demonstrates how to acquire and process time series of surface reflectance data in an efficient manner. Our method ac...
Satellite remote sensing data have become available in meteorology, agriculture, forestry, geolog... more Satellite remote sensing data have become available in meteorology, agriculture, forestry, geology, regional planning, hydrology or natural environment sciences since several decades ago, because satellites provide routinely high quality images with different temporal and spatial resolutions. Joining, combining or smoothing these images for a better quality of information is a challenge not always properly solved. In this regard, geostatistics, as the spatio-temporal stochastic techniques of geo-referenced data, is a very helpful and powerful tool not enough explored in this area yet. Here, we analyze the current use of some of the geostatistical tools in satellite image analysis, and provide an introduction to this subject for potential researchers.
Detecting change-points and trends are common tasks in the analysis of remote sensing data. Over ... more Detecting change-points and trends are common tasks in the analysis of remote sensing data. Over the years, many different methods have been proposed for those purposes, including (modified) Mann–Kendall and Cox–Stuart tests for detecting trends; and Pettitt, Buishand range, Buishand U, standard normal homogeneity (Snh), Meanvar, structure change (Strucchange), breaks for additive season and trend (BFAST), and hierarchical divisive (E.divisive) for detecting change-points. In this paper, we describe a simulation study based on including different artificial, abrupt changes at different time-periods of image time series to assess the performances of such methods. The power of the test, type I error probability, and mean absolute error (MAE) were used as performance criteria, although MAE was only calculated for change-point detection methods. The study reveals that if the magnitude of change (or trend slope) is high, and/or the change does not occur in the first or last time-periods,...
IEEE Transactions on Geoscience and Remote Sensing, 2019
When monitoring time series of remote sensing data, it is advisable to fill gaps, i.e., missing o... more When monitoring time series of remote sensing data, it is advisable to fill gaps, i.e., missing or distorted data, caused by atmospheric effects or technical failures. In this paper, a new method for filling these gaps called interpolation of the mean anomalies (IMA) is proposed and compared with some competitors. The method consists of: 1) defining a neighborhood for the target image from previous and subsequent images across previous and subsequent years; 2) computing the mean target image of the neighborhood; 3) estimating the anomalies in the target image by subtracting the mean image from the target image; 4) filtering the anomalies; 5) averaging the anomalies over a predefined window; 6) interpolating the averaged anomalies; and 7) adding the interpolated anomalies to the mean image. To assess the performance of the IMA method, both a real example and a simulation study are conducted with a time series of Moderate Resolution Imaging Spectroradiometer (MODIS) TERRA and MODIS AQUA images captured over the region of Navarre (Spain) from 2011 to 2013. We analyze the land surface temperature (LST) day and night, and the normalized difference vegetation index (NDVI). In the simulation study, seven sizes of artificial clouds are randomly introduced to each image in the studied time series. The square root of the mean-squared prediction error (RMSE) between the original and the filled data is chosen as an indicator of the goodness of fit. The results show that the IMA method outperforms Timesat, Hants, and Gapfill (GF) in filling small, moderate, and big cloud gaps in both the day and night LST and NDVI data, reaching RMSE reductions of up to 23%.
To predict accumulated daily rainfall in a particular location where no rain gauges are available... more To predict accumulated daily rainfall in a particular location where no rain gauges are available is important for agriculture, meteorology, traffic networks, environment and many other areas. However, statistical modelling of rain is not trivial because a high variability is presented within and between days. In this work we analyze the performance of alternative spatio-temporal models. The sampled data consist of daily observations taken in 87 manual rainfall gauges during the 1990-2010 period in Navarre, Spain. The accuracy and precision of the interpolated data is checked with data of 28 automated rainfall non-sampled gauges of the same region but placed in different locations than the manual rainfall gauges. Interpolations will be mapped on a squared grid of 1km 2 grid over the whole study region and to assess the prediction performance of the models a continuous ranked probability score (CRPS) has also been calculated.
In this paper, space-time patterns of colorectal cancer (CRC) mortality risks are studied by sex ... more In this paper, space-time patterns of colorectal cancer (CRC) mortality risks are studied by sex and age group (50-69, ≥70) in Spanish provinces during the period 1975-2008. Space-time conditional autoregressive models are used to perform the statistical analyses. A pronounced increase in mortality risk has been observed in males for both age-groups. For males between 50 and 69 years of age, trends seem to stabilize from 2001 onward. In females, trends reflect a more stable pattern during the period in both age groups. However, for the 50-69 years group, risks take an upward trend in the period 2006-2008 after the slight decline observed in the second half of the period. This study offers interesting information regarding CRC mortality distribution among different Spanish provinces that could be used to improve prevention policies and resource allocation in different regions.
Spatio-temporal disease mapping comprises a wide range of models used to describe the distributio... more Spatio-temporal disease mapping comprises a wide range of models used to describe the distribution of a disease in space and its evolution in time. These models have been commonly formulated within a hierarchical Bayesian framework with two main approaches: an empirical Bayes (EB) and a fully Bayes (FB) approach. The EB approach provides point estimates of the parameters relying on the well-known penalized quasi-likelihood (PQL) technique. The FB approach provides the posterior distribution of the target parameters. These marginal distributions are not usually available in closed form and common estimation procedures are based on Markov chain Monte Carlo (MCMC) methods. However, the spatio-temporal models used in disease mapping are often very complex and MCMC methods may lead to large Monte Carlo errors and a huge computation time if the dimension of the data at hand is large. To circumvent these potential inconveniences, a new technique called integrated nested Laplace approximati...
The results of the four trials to determine the effects of different defoliation treatments on ga... more The results of the four trials to determine the effects of different defoliation treatments on garlic yield, carried out in the Central Ebro Valley (Spain), are presented. Four defoliation levels of 0 (control), 33, 66 and 100% were applied at seven different developmental stages. The results demonstrate a close relationship between yield reduction and the defoliation treatment in¯icted. The higher the defoliation level, the higher the yield reduction. Defoliation imposed at the onset of bulb formation resulted in maximum yield reduction.
Hemos elaborado un estudio descriptivo de algunas de las afecciones respiratorias con mayor trasc... more Hemos elaborado un estudio descriptivo de algunas de las afecciones respiratorias con mayor trascendencia clínica en el niño, como son neumonías, bronquitis y bronquiolitis.
Background: Few studies on occupational mortality have been conducted in Spain. The objective of ... more Background: Few studies on occupational mortality have been conducted in Spain. The objective of this work was to analyse inequalities on global mortality and on mortality due to specific causes according to occupation in a historical cohort of males from the province of Navarra, Spain. Methods: The base population for this historical cohort comprised all employed men over age 34 from Navarra in the 1986 population register. Age-standardised point estimates and confidence intervals for occupational-specific mortality risks were computed. Results: There exist differences in mortality risks with respect to the overall risk of Navarra in certain occupational activities for several major causes of mortality. Some of the results corroborate previous findings in other works, such as the significant high risk that presents in leather, clothing workers and shoemakers when analysing kidney, bladder and other urinary malignant tumours, while others present a certain degree of novelty. Conclusion: This work contributes to filling the gap in the lack of works on occupational mortality in Spain. It also complements the information that other monitoring systems may provide on occupational health.
Stochastic Environmental Research and Risk Assessment, 2015
In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in de... more In some diseases it is well-known that a unimodal mortality pattern exists. A clear example in developed countries is breast cancer, where mortality increased sharply until the nineties and then decreased. This clear unimodal pattern is not necessarily applicable to all regions within a country. In this paper, we develop statistical tools to check if the unimodality pattern persists within regions using order restricted inference. Break points as well as confidence intervals are also provided. In addition, a new test for checking monotonicity against unimodality is derived allowing to discriminate between a simple increasing pattern and an up-then-down response pattern. A comparison with the widely used joinpoint regression technique under unimodality is provided. We show that the joinpoint technique could fail when the underlying function is not piecewise linear. Results will be illustrated using age-specific breast cancer mortality data from Spain in the period 1975-2005.
The main goal of spatio-temporal disease mapping is describing the evolution of geographical patt... more The main goal of spatio-temporal disease mapping is describing the evolution of geographical patterns of mortality or incidence risks (rates). This could give clues to epidemiologists and public health researchers to formulate etiologic hypothesis of the disease. However, the ability of disease mapping models to make predictions about future mortality or incidence risks has not been widely explored. In this work, a flexible spatio-temporal model is considered for risk estimation and forecasting. The prediction MSE of both fitted and forecast values, as well as estimators of those quantities, will be derived. Spanish cancer mortality data will be used for illustration.
Accurate and precise knowledge about the distribution and evolution of a disease in space and tim... more Accurate and precise knowledge about the distribution and evolution of a disease in space and time is crucial to develop health policies and to help researchers to look into risk factors related to the disease. During the last years, the availability of modern computers has made it possible the development of statistical models and estimation techniques to analyze spatio-temporal data. Such spatio-temporal models have been used in disease mapping to study how a disease evolves in space throughout the years. It is common in practice to study age-standardized mortality or incidence risks or rates such that a single measure is provided for the whole region and all age groups. However, if the evolution of the disease is not the same among the age groups, age-specific rates within each region should be provided. In this work, several age-space-time models are considered and fitted to study the evolution of age-specific rates along time in different small areas. Spanish prostate cancer mo...
Stochastic Environmental Research and Risk Assessment, 2021
We propose an adaptive-sliding-window approach (LACPD) for the problem of change-point detection ... more We propose an adaptive-sliding-window approach (LACPD) for the problem of change-point detection in a set of time-ordered observations. The proposed method is combined with sub-sampling techniques to compensate for the lack of enough data near the time series’ tails. Through a simulation study, we analyse its behaviour in the presence of an early/middle/late change-point in the mean, and compare its performance with some of the frequently used and recently developed change-point detection methods in terms of power, type I error probability, area under the ROC curves (AUC), absolute bias, variance, and root-mean-square error (RMSE). We conclude that LACPD outperforms other methods by maintaining a low type I error probability. Unlike some other methods, the performance of LACPD does not depend on the time index of change-points, and it generally has lower bias than other alternative methods. Moreover, in terms of variance and RMSE, it outperforms other methods when change-points are ...
Stochastic Environmental Research and Risk Assessment, 2019
Outliers and missing data are commonly found in satellite imagery. These are usually caused by at... more Outliers and missing data are commonly found in satellite imagery. These are usually caused by atmospheric or electronic failures, hampering the correct monitoring of remote-sensing data. To avoid distorted data, we propose a procedure called "spatial functional prediction" (SFP). The SFP procedure consists of the following: 1) aggregating remote-sensing data for reducing the number of missing data and/or outliers; 2) additively decomposing the time series of images into a trend, a seasonal, and an error component; 3) defining the spatial functional data and predicting the trend component using an ordinary kriging; and 4) adding back the seasonal and error components to the predicted trend. The benefits of the SFP procedure are illustrated in the following scenarios: introducing random outliers, random missing data, mixtures of both, and artificial clouds in an extensive simulation study of composite images, and using daily images with real clouds. The following two derived variables are considered: land surface temperature (LST day) and normalized vegetation index (NDVI), which are obtained as remote-sensing data in a region in northern Spain during 2003-2016. The performance of SFP was checked using the root mean square error (RMSE). A comparison with a procedure based on predicting with thin-plate splines (TpsP) is also made. We conclude that SFP is simpler and faster than TpsP, and provides smaller values of RMSE.
Uploads
Papers by Ana Fernandez