scholarly journals Spatial analysis of two aquatic invaders in Adirondack Lakes: a modelling approach for environmental management

2021 ◽  
Author(s):  
Valerie Elaine Bowler
Keyword(s):  
Invasive Species ◽  
Logistic Regression ◽  
Predictive Models ◽  
Global Analysis ◽  
New York State ◽  
Natural World ◽  
Ordinary Least Squares ◽  
Adirondack Park ◽  
Eurasian Watermilfoil ◽  
Curly Leaf

The global expansion of humans has stressed the natural world, removed boundaries between continents and habitats and exposed natural areas to invasive species. These cause billions of dollars of damage yet there are limited funds given for their management. Predictive tools can be used to develop pro-active strategies for managing invasive species and this study developed such a tool. Publicly available data were used to build predictive models for the presence of two invasive species, curly-leaf pondweed (Potamogeton crispus) and Eurasian watermilfoil (Myriophyllum spicatum) within the Adirondack Park (New York State). Predictors were identified through: bivariate analysis to test the variables; ordinary least squares regression to build predictive models and logistic regression to validate those models; geographically weighted logistic regression to evaluate local impacts. Models were ranked by Aikake information criterion minimization and evaluated with McFadden’s rho-squared, standard coefficients and variance inflation factors. The top five models for each invasive species established seven predictors for curly-leaf pondweed and nine predictors for Eurasian watermilfoil. Geographically weighted regression, a local analysis, was found to be a definite improvement over the global analysis for watermilfoil but not for pondweed. Two predictors (lake elevation and distance to Interstate-87) were significant in all the top models for both species. The identified predictors provided a group of characteristics that could be used to identify vulnerable lakes and prioritize management strategies. Even though these findings were specific to the Adirondack Park, this approach could be applied to other invasive species or other areas to help in the decision-making process for management.

scholarly journals Spatial analysis of two aquatic invaders in Adirondack Lakes: a modelling approach for environmental management

2021 ◽  
Author(s):  
Valerie Elaine Bowler
Keyword(s):  
Invasive Species ◽  
Logistic Regression ◽  
Predictive Models ◽  
Global Analysis ◽  
New York State ◽  
Natural World ◽  
Ordinary Least Squares ◽  
Adirondack Park ◽  
Eurasian Watermilfoil ◽  
Curly Leaf

The global expansion of humans has stressed the natural world, removed boundaries between continents and habitats and exposed natural areas to invasive species. These cause billions of dollars of damage yet there are limited funds given for their management. Predictive tools can be used to develop pro-active strategies for managing invasive species and this study developed such a tool. Publicly available data were used to build predictive models for the presence of two invasive species, curly-leaf pondweed (Potamogeton crispus) and Eurasian watermilfoil (Myriophyllum spicatum) within the Adirondack Park (New York State). Predictors were identified through: bivariate analysis to test the variables; ordinary least squares regression to build predictive models and logistic regression to validate those models; geographically weighted logistic regression to evaluate local impacts. Models were ranked by Aikake information criterion minimization and evaluated with McFadden’s rho-squared, standard coefficients and variance inflation factors. The top five models for each invasive species established seven predictors for curly-leaf pondweed and nine predictors for Eurasian watermilfoil. Geographically weighted regression, a local analysis, was found to be a definite improvement over the global analysis for watermilfoil but not for pondweed. Two predictors (lake elevation and distance to Interstate-87) were significant in all the top models for both species. The identified predictors provided a group of characteristics that could be used to identify vulnerable lakes and prioritize management strategies. Even though these findings were specific to the Adirondack Park, this approach could be applied to other invasive species or other areas to help in the decision-making process for management.

scholarly journals Predicting Eurasian Watermilfoil and Curly-leaf Pondweed Presence-Absence in Adirondack Lakes Using Geographically Weighted Logistic Regression

2021 ◽  
Author(s):  
Jelena Grbic
Keyword(s):  
New York ◽  
Logistic Regression ◽  
Model Performance ◽  
Adirondack Park ◽  
Eurasian Watermilfoil ◽  
Weighted Logistic Regression ◽  
Adirondack Lakes ◽  
Geographically Weighted Logistic Regression ◽  
Forest Preserve ◽  
Curly Leaf

Aquatic invasive species, Eurasian Watermilfoil (EWM) and Curly-leaf Pondweed (CLP), have been dispersing across New York, USA and are threatening the ecosystem of Adirondack Park, a state park with a large forest preserve and heavily frequented by tourists. In this study, the prediction of EWM and CLP invasion across Adirondack Park lakes is modeled using logistic regression (LR) and geographically weighted logistic regression (GWLR) with lake, landscape, and climate variable predictors. EWM presence-absence is found to be best predicted by nearby invaded lakes, human presence, and elevation. The presence-absence of CLP models have similar findings, with the addition of game-fish abundance being important. GWLR increases model performance and prediction, with explained variation of EWM and CLP increasing by 23% and 16% and the percent correctly predicted increasing by 2.6% and 0.9%. The study shows that GWLR, a relatively novel methodology, works better than common LR models for predicting invasion of EWM and CLP across Adirondack Park, and corroborates anthropogenic influences on dispersal of aquatic invaders.

scholarly journals Predicting Eurasian Watermilfoil and Curly-leaf Pondweed Presence-Absence in Adirondack Lakes Using Geographically Weighted Logistic Regression

2021 ◽  
Author(s):  
Jelena Grbic
Keyword(s):  
New York ◽  
Logistic Regression ◽  
Model Performance ◽  
Adirondack Park ◽  
Eurasian Watermilfoil ◽  
Weighted Logistic Regression ◽  
Adirondack Lakes ◽  
Geographically Weighted Logistic Regression ◽  
Forest Preserve ◽  
Curly Leaf

Aquatic invasive species, Eurasian Watermilfoil (EWM) and Curly-leaf Pondweed (CLP), have been dispersing across New York, USA and are threatening the ecosystem of Adirondack Park, a state park with a large forest preserve and heavily frequented by tourists. In this study, the prediction of EWM and CLP invasion across Adirondack Park lakes is modeled using logistic regression (LR) and geographically weighted logistic regression (GWLR) with lake, landscape, and climate variable predictors. EWM presence-absence is found to be best predicted by nearby invaded lakes, human presence, and elevation. The presence-absence of CLP models have similar findings, with the addition of game-fish abundance being important. GWLR increases model performance and prediction, with explained variation of EWM and CLP increasing by 23% and 16% and the percent correctly predicted increasing by 2.6% and 0.9%. The study shows that GWLR, a relatively novel methodology, works better than common LR models for predicting invasion of EWM and CLP across Adirondack Park, and corroborates anthropogenic influences on dispersal of aquatic invaders.

Assessing natural disaster survivor evacuation attitudes to inform social policy

2014 ◽  
Vol 34 (7/8) ◽  
pp. 485-510 ◽  
Author(s):  
Kenneth David Strang
Keyword(s):  
Logistic Regression ◽  
Social Policy ◽  
Natural Disaster ◽  
Online Survey ◽  
New York State ◽  
Ordinary Least Squares ◽  
Near Miss ◽  
Test Theory ◽  
Least Squares Regression ◽  
Content Type

Purpose – The literature was reviewed to locate the most relevant social-psychology theories, factors, and instruments in order to measure New York State resident attitudes and social norms (SNs) concerning their intent to evacuate Hurricane Irene in the summer of 2011. The purpose of this paper is to develop a model which could be generalized to improve social policy determination for natural disaster preparation. Design/methodology/approach – A post-positivist ideology was employed, quantitative data were collected from an online survey (nominal, binary, interval, and ratio), and inferential statistical techniques were applied to test theory-deductive hypotheses (Strang, 2013b). Since the questions for each hypothesized factor were customized using a pilot for this study, exploratory factor analysis were conducted to ensure the item validity and reliabilities were compared to a priori benchmarks (Gill et al., 2010). Correlation analysis along with logistic and multiple regression were applied to test the hypothesis at the 95 percent confidence level. Findings – A statistically significant model was developed using correlation, stepwise regression, ordinary least squares regression, and logistic regression. Only two composite factors were needed to capture 55.4 percent of the variance for behavioral intent (BI) to evacuate. The model predicted 43.9 percent of the evacuation decisions, with 13.3 percent undecided, leaving 42.8 incorrectly classified), using logistic regression (n=401 surveyed participants). Research limitations/implications – Municipal planners can use this information by creating surveys and collecting BI indicators from citizens, during risk planning, in advance of a natural disaster. The concepts could also apply to man-made disasters. Planners can use the results from these surveys to predict the overall likelihood that residents with home equity (e.g. home owners) intend to leave when given a public evacuation order. Practical implications – Once municipal planners know the indicators for personal attitudes (PAs) (in particular) and SNs, they could sort these by region, to identify areas where the PAs were too low. Then additional evacuation preparation efforts can be focussed on those regions. According to these findings, the emphasis must be focussed on a PA basis, describing the extreme negative impacts of previous disasters, rather than using credible spokespersons, to persuade individuals to leave. Originality/value – A new model was created with a “near miss disaster” severity factor as an extension to the theory of reasoned action.

scholarly journals Least Likely Observations in Regression Models for Categorical Outcomes

2002 ◽  
Vol 2 (3) ◽  
pp. 296-300 ◽  
Author(s):  
Jeremy Freese
Keyword(s):  
Logistic Regression ◽  
Regression Models ◽  
Binary Logistic Regression ◽  
Positive Outcome ◽  
Ordinary Least Squares ◽  
Parameter Estimates ◽  
Least Squares Regression ◽  
Binary Logistic Regression Model ◽  
Ordered Logistic Regression ◽  
Negative Outcome

This article presents a method and program for identifying poorly fitting observations for maximum-likelihood regression models for categorical dependent variables. After estimating a model, the program leastlikely will list the observations that have the lowest predicted probabilities of observing the value of the outcome category that was actually observed. For example, when run after estimating a binary logistic regression model, leastlikely will list the observations with a positive outcome that had the lowest predicted probabilities of a positive outcome and the observations with a negative outcome that had the lowest predicted probabilities of a negative outcome. These can be considered the observations in which the outcome is most surprising given the values of the independent variables and the parameter estimates and, like observations with large residuals in ordinary least squares regression, may warrant individual inspection. Use of the program is illustrated with examples using binary and ordered logistic regression.

scholarly journals Clinical response scores to predict blood pressure and renal function improvement following renal artery stenting for atherosclerotic renal artery stenosis

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
A Kablak-Ziembicka ◽  
A Roslawiecka ◽  
R Badacz ◽  
A Sokolowski ◽  
P Musialek ◽  
...  
Keyword(s):  
Blood Pressure ◽  
Logistic Regression ◽  
Renal Function ◽  
Renal Artery ◽  
Renal Artery Stenosis ◽  
Predictive Models ◽  
Artery Stenosis ◽  
Atherosclerotic Renal Artery Stenosis ◽  
Baseline Serum ◽  
Kidney Length

Abstract Background It is little known about predictors of systolic (SBP) and diastolic (DBP) blood pressure or renal function (eGFR) improvement in patients with atherosclerotic renal artery stenosis (ARAS) undergoing stent-assisted angioplasty (PTA). Therefore, we aimed to build a prediction scores that would indicate characteristics of patient subsets with ARAS most likely to have clinical improvement following PTA. Methods 201 patients who underwent PTA for ARAS (2003–2018) were categorized as eGFR or SBP/DBP responders based on eGFR increase of ≥11 ml/min/1.73m2, decrease of SBP ≥20mmHg and DBP ≥5mmHg at 12-months following PTA. The remaining patients were classified as non-responders. The performance of logistic regression models were evaluated by basic decision characteristics. Continuous data have been transformed into binary coding with help of operating characteristic (ROC) curve. Predictive models have been constructed for each followed by construction of predictive models in each of 3 categories. Results Logistic regression analysis showed that: baseline SBP>145 mmHg, DBP >82 mmHg, previous myocardial infarction and Renal-Aotric-Ratio >5.1 were independent influencing factors of SBP response, with relative risk percentage shares of 69.8%; 12.1%; 10.9%; and 7.2%, respectively (sensitivity: 82%, specificity: 86.3%, positive (PPV):82% and negative (NPV) predictive values: 86.3%). The DBP decrease prediction model included baseline SBP >145 mmHg and DBP >82 mmHg, the ARAS progression, index kidney length >106 mm, and bilateral PTA with respective shares of 35.0%; 21.8%; 18.2%; 13.3% and 11.8%. (sensitivity: 76%, specificity: 77.8%, PPV: 80.7% and NPV: 72.6%). The eGFR increase was associated with baseline serum creatinine >122 μmol/L but eGFR greater than 30 ml/min/1.73m2, index kidney length >98 mm, end-diastolic velocity in index renal artery, renal resistive index <0.74, and requirement for >3 BP medications, with respective shares of 24.4%; 24.4%; 21.2%; 15% and 15% (sensitivity: 33.3%, specificity: 93.5%, PPV: 65.6% and NPV: 78.9%). Conclusions Current study identified clinical characteristics of patients who most likely to respond to PTA for ARAS. The sutability of the score should be verified in a prospective cohort of patients referred to PTA of ARAS Funding Acknowledgement Type of funding source: None

The Euphoric Sensations of Cigarettes: Exploring How Operant Conditioning May Be the Missing Link in the Theoretical Understanding of Antidrug Policies

2018 ◽  
Vol 48 (3) ◽  
pp. 485-502
Author(s):  
Elizabeth L. Borkowski ◽  
Wanda E. Leal
Keyword(s):  
Logistic Regression ◽  
Operant Conditioning ◽  
Smoking Behavior ◽  
Add Health ◽  
Ordinary Least Squares ◽  
Cigarette Use ◽  
Theoretical Understanding ◽  
Negative Experiences ◽  
Drug Policies ◽  
Positive Experiences

This study aims to examine how positive and negative reinforcers during an individual’s first few cigarettes (cigarette initiation experiences) are associated with adulthood smoking behavior. Respondents from the Add Health were asked about subjective feelings during their first few cigarettes. Using ordinary least squares (OLS) and logistic regression, we examine the differential effects of positive and negative cigarette initiation experiences on 30-day cigarette use in adulthood and lifetime nicotine dependence. The results indicate that all measures of positive cigarette initiation experiences are positively associated with both cigarette measures; however, the opposite is not true of negative cigarette initiation experiences. The results highlight the misconceptions of antidrug policies aimed at punishment of users, by indicating that positive experiences influence later cigarette use more than negative experiences. These findings suggest that drug policies and initiatives aimed at punishment may be misguided and could benefit from adopting operant conditioning concepts that emphasize reinforcements.

Χρησιμοποίηση υπηρεσιών υγείας στην Ελλάδα

2009 ◽  
Author(s):  
Δημήτρης Ζάβρας
Keyword(s):  
Logistic Regression ◽  
Least Squares ◽  
Ordinary Least Squares

Σύμφωνα με τη διεθνή βιβλιογραφία, η επιλογή μεταξύ υπηρεσιών υγείας του ιδιωτικού και δημόσιου τομέα, ερμηνεύεται κυρίως από το κοινωνικοοικονομικό επίπεδο. Σκοπός της συγκεκριμένης διατριβής είναι ο προσδιορισμός των παραγόντων οι οποίοι επηρεάζουν τις επιλογές των χρηστών ως προς τον τύπο και την δομή της φροντίδας υγείας που θα χρησιμοποιήσουν. Για τον σκοπό της διατριβής χρησιμοποιήθηκαν δεδομένα από πανελλαδική έρευνα, η οποία πραγματοποιήθηκε με τη μέθοδο των προσωπικών συνεντεύξεων σε τυχαίο δείγμα 4003 ενηλίκων ατόμων το οποίο στρωματοποιήθηκε ανά γεωγραφική περιφέρεια, βαθμό αστικότητας τόπου διαμονής, ηλικία και φύλο.Η διατριβή εστίασε σε δύο διαφορετικές ομάδες χρηστών ανάλογα με τις εναλλακτικές υπηρεσίες υγείας που είχαν στη διάθεσή τους: α) ασφαλισμένους του Ι.Κ.Α οι οποίοι είναι σε θέση να χρησιμοποιήσουν εκτός από τις δημόσιες και ιδιωτικές υπηρεσίες υγείας και υπηρεσίες υγείας του Ι.Κ.Α και β) ασφαλισμένους σε ασφαλιστικούς οργανισμούς πλην Ι.Κ.Α, οι οποίοι είναι σε θέση να χρησιμοποιήσουν μόνο υπηρεσίες υγείας του ιδιωτικού και δημόσιου τομέα. Οι μεταβλητές οι οποίες αφορούσαν στη χρήση εναλλακτικών υπηρεσιών υγείας εκφράστηκαν ως ποσοστά επί της συνολικής χρησιμοποίησης. Τα συγκεκριμένα ποσοστά έπαιρναν τιμές στο διάστημα [0, 1] και είχαν άθροισμα ίσο με 1. Από τις συγκεκριμένες μεταβλητές προέκυψαν με κωδικοποίηση ονομαστικές μεταβλητές, οι κατηγορίες των οποίων εξέφραζαν τη χρήση των εναλλακτικών υπηρεσιών υγείας. Πιο συγκεκριμένα, η τιμή 0 των προαναφερόμενων ποσοστών αντιστοιχούσε σε μηδενική χρήση υπηρεσιών υγείας κάθε τομέα παροχής υπηρεσιών υγείας, η τιμή 1 αντιστοιχούσε σε αποκλειστική χρήση υπηρεσιών υγείας κάθε τομέα παροχής υπηρεσιών υγείας και τιμές στο διάστημα [0.01-0.99] αντιστοιχούσε σε συνδυαστική χρήση υπηρεσιών υγείας κάθε τομέα παροχής υπηρεσιών υγείας. Οι μεταβλητές της ανάλυσης των δεδομένων επομένως ήταν: α) οι ονομαστικές μεταβλητές οι κατηγορίες των οποίων εξέφραζαν τη μηδενική, αποκλειστική ή συνδυαστική χρήση των εναλλακτικών υπηρεσιών υγείας και β) τα ποσοστά με τιμές στο διάστημα (0, 1) οι οποίες εξέφραζαν το βαθμό συνδυαστικής χρήσης των υπηρεσιών υγείας κάθε τομέα παροχής υπηρεσιών υγείας. Επιπρόσθετα, η ανάλυση εστίασε σε δύο ομάδες χρηστών: α) χρήστες με συνολική χρησιμοποίηση μεγαλύτερη του 1 και β) χρήστες με συνολική χρησιμοποίηση ίση με 1. Για την ανάλυση των ονομαστικών μεταβλητών της μελέτης χρησιμοποιήθηκε η μέθοδος Individualized Logistic Regression, ενώ για την ανάλυση των ποσοστών χρησιμοποιήθηκε η μέθοδος Ordinary Least Squares. Ως ανεξάρτητες μεταβλητές της ανάλυσης χρησιμοποιήθηκαν οι μεταβλητές: α) γεωγραφική περιφέρεια, β) βαθμός αστικότητας, γ) ηλικία, δ) φύλο, ε) οικογενειακή κατάσταση, στ) εισόδημα, ζ) απασχόληση, η) εκπαίδευση, θ) αυτοαξιολογούμενο επίπεδο υγείας, ι) ύπαρξη χρόνιου προβλήματος υγείας, κ) δημόσια ασφαλιστική κάλυψη, λ) ιδιωτική ασφαλιστική κάλυψη. Επιπρόσθετα για την ομάδα χρηστών με συνολική χρησιμοποίηση μεγαλύτερη της μονάδας, η συγκεκριμένη μεταβλητή χρησιμοποιήθηκε ως ανεξάρτητη μεταβλητή.Βασικά συμπεράσματα της διδακτορικής διατριβής αποτελούν η επίδραση του κοινωνικοοικονομικού επιπέδου στην επιλογή μεταξύ υπηρεσιών υγείας του δημόσιου και ιδιωτικού τομέα καθώς και το γεγονός ότι άτομα με υψηλή συνολική χρησιμοποίηση έχουν μεγαλύτερη πιθανότητα να κάνουν συνδυαστική χρήση εναλλακτικών υπηρεσιών υγείας. Συμπερασματικά θα μπορούσε να ειπωθεί ότι η επιλογή μεταξύ εναλλακτικών υπηρεσιών υγείας προσδιορίζεται τόσο από τα χαρακτηριστικά των χρηστών και ιδίως το κοινωνικοοικονομικό επίπεδο αλλά και από την συνολική χρησιμοποίηση υπηρεσιών υγείας.

Landscape Influences on Stream Habitats and Biological Assemblages

2006 ◽  
Keyword(s):  
Neural Network ◽  
New York ◽  
Linear Regression ◽  
Multiple Linear Regression ◽  
Predictive Models ◽  
New York State ◽  
Species Abundance ◽  
Dorosoma Cepedianum ◽  
Linear Discriminant ◽  
Potential Applications

<em>Abstract.</em>—We describe a methodology for developing species–habitat models using available fish and stream habitat data from New York State, focusing on the Genesee basin. Electrofishing data from the New York Department of Environmental Conservation were standardized and used for model development and testing. Four types of predictive models (multiple linear regression, stepwise multiple linear regression, linear discriminant analysis, and neural network) were developed and compared for 11 fish species. Predictive models used as many as 25 habitat variables and explained 35–91% of observed species abundance variability. Omission rates were generally low, but commission rates varied widely. Neural network models performed best for all species, except for rainbow trout <em>Oncorhynchus mykiss</em>, gizzard shad <em>Dorosoma cepedianum</em>, and brown trout <em>Salmo trutta</em>. Linear discriminant functions generally performed poorly. The species–environment models we constructed performed well and have potential applications to management issues.

Sign in / Sign up

Export Citation Format

Share Document