Simple Linear Models With Polytomous Categorical Dependent Variables: Multinomial and Ordinal Logistic Regression

2013 ◽  
Vol 35 (1) ◽  
pp. 98
Author(s):  
Angela Radünz Lazzari

Air pollution is a risk factor for the population health. Its harmful effects on the population are observed even when the atmospheric pollutants are within the parameters set out in specific legislation, and they develop mainly through respiratory diseases. The aim of this study was to analyze the relationship between the concentrations of air pollutants and the incidence of respiratory diseases in the city of Porto Alegre, in 2005 and 2006. Applied multiple linear regression analysis, ordinal logistic regression and generalized linear models were used in the work. The results show good adjustment by the three techniques. The ordinal logistic regression detected only positive influence of air temperature and relative humidity in hospitalizations for respiratory diseases. Multiple linear regression related negatively hospitalizations with meteorological variables and positively with the particulate matter (PM10). The generalized linear model detected negative influence of meteorological variables and positive of pollutants, tropospheric ozone (O3) and PM10 in hospitalizations. Comparing the three statistical techniques to analyze the same data set, it can be concluded that all of them had a model with good fit to the data, but the technique of generalized linear models showed higher sensitivity in capturing the influence of pollutants, except in ordinal logistic regression and multiple linear regression.


2017 ◽  
Vol 12 (2) ◽  
pp. 243-264 ◽  
Author(s):  
Rahul Kumar ◽  
Pradip Kumar Bala

Purpose Collaborative filtering (CF), one of the most popular recommendation techniques, is based on the principle of word-of-mouth communication between other like-minded users. The process of identifying these like-minded or similar users remains crucial for a CF framework. Conventionally, a neighbor is the one among the similar users who has rated the item under consideration. To select neighbors by the existing practices, their similarity deteriorates as many similar users might not have rated the item under consideration. This paper aims to address the drawback in the existing CF method where “not-so-similar” or “weak” neighbors are selected. Design/methodology/approach The new approach proposed here selects neighbors only on the basis of highest similarity coefficient, irrespective of rating the item under consideration. Further, to predict missing ratings by some neighbors for the item under consideration, ordinal logistic regression based on item–item similarity is used here. Findings Experiments using the MovieLens (ml-100) data set prove the efficacy of the proposed approach on different performance evaluation metrics such as accuracy and classification metrics. Apart from higher prediction quality, coverage values are also at par with the literature. Originality/value This new approach gets its motivation from the principle of the CF method to rely on the opinion of the closest neighbors, which seems more meaningful than trusting “not-so-similar” or “weak” neighbors. The static nature of the neighborhood addresses the scalability issue of CF. Use of ordinal logistic regression as a prediction technique addresses the statistical inappropriateness of other linear models to make predictions for ordinal scale ratings data.


2020 ◽  
Vol 4 (Supplement_1) ◽  
pp. 414-414
Author(s):  
Anna Huang ◽  
Kristen Wroblewski ◽  
Ashwin Kotwal ◽  
Linda Waite ◽  
Martha McClintock ◽  
...  

Abstract The classical senses (vision, hearing, touch, taste, and smell) play a key role in social function by allowing interaction and communication. We assessed whether sensory impairment across all 5 modalities (global sensory impairment [GSI]) was associated with social function in older adults. Sensory function was measured in 3,005 home-dwelling older U.S. adults at baseline in the National Social Life, Health, and Aging Project and GSI, a validated measure, was calculated. Social network size and kin composition, number of close friends, and social engagement were assessed at baseline and 5- and 10-year follow-up. Ordinal logistic regression and mixed effects ordinal logistic regression analyzed cross-sectional and longitudinal relationships respectively, controlling for demographics, physical/mental health, disability, and cognitive function (at baseline). Adults with worse GSI had smaller networks (β=-0.159, p=0.021), fewer close friends (β=-0.262, p=0.003) and lower engagement (β=-0.252, p=0.006) at baseline, relationships that persisted at 5 and 10 year follow-up. Men, older people, African-Americans, and those with less education, fewer assets, poor mental health, worse cognitive function, and more disability had worse GSI. Men and those with fewer assets, worse cognitive function, and less education had smaller networks and lower engagement. African-American and Hispanic individuals had smaller networks and fewer close friends, but more engagement. Older respondents also had more engagement. In summary, GSI independently predicts smaller social networks, fewer close friends, and lower social engagement over time, suggesting that sensory decline results in decreased social function. Thus, rehabilitating multisensory impairment may be a strategy to enhance social function as people age.


2021 ◽  
pp. 1-11
Author(s):  
Guilian Wang ◽  
Liyan Zhang ◽  
Jing Guo

This paper try to fully reveal the key factors affecting the the level of AMT application in micro- and small enterprises (MSEs) from its organizational factors by ordinal logistic regression. The results show that MSEs have a relatively high level of AMT application as a whole due to the maturity and cost reduction of basic technologies such as artificial intelligence, digital manufacturing and industrial robots. In this paper we propose manufacturing world analysis at Application using Logistic Regression and best AMT selection using Fuzzy-TOPSIS Integration approach.Considering the influence mechanism of each factor, the important factors that affect the application level of AMT are the enterprise’s market pricing power, the main production types, technical, market and management capabilities, organization development incentives and the interaction with external stakeholders. Based on the results above, the following policy implications are proposed: further expanding the customized production in MSEs to gradually improve the market pricing power, expanding the core competence of enterprises, enhancing the employee autonomy, and strengthening the interaction with industry organizations.


2007 ◽  
Vol 86 (9) ◽  
pp. 852-856 ◽  
Author(s):  
M.T. John ◽  
W. Micheelis ◽  
J.G. Steele

Depression is associated with impaired health outcomes. This study investigated whether there is a significant association between depression and dissatisfaction with dentures in older adults. In a population-based study (1180 adults aged 65–74 yrs), depression was measured by an abbreviated Geriatric Depression Scale. Denture dissatisfaction was assessed with a five-point Likert-type question ("very dissatisfied" to "very satisfied"). The depression-denture dissatisfaction association was analyzed with simple (dissatisfied vs. not dissatisfied outcome) and ordinal logistic regression (based on outcome’s full range). For each unit increase on the 15-point depression scale, the probability of denture dissatisfaction increased by 24% [95% confidence interval, 15–34%, P < 0.001 (simple logistic regression)] and the probability for higher levels on the five-point dissatisfaction scale increased by 16% [95% CI, 11–22%, P < 0.001 (ordinal logistic regression)], adjusted for potential confounding variables. The likely causal association in older adults has major implications for the evaluation of treatment effects and the demand for prosthodontic therapy.


Sign in / Sign up

Export Citation Format

Share Document