Impact of Influenza on Outpatient Visits, Hospitalizations, and Deaths by Using a Time Series Poisson Generalized Additive Model

The associations between ambient air pollutants and tuberculosis seasonality are unclear. We assessed the temporal cross-correlations between ambient air pollutants and tuberculosis seasonality. Monthly tuberculosis incidence data and ambient air pollutants (PM2.5, PM10, carbon monoxide (CO), nitrogen dioxide (NO2), ozone (O3), sulfur dioxide (SO2)) and air quality index (AQI) from 2013 to 2017 in Shanghai were included. A cross-correlogram and generalized additive model were used. A 4-month delayed effect of PM2.5 (0.55), PM10 (0.52), SO2 (0.47), NO2 (0.40), CO (0.39), and AQI (0.45), and a 6-month delayed effect of O3 (−0.38) on the incidence of tuberculosis were found. The number of tuberculosis cases increased by 8%, 4%, 18%, and 14% for a 10 μg/m3 increment in PM2.5, PM10, SO2, and NO2; 4% for a 10 unit increment in AQI; 8% for a 0.1 mg/m3 increment in CO; and decreased by 4% for a 10 μg/m3 increment in O3. PM2.5 concentrations above 50 μg/m3, 70 μg/m3 for PM10, 16 μg/m3 for SO2, 47 μg/m3 for NO2, 0.85 mg/m3 for CO, and 85 for AQI, and O3 concentrations lower than 95 μg/m3 were positively associated with the incidence of tuberculosis. Ambient air pollutants were correlated with tuberculosis seasonality. However, this sort of study cannot prove causality.

Download Full-text

GENERALIZED ADDITIVE MODEL FOR COUNT TIME SERIES: AN APPLICATION TO QUANTIFY THE IMPACT OF AIR POLLUTANTS ON HUMAN HEALTH

Pesquisa Operacional ◽

10.1590/0101-7438.2021.041.00241120 ◽

2021 ◽

Vol 41 ◽

Author(s):

Ana Julia A. Camara ◽

Glaura C. Franco ◽

Valderio A. Reisen ◽

Pascal Bondon

Keyword(s):

Time Series ◽

Human Health ◽

Air Pollutants ◽

Generalized Additive Model ◽

Additive Model ◽

Count Time Series ◽

The Impact

Download Full-text

Importance of applying Mixed Generalized Additive Model (MGAM) as a method for assessing the environmental health impacts: Ambient temperature and Acute Myocardial Infarction (AMI), among elderly in Shanghai, China

PLoS ONE ◽

10.1371/journal.pone.0255767 ◽

2021 ◽

Vol 16 (8) ◽

pp. e0255767

Author(s):

Xiaoqian Huang ◽

Weiping Ma ◽

Chikin Law ◽

Jianfeng Luo ◽

Naiqing Zhao

Keyword(s):

Myocardial Infarction ◽

Acute Myocardial Infarction ◽

Time Series ◽

Ambient Temperature ◽

Autocorrelation Function ◽

Risk Ratio ◽

Generalized Additive Model ◽

Seasonal Pattern ◽

Additive Model ◽

Degree Of Freedom

Association between acute myocardial infarction (AMI) morbidity and ambient temperature has been examined with generalized linear model (GLM) or generalized additive model (GAM). However, the effect size by these two methods might be biased due to the autocorrelation of time series data and arbitrary selection of degree of freedom of natural cubic splines. The present study analyzed how the climatic factors affected AMI morbidity for older adults in Shanghai with Mixed generalized additive model (MGAM) that addressed these shortcomings mentioned. Autoregressive random effect was used to model the relationship between AMI and temperature, PM10, week days and time. The degree of freedom of time was chosen based on the seasonal pattern of temperature. The performance of MGAM was compared with GAM on autocorrelation function (ACF), partial autocorrelation function (PACF) and goodness of fit. One-year predictions of AMI counts in 2011 were conducted using MGAM with the moving average. Between 2007 and 2011, MGAM adjusted the autocorrelation of AMI time series and captured the seasonal pattern after choosing the degree of freedom of time at 5. Using MGAM, results were well fitted with data in terms of both internal (R2 = 0.86) and external validity (correlation coefficient = 0.85). The risk of AMI was relatively high in low temperature (Risk ratio = 0.988 (95% CI 0.984, 0.993) for under 12°C) and decreased as temperature increased and speeded up within the temperature zone from 12°C to 26°C (Risk ratio = 0.975 (95% CI 0.971, 0.979), but it become increasing again when it is 26°C although not significantly (Risk ratio = 0.999 (95% CI 0.986, 1.012). MGAM is more appropriate than GAM in the scenario of response variable with autocorrelation and predictors with seasonal variation. The risk of AMI was comparatively higher when temperature was lower than 12°C in Shanghai as a typical representative location of subtropical climate.

Download Full-text

Abstract 345: Forecasting the Number of Out-Of-Hospital Cardiac Arrest in the Young Using Generalized Additive Model

Circulation ◽

10.1161/circ.140.suppl_2.345 ◽

2019 ◽

Vol 140 (Suppl_2) ◽

Author(s):

Alyssa Vermeulen ◽

Marina Del Rios ◽

Teri L Campbell ◽

Hai Nguyen ◽

Hoang H Nguyen

Keyword(s):

Time Series ◽

Cardiac Arrest ◽

Generalized Additive Model ◽

Moving Average ◽

Additive Model ◽

Resource Planning ◽

Model Framework ◽

Auto Regressive ◽

Hospital Cardiac Arrest ◽

Causality Hypothesis

Introduction: Accurate forecasting could help in resource planning and evaluation of intervention efforts to reduce out-of-hospital cardiac arrest. Hypothesis: Generalized additive model can rapidly and accurately forecast the number of OHCA in the young (1-35 years old) when compared to other auto regressive moving average (ARIMA) based models. Methods: Data were obtained from CARES in Chicago from May 2013 to December 2017. Monthly forecasts of the number of OHCA were performed using a generative additive model framework using the open source software Prophet and R. The first 50 months served as training and the last 6 months served as testing. Results: Figure 1 shows the distribution of the number of cases over the 3-year study period showing yearly seasonality and upward trend. Figure 2 shows the forecast of the number of arrest (middle line) along with the actual number of arrest (dots). The shaded band represents the 95% confidence interval of the prediction. Figure 3 show the comparison of the 6-month forecast and actual data. This model has low root mean square error score when compared with other ARIMA based models (2 vs. 4). Conclusion: Accurate time series forecasting of the number of OHCA arrests could be achieved. Time series analysis and forecasting are essential tools in evaluating the effectiveness of intervention efforts to reduce OHCA as they allow for causality hypothesis testing.

Download Full-text

Applying generalized additive model to integrate digital elevation model and remotely sensed data to predict the vegetation distribution

Chinese Journal of Plant Ecology ◽

10.3724/sp.j.1258.2012.01106 ◽

2013 ◽

Vol 36 (10) ◽

pp. 1106-1119

Author(s):

Chuang-Ye SONG ◽

Hui-Ming LIU ◽

Gao-Huan LIU ◽

Chong HUANG

Keyword(s):

Digital Elevation Model ◽

Generalized Additive Model ◽

Additive Model ◽

Remotely Sensed ◽

Vegetation Distribution ◽

Remotely Sensed Data ◽

Digital Elevation ◽

Elevation Model

Download Full-text

Wages, BMI, and Age: A Generalized Additive Model Using the Oracle Estimator

SSRN Electronic Journal ◽

10.2139/ssrn.1975044 ◽

2011 ◽

Author(s):

Christian Gregory

Keyword(s):

Generalized Additive Model ◽

Additive Model

Download Full-text

Assessing the Performance of Random Forests for Modeling Claim Severity in Collision Car Insurance

Risks ◽

10.3390/risks9030053 ◽

2021 ◽

Vol 9 (3) ◽

pp. 53

Author(s):

Yves Staudt ◽

Joël Wagner

Keyword(s):

Random Forests ◽

Goodness Of Fit ◽

Generalized Additive Model ◽

Mean Squared Error ◽

Additive Model ◽

Continuous Variables ◽

Insurance Portfolio ◽

Normal Transformation ◽

Car Insurance ◽

Log Normal

For calculating non-life insurance premiums, actuaries traditionally rely on separate severity and frequency models using covariates to explain the claims loss exposure. In this paper, we focus on the claim severity. First, we build two reference models, a generalized linear model and a generalized additive model, relying on a log-normal distribution of the severity and including the most significant factors. Thereby, we relate the continuous variables to the response in a nonlinear way. In the second step, we tune two random forest models, one for the claim severity and one for the log-transformed claim severity, where the latter requires a transformation of the predicted results. We compare the prediction performance of the different models using the relative error, the root mean squared error and the goodness-of-lift statistics in combination with goodness-of-fit statistics. In our application, we rely on a dataset of a Swiss collision insurance portfolio covering the loss exposure of the period from 2011 to 2015, and including observations from 81 309 settled claims with a total amount of CHF 184 mio. In the analysis, we use the data from 2011 to 2014 for training and from 2015 for testing. Our results indicate that the use of a log-normal transformation of the severity is not leading to performance gains with random forests. However, random forests with a log-normal transformation are the favorite choice for explaining right-skewed claims. Finally, when considering all indicators, we conclude that the generalized additive model has the best overall performance.

Download Full-text