scholarly journals Geographically Weighted Multivariate Logistic Regression Model and Its Application

2020 ◽  
Vol 2020 ◽  
pp. 1-10
Author(s):  
M. Fathurahman ◽  
Purhadi ◽  
Sutikno ◽  
Vita Ratnasari
Keyword(s):  
Logistic Regression ◽  
Parameter Estimation ◽  
Maximum Likelihood ◽  
Hypothesis Testing ◽  
Multinomial Distribution ◽  
Multivariate Logistic Regression Model ◽  
Ratio Test ◽  
Multivariate Logistic Regression ◽  
Testing Procedures ◽  
Mlr Model

This study investigates the geographically weighted multivariate logistic regression (GWMLR) model, parameter estimation, and hypothesis testing procedures. The GWMLR model is an extension to the multivariate logistic regression (MLR) model, which has dependent variables that follow a multinomial distribution along with parameters associated with the spatial weighting at each location in the study area. The parameter estimation was done using the maximum likelihood estimation and Newton-Raphson methods, and the maximum likelihood ratio test was used for hypothesis testing of the parameters. The performance of the GWMLR model was evaluated using a real dataset and it was found to perform better than the MLR model.

scholarly journals Multivariate Gamma Regression: Parameter Estimation, Hypothesis Testing, and Its Application

Symmetry ◽  
2020 ◽  
Vol 12 (5) ◽  
pp. 813
Author(s):  
Anita Rahayu ◽  
Purhadi ◽  
Sutikno ◽  
Dedy Dwi Prastyo
Keyword(s):  
Parameter Estimation ◽  
Maximum Likelihood ◽  
Hypothesis Testing ◽  
Likelihood Estimation ◽  
Estimation Procedure ◽  
Wald Test ◽  
Positive Outcome ◽  
Three Dimensions ◽  
Predictor Variables ◽  
Ratio Test

Gamma distribution is a general type of statistical distribution that can be applied in various fields, mainly when the distribution of data is not symmetrical. When predictor variables also affect positive outcome, then gamma regression plays a role. In many cases, the predictor variables give effect to several responses simultaneously. In this article, we develop a multivariate gamma regression (MGR), which is one type of non-linear regression with response variables that follow a multivariate gamma (MG) distribution. This work also provides the parameter estimation procedure, test statistics, and hypothesis testing for the significance of the parameter, partially and simultaneously. The parameter estimators are obtained using the maximum likelihood estimation (MLE) that is optimized by numerical iteration using the Berndt–Hall–Hall–Hausman (BHHH) algorithm. The simultaneous test for the model’s significance is derived using the maximum likelihood ratio test (MLRT), whereas the partial test uses the Wald test. The proposed MGR model is applied to model the three dimensions of the human development index (HDI) with five predictor variables. The unit of observation is regency/municipality in Java, Indonesia, in 2018. The empirical results show that modeling using multiple predictors makes more sense compared to the model when it only employs a single predictor.

scholarly journals Parameter Estimation and Hypothesis Testing of Multivariate Poisson Inverse Gaussian Regression

Symmetry ◽  
2020 ◽  
Vol 12 (10) ◽  
pp. 1738
Author(s):  
Selvi Mardalena ◽  
Purhadi Purhadi ◽  
Jerry Dwi Trijoyo Purnomo ◽  
Dedy Dwi Prastyo
Keyword(s):  
Parameter Estimation ◽  
Maximum Likelihood ◽  
Hypothesis Testing ◽  
Poisson Regression ◽  
Ratio Test ◽  
Inverse Gaussian ◽  
Count Response ◽  
The Mean ◽  
Gaussian Regression ◽  
Response Variables

Multivariate Poisson regression is used in order to model two or more count response variables. The Poisson regression has a strict assumption, that is the mean and the variance of response variables are equal (equidispersion). Practically, the variance can be larger than the mean (overdispersion). Thus, a suitable method for modelling these kind of data needs to be developed. One alternative model to overcome the overdispersion issue in the multi-count response variables is the Multivariate Poisson Inverse Gaussian Regression (MPIGR) model, which is extended with an exposure variable. Additionally, a modification of Bessel function that contain factorial functions is proposed in this work to make it computable. The objective of this study is to develop the parameter estimation and hypothesis testing of the MPIGR model. The parameter estimation uses the Maximum Likelihood Estimation (MLE) method, followed by the Newton–Raphson iteration. The hypothesis testing is constructed using the Maximum Likelihood Ratio Test (MLRT) method. The MPIGR model that has been developed is then applied to regress three response variables, i.e., the number of infant mortality, the number of under-five children mortality, and the number of maternal mortality on eight predictors. The unit observation is the cities and municipalities in Java Island, Indonesia. The empirical results show that three response variables that are previously mentioned are significantly affected by all predictors.

scholarly journals Parameter estimation and hypothesis testing of geographically and temporally weighted bivariate Poisson inverse Gaussian regression model

2021 ◽  
Vol 880 (1) ◽  
pp. 012045
Author(s):  
Meylita Sari ◽  
Sutikno ◽  
Purhadi
Keyword(s):  
Parameter Estimation ◽  
Maximum Likelihood ◽  
Hypothesis Testing ◽  
Count Data ◽  
Poisson Regression ◽  
Statistical Test ◽  
Ratio Test ◽  
Inverse Gaussian ◽  
Test Statistics ◽  
Gaussian Regression

Abstract One of the appropriate methods used to model count data response and its corresponding predictors is Poisson regression. Poisson regression strictly assumes that the mean and variance of response variables should be equal (equidispersion). Nonetheless, some cases of the count data unsatisfied this assumption because variance can be larger than mean (over-dispersion). If overdispersion is violated, causing the underestimate standard error. Furthermore, this will lead to incorrect conclusions in the statistical test. Thus, a suitable method for modelling this kind of data needs to develop. One alternative model to outcome the overdispersion issue in bivariate response variable is the Bivariate Poisson Inverse Gaussian Regression (BPIGR) model. The BPIGR model can produce a global model for all locations. On the other hand, each location and time have different geographic conditions, social, cultural, and economical so that Geographically and Temporally Bivariate Poisson Inverse Gaussian Regression (GTWBPIGR)) is needed. The weighting function spatial-temporal in GTWBPIGR generates a different local model for each period. GTWBPIGR model solves the overdispersion case and generates global models for each period and location. The parameter estimation of the GTWBPIGR model uses the Maximum Likelihood Estimation (MLE) method, followed by Newton Raphson iteration. Meanwhile, the test statistics on the hypothesis testing is simultaneously testing of the GTWBPIGR model is obtained with the Maximum Likelihood Ratio Test (MLRT) approach, using n large samples of the statistical test is chi-square distribution. Moreover, the test statistics for partially testing used the Z-test statistic.

Preoperative Circulating MiR-210, a Risk Factor for Postoperative Delirium Among Elderly Patients with Gastric Cancer Undergoing Curative Resection

2020 ◽  
Vol 26 (40) ◽  
pp. 5213-5219
Author(s):  
Yun Chen ◽  
Jinwei Zheng ◽  
Junping Chen
Keyword(s):  
Gastric Cancer ◽  
Logistic Regression ◽  
Risk Factor ◽  
Elderly Patients ◽  
Curative Resection ◽  
Postoperative Delirium ◽  
Area Under The Curve ◽  
Multivariate Logistic Regression Model ◽  
Multivariate Logistic Regression ◽  
Clinicopathologic Characteristics

Background: Postoperative delirium (POD) is a very common complication in elderly patients with gastric cancer (GC) and associated with poor prognosis. MicroRNAs (miRNAs) serve as key post-transcriptional regulators of gene expression via targeting mRNAs and play important roles in the nervous system. This study aimed to investigate the potential predictive role of miRNAs for POD. Methods: Elderly GC patients who were scheduled to undergo elective curative resection were consequently enrolled in this study. POD was assessed at 1 day before surgery and 1-7 days after surgery following the guidance of the 5th edition of Diagnostic and Statistical Manual of Mental Disorders (DSM V, 2013). The demographics, clinicopathologic characteristics and preoperative circulating miRNAs by quantitative reverse transcription-polymerase chain reaction (qRT-PCR) were compared between patients with or without POD. Risk factors for POD were assessed via univariate and multivariate logistic regression analyses. Results: A total of 370 participants were enrolled, of which 63 had suffered from POD within postoperative 7 days with an incidence of 17.0%. Preoperative miR-210 was a predictor for POD with an area under the curve (AUC) of 0.921, a cut-off value of 1.67, a sensitivity of 95.11%, and a specificity of 92.06%, (P<0.001). In the multivariate logistic regression model, the relative expression of serum miR-210 was an independent risk factor for POD (OR: 3.37, 95%CI: 1.98–5.87, P=0.003). Conclusions: In conclusion, the present study highlighted that preoperative miR-210 could serve as a potential predictor for POD in elderly GC patients undergoing curative resection.

scholarly journals The second pregnancy has no effect in the incidence of macrosomia: a cross-sectional survey in two western Chinese regions

2021 ◽  
Vol 40 (1) ◽  
Author(s):  
Li Luo ◽  
Huan Zeng ◽  
Mao Zeng ◽  
Xueqing Liu ◽  
Xianglong Xu ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Western China ◽  
Fetal Macrosomia ◽  
Multivariate Logistic Regression Model ◽  
Multivariate Logistic Regression ◽  
Cross Sectional ◽  
Second Child ◽  
Child Policy ◽  
Significant Difference ◽  
The Impact

Abstract Background After the implementation of the universal two-child policy in China, the increase in parity has led to an increase in adverse pregnancy outcomes. The impact of one and two fetuses on the incidence of fetal macrosomia has not been fully confirmed in China. This study aimed to explore the differences in the incidence of fetal macrosomia in first and second pregnancies in Western China after the implementation of the universal two-child policy. Methods A total of 1598 pregnant women from three hospitals were investigated by means of a cross-sectional study from August 2017 to January 2018. Participants were recruited by convenience and divided into first and second pregnancy groups. These groups included 1094 primiparas and 504 women giving birth to their second child. Univariate and multivariate logistic regression analyses were performed to discuss the differences in the incidence of fetal macrosomia in first and second pregnancies. Results No significant difference was found in the incidence of macrosomia in the first pregnancy group (7.2%) and the second pregnancy group (7.1%). In the second-time pregnant mothers, no significant association was found between the macrosomia of the second child (5.5%) and that of the first child (4.7%). The multivariate logistic regression model showed that mothers older than 30 years are not likely to give birth to children with macrosomia (odds ratio (OR) 0.6, 95% confidence interval (CI) 0.4,0.9). Conclusions The incidence of macrosomia in Western China is might not be affected by the birth of the second child and is not increased by low parity.

scholarly journals ESTIMATION OF THE BINARY LOGISTIC REGRESSION MODEL PARAMETER USING BOOTSTRAP RE-SAMPLING

2018 ◽  
Vol 48 (3) ◽  
pp. 199-204 ◽  
Author(s):  
R. LI ◽  
J. ZHOU ◽  
L. WANG
Keyword(s):  
Logistic Regression ◽  
Parameter Estimation ◽  
Maximum Likelihood ◽  
Regression Model ◽  
Logistic Regression Model ◽  
Binary Logistic Regression ◽  
Parametric Bootstrap ◽  
Binary Logistic Regression Model ◽  
Bayesian Bootstrap ◽  
Non Parametric

In this paper, the non-parametric bootstrap and non-parametric Bayesian bootstrap methods are applied for parameter estimation in the binary logistic regression model. A real data study and a simulation study are conducted to compare the Nonparametric bootstrap, Non-parametric Bayesian bootstrap and the maximum likelihood methods. Study results shows that three methods are all effective ways for parameter estimation in the binary logistic regression model. In small sample case, the non-parametric Bayesian bootstrap method performs relatively better than the non-parametric bootstrap and the maximum likelihood method for parameter estimation in the binary logistic regression model.

Sex-based differences in outcomes for adult patients presenting to the emergency department with a concussion

2021 ◽  
pp. 1-10
Author(s):  
Brian H. Rowe ◽  
Esther H. Yang ◽  
Lindsay A. Gaudet ◽  
Leeor Eliyahu ◽  
Daniela R. Junqueira ◽  
...  
Keyword(s):  
Emergency Department ◽  
Logistic Regression ◽  
Motor Vehicle ◽  
Adult Population ◽  
Treatment Options ◽  
Multivariate Logistic Regression Model ◽  
Adult Women ◽  
Multivariate Logistic Regression ◽  
Persistent Symptoms ◽  
Postconcussion Symptoms

OBJECTIVE Patients with concussion frequently present to the emergency department (ED). Studies of athletes and children indicate that concussion symptoms are often more severe and prolonged in females compared with males. Given infrequent study of concussion symptoms in the general adult population, the authors conducted a sex-based comparison of patients with concussion. METHODS Adults (≥ 17 years of age) presenting with concussion to one of three urban Canadian EDs were recruited. Discharged patients were contacted by telephone 30 and 90 days later to capture the extent of persistent postconcussion symptoms using the Rivermead Post Concussion Symptoms Questionnaire (RPQ). A multivariate logistic regression model for persistent symptoms that included biological sex was developed. RESULTS Overall, 250 patients were included; 131 (52%) were women, and the median age of women was significantly higher than that of men (40 vs 32 years). Women had higher RPQ scores at baseline (p < 0.001) and the 30-day follow-up (p = 0.001); this difference resolved by 90 days. The multivariate logistic regression identified that women, patients having a history of sleep disorder, and those presenting to the ED with concussions after a motor vehicle collision were more likely to experience persistent symptoms. CONCLUSIONS In a community concussion sample, inconsequential demographic differences existed between adult women and men on ED presentation. Based on self-reported and objective outcomes, work and daily activities may be more affected by concussion and persistent postconcussion symptoms for women than men. Further analysis of these differences is required to identify different treatment options and ensure adequate care and management of injury.

Identification of characteristics associated with long-term survival in patients with metastatic urothelial carcinoma (mUC) who received durvalumab (D) with or without tremelimumab (T) in clinical studies.

2021 ◽  
Vol 39 (6_suppl) ◽  
pp. 441-441
Author(s):  
Marie Alt ◽  
Carlos Stecca ◽  
Shaum Kabadi ◽  
Benga Kazeem ◽  
Srikala S. Sridhar
Keyword(s):  
Logistic Regression ◽  
Phase I ◽  
Regression Model ◽  
Clinical Studies ◽  
Logistic Regression Model ◽  
Multivariate Logistic Regression Model ◽  
Multivariate Logistic Regression ◽  
Term Survival ◽  
Long Term Survival

441 Background: Immune checkpoint inhibitors (ICI) have changed the landscape of mUC, yet outcomes are variable as some patients (pts) do not respond to treatment while others have a durable response. To optimally select pts who may derive benefit from ICIs, predictive factors are required. This retrospective, post-hoc analysis evaluated pt characteristics to determine differences between short and long-term survivors among pts with mUC who received D (anti–PD-L1) with or without T (anti–CTLA-4) in two clinical studies. Methods: Pts with platinum-refractory mUC who received D monotherapy in the phase I/II study 1108 (10 mg/kg Q2W, up to 12 mo) or D+T in the phase I study 10 (D at 20 mg/kg + T at 1 mg/kg Q4W for 4 mo, then D at 10 mg/kg Q2W for 12 mo) were included. Pt characteristics, tumor characteristics, radiological assessments, and biological assessments were collected. The primary outcome measure was long-term overall survival (OS). Pts were categorized as OS ≥2 yrs (from 1st dose of study drug) or OS <2 yrs. A univariate analysis was conducted on each baseline characteristic to assess independent associations with long-term OS; a multivariate logistic regression model was employed including each variable with a p-value ≤0.1 as factors or covariates. Results: A total of 367 pts with mUC were included in the analysis: 88 (24.0%) had OS ≥2 yrs (range: 2.09–4.99) and 279 (76.0%) had OS <2 yrs (range: 0.03–1.98). Pts with OS ≥2 yrs had a significantly higher objective response rates than those with OS <2 yrs (71.6% vs 5.7%; p<0.0001) and a significantly longer duration of response (median 2.3 yrs vs 0.39 yrs; p<0.0001). The characteristics included in the multivariate logistic regression model are listed in the Table. Long-term OS was significantly associated with ECOG PS, PD-L1 status, baseline hemoglobin level, and baseline absolute neutrophils count. Conclusions: Our analyses show that several characteristics, including tumor response to treatment, are associated with long-term OS for pts with mUC treated with D or D+T. Further investigation into these and other characteristics may provide additional insights into long-term survival outcomes with ICIs. [Table: see text]

Sign in / Sign up

Export Citation Format

Share Document