No abstract available.
Regression Analysis*

No abstract available.
Humans ; Regression Analysis*

Multicollinearity is an important issue affecting the results of regression analysis. LASSO developed in recent years has great advantages in selecting explanatory variables, processing high-dimensional data, and solving multicollinearity problems. This method adds a penalty term to the model estimation, which can compress the regression coefficients of some unnecessary variables to zero and then remove them from the model to achieve the purpose of variable screening. This paper focuses on the LASSO method and compares it with optimal subsets, ridge regression, adaptive LASSO, and elastic net results. It is found that both LASSO and adaptive LASSO have good performance in solving independent variable multicollinearity problems and enhancing model interpretation and prediction accuracy.
Humans ; Regression Analysis

Humans ; Influenza, Human ; mortality ; Regression Analysis

Heart rate variability (HRV) is an important point to judge a person's state in modern medicine. This paper is aimed to research a person's fatigue level connected with vagal nerve based on the HRV using the improved Welch method. The process of this method is that it firstly uses a time window function on the signal to be processed, then sets the length of time according to the requirement, and finally makes frequency domain analysis. Compared with classical periodogram method, the variance and consistency of the present method have been improved. We can set time span freely using this method (at present, the time of international standard to measure HRV is 5 minutes). This paper analyses the HRV's characteristics of fatigue crowd based on the database provided by Physio-Net. We therefore draw the conclusion that the accuracy of Welch analyzing HRV combining with appropriate window function has been improved enormously, and when the person changes to fatigue, the vagal activity is diminished and sympathetic activity is raised.
Fatigue ; diagnosis ; Heart Rate ; Humans ; Regression Analysis

Prevalence ; Proportional Hazards Models ; Regression Analysis

Background: The aims of this study were to determine the average length of stay in the Post-anesthesia Care Unit (PACU LOS) in the Philippine General Hospital (PGH) and to create a model that will predict the PACU LOS based on the factors that significantly affect the LOS. Determination and prediction of PACU LOS is essential in resource utilization, and in cost containment and reduction. Addressing the modifiable variables that affect the PACU LOS may lead to an improvement in the LOS of patients in the PACU and, consequently, to better recovery room staffing and a reduced cost for the patients and the hospital. Methods: A prospective chart review of 400 postoperative patients admitted in the PGH PACU was done. Summary statistics were presented. Using the set of variables found to be significant, a regression model was formulated to estimate the PACU LOS. Results: The mean PACU LOS was 4.59 hours. There were significant differences in the mean PACU LOS based on the occurrence of complications. There were also significant differences in the median PACU LOS based on the type and duration of surgery, anesthetic technique, and duration of anesthesia. The multiple linear regression model that best predicted PACU LOS included ASA-PS classification, type of surgery, duration of surgery, anesthetic technique, and occurrence of intraoperative or postoperative complications. Conclusions The mean PACU LOS of the Philippine Genera Hospital is higher than that of published data. The factors included in the model that best predicts PACU LOS may be studied to improve the PACU LOS.
Length of Stay ; Anesthesia ; Anesthesiology ; Regression Analysis

Objective: The study aims to determine the factors associated with unsuccessful treatment outcome among patients who have undergone tuberculosis treatment. Methods: An analytic cross-sectional study was employed through secondary data analysis of administrative data collected by the National Tuberculosis Control Program from October 2015 to September 2016. Using multiple logistic regression analysis, factors associated with unsuccessful treatment outcome were determined. Results: Multiple logistic regression analysis revealed that belonging to age groups 25 to 34 (aOR=0.73; 95%CI 0.54-0.99) or 35 to 44 (aOR=0.75; 95%CI 0.56-0.99), being male (aOR=1.30; 95%CI 1.03-1.64), doing crafts and related trades work (aOR=0.66; 95%CI 0.46-0.94), living in either a 4th class city (aOR=0.46; 95%CI 0.26-0.82), 1st class municipality (aOR=0.75; 95%CI 0.57-0.98), 4th class municipality (aOR=0.59; 95%CI 0.38-0.93), having a positive sputum smear result (aOR=1.60; 95%CI 1.29-2.00), having rifampicin-resistant/ multidrug-resistant tuberculosis (aOR=9.32; 95%CI 7.28-11.93), being a treatment after lost to follow-up case (aOR=1.84; 95%CI 1.37-2.47) or a case with previously unknown treatment outcome (aOR=1.42; 95%CI 1.00-2.01) were significant correlates of unsuccessful treatment outcome. Conclusion The study found that age, sex, occupation, residence, sputum smear results, drug resistance, and history of previous treatment were associated with unsuccessful treatment outcome.
Philippines ; Tuberculosis, Pulmonary ; Treatment Outcome ; Regression Analysis

OBJECTIVE: To realize batch processing of univariate Cox regression analysis for great database by SAS marco program. METHODS: We wrote a SAS macro program, which can filter, integrate, and export P values to Excel by SAS9.2. The program was used for screening survival correlated RNA molecules of ovarian cancer. RESULTS: A SAS marco program could finish the batch processing of univariate Cox regression analysis, the selection and export of the results. CONCLUSION The SAS macro program has potential applications in reducing the workload of statistical analysis and providing a basis for batch processing of univariate Cox regression analysis.
Proportional Hazards Models ; Regression Analysis ; Software

The purposes of this study were 1) to analyze the effect of the information quality of standard menu system of ERP on the user's satisfaction 2) to analyze the effect of the information quality of standard menu system of ERP on the user's performance. The questionnaires were distributed to 260 end users(dietitians, foodservice management managers) in charge of managing the institutional foodservice such as the office(government agencies, factories), schools(universities, middle and high school) and hospitals which were managed by contract foodservice management company "C" in Seoul and Kyunggi. The surveys were performed from July 26 to July 30, 2004. Two hundred and fifty questionnaires were responded(response rate : 96.2%). The statistical analysis was conducted by using SPSS Win(11.5). And the descriptive analysis, factor analysis, reliability test, pearson correlation and simple regression analysis were performed. The results showed that the information quality of standard menu system of ERP have a significant positive effect on the user's satisfaction(p<0.001) and user's performance(p<0.001). When constructing standard menu system on ERP in the contract foodservice management company, the information quality of standard menu system of ERP should be guaranteed in order to satisfy the end-user and to improve the user's performance.
Factor Analysis, Statistical ; Gyeonggi-do ; Surveys and Questionnaires ; Regression Analysis ; Seoul