Diagnostics

Author(s):  
Youseop Shin

Chapter Three talks about diagnostics. As in regression analysis, several properties of residuals should be satisfied, if the fitted model is appropriate. Residuals should be a realization of a white or IID sequence. They should have zero as a mean. Their variance, σ‎2, should be constant for all values of the Time variable. They should be normally distributed as well. This chapter explains how to test these points.

2019 ◽  
Vol 35 (2) ◽  
pp. 977-1001 ◽  
Author(s):  
Shu-Hsien Chao ◽  
Yi-Hau Chen

Regression analysis is a basic and essential tool for developing the ground motion prediction equation (GMPE). Generally, the probability of intensity measurement for a given ground motion scenario described by several predictors is assumed to be normally distributed. However, because of the triggering threshold of the strong-motion station, ground motion records below the triggering threshold are truncated (i.e., not recorded), and the truncated intensity levels of spectral accelerations at different periods are random variables. Consequently, the sampling of the ground motion data used in GMPE development is biased, and the observed probability of the intensity measurement is no longer normally distributed. Therefore, a novel two-step maximum-likelihood method is proposed in this paper as a regression tool to overcome this problem in GMPE development. The advantage of the proposed method is that the correlation between records from the same events and those from the same sites as well as the biased sampling problem can be considered simultaneously, and more ground motion data can be considered to derive more reliable analysis results.


Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 3381-3381
Author(s):  
Wei Xiao ◽  
Linlu Ma ◽  
Yufeng Shang ◽  
Yuxin Tan ◽  
Fuling Zhou

Abstract Background: More and more cases of leukemia with heart injury have been reported, mostly due to tumor cell infiltration, endothelial injury and high viscosity of leukemia cells in circulating blood. To evaluate whether the newly diagnosed acute leukemia patients have suffered heart-related injuries, and determine the possible related factors, we carried out this study. Methods: From January 2015 to August 2019, we retrospectively selected 408 patients from the Department of Hematology and Cardiology, Zhongnan Hospital of Wuhan University. There were 200 newly diagnosed acute leukemia patients (including 15 acute hyperleukocytic leukemia patients), 105 patients without cancer and no known heart disease, and 103 cases of confirmed coronary heart disease. We collected the basic information of patients, echocardiographic data and myocardial enzyme results. We also followed up all newly diagnosed acute leukemia patients for prognostic analysis. All data were logarithmically transformed to make the data normally distributed. Normally distributed samples were compared using independent sample t-tests, Mann-Whitney U rank sum test was used for non-normally distributed samples and qualitative data was using chi-square test. The log-rank test was used to perform univariate analysis through Kaplan-Meier curves, and meaningful values were screened for multivariate COX regression analysis. The statistical analysis was performed with the SPSS software. Results: Compared with the control group, patients with newly diagnosed acute leukemia had already experienced some heart-related injuries. The average values of myocardial enzyme indexes such as CKMB,LDH,hs-cTnI,BNP and echocardiography indexes such as LV,EDV,ESV,SV,EF were all significantly different. But the degree of heart damage was lower than that of the coronary heart disease group. Compared with hyperleukocytic leukemia, non-hyperleukocytic leukemia had suffered more serious heart damage, especially myocardial enzyme indexes such as HBDH and LDH were significantly increased. Analyze the relationship between age, gender, cardiovascular risk factors (hypertension, diabetes and smoking history), myocardial enzyme indexes, echocardiography indexes and survival, to screen for risk factors related to the prognosis of the disease. Variables with statistically significant effect measurement values include age (p<0.001), LDH (p=0.018), and EF (p=0.053). In addition, the meaningful variables in the previous single-factor analysis were subjected to multi-factor COX regression analysis, and age (HR = 1.515, 95 %CI, 0.994-2.311, P = 0.001), EF (HR = 0.526, 95% CI, 0.361-0.766, P=0.05) were independent factors related to the prognosis of patients. Classification based on the clinical diagnosis of echocardiography in newly diagnosed acute leukemia patients, significant differences in myocardial enzymes and related indicators of echocardiography were included for ROC plots. The 3 indicators with the highest results for the area under the curve (AUC) from ROC plots were LV, EDV and HBDH, AUC were 0.721,0.619 and 0.615, respectively. This result revealed that the best predictor of leukemia myocardial damage was LV. It showed the predict ability of LV (AUC = 0.721, 95%CI = 0.643-0.799, P <0.001; sensitivity = 42.4%; specificity = 91.8%). Conclusions: According to our research, we have confirmed that the cardiac function of newly diagnosed acute leukemia patients is significantly different from that of patients without leukemia, and these differences will have an impact on the prognosis of patients. Among them, the best predictor of leukemia heart damage is LV, and EF is an independent risk factor for the prognosis of leukemia patients. In addition, newly diagnosed acute leukemia patients are more likely to have an increase in left ventricular diameter and a significant decrease in ejection fraction. Further research is needed to explore the specific mechanism in the future. Disclosures No relevant conflicts of interest to declare.


2012 ◽  
Vol 3 (2) ◽  
pp. 934
Author(s):  
Kasiyat Kasiyat ◽  
Pariang Siagian

This article aims to analyze the contribute of banking as intermediate institutions by studying if there is the significance relationship between the amount of funds raised and amount of outstanding loans or given credits as the two main functions of banks. The benefits are useful for banking authorities, management of banks, costumers, and scientific people according to their interests. Population as object in this research were all of commercial banks listed in Bank Indonesia website. The representative samples were 10 banks which have been go public. The data were got from the statements of financial position, 2007 to 2011 periods by using technical sampling. The research methods, first were used classical asumption, then regression analysis to test the hypothesis t-test. From the research, it showed that amount of funds raised and the amount of outstanding loans were normally distributed with a significance value 2,000. Autocorrelation test results were obtained that the data were free of the problem of autocorrelation, where the DW-value is 2,946. Correlation analyze was performed on the two variables have the positive outcome or significance to the value of r = 0,947. And the results from the analyzed of test significance of correlation coefficient, showed that the relation between the two variables can be generalized to all commercial banks because the value of t was 23,274. 


2012 ◽  
Vol 02 (04) ◽  
pp. 389-400 ◽  
Author(s):  
Sara M. Gustavsson ◽  
Sandra Johannesson ◽  
Gerd Sallsten ◽  
Eva M. Andersson

2019 ◽  
Vol 11 (1) ◽  
pp. 1
Author(s):  
Sugito Sugito ◽  
Y Suyitno ◽  
Kuntoro Kuntoro

This research aims to identify: 1) the influence of working time on teacherperformance, 2) the influence of motivation on teacher performance, and 3) theinfluence of working time and motivation on performance of elementary school teachersin the villages of Samudra and Samudra Kulon. This research is ex-post facto lookingfor causal relationships, to reveal facts based on measurements of symptoms that haveoccurred in each respondent. Determination of the sample in this study is a saturatedsample, where all members of the population become the object of research. Totalpopulation is 50 people. The analysis carried out was a simple and multiple correlationanalysis and regression analysis. Respondents in this research have a working time of 2to 32 years. The results of this research are (1) there is a negative influence on theworking time variable on teacher performance; (2) there is a positive influence on themotivation variable on teacher performance; and (3) there is a positive influence onworking time and motivation on performance. The effective contribution of working timeand motivation to performance are 8.9% and 18%, respectively. The relativecontribution of working time and motivation to performance are 33% and 67%,respectively.


Rheumatology ◽  
2021 ◽  
Author(s):  
Ciara M O’Brien ◽  
Nikos Ntoumanis ◽  
Joan L Duda ◽  
George D Kitas ◽  
Jet J C S Veldhuijzen van Zanten ◽  
...  

Abstract Objectives The aims of this study were to examine the longitudinal and bi-directional associations of pain and fatigue with sedentary, standing and stepping time in RA. Methods People living with RA undertook identical assessments at baseline (T1, n = 104) and 6-month follow-up (T2, n = 54). Participants completed physical measures (e.g. height, weight, BMI) and routine clinical assessments to characterize RA disease activity (DAS-28). Participants also completed questionnaires to assess physical function (HAQ), pain (McGill Pain Questionnaire) and fatigue (Multidimensional Assessment of Fatigue Scale). Participants’ free-living sedentary, standing and stepping time (min/day) were assessed over 7 days using the activPAL3µ™. For the statistical analysis, hierarchical regression analysis was employed to inform the construction of path models, which were subsequently used to examine bi-directional associations of pain and fatigue with sedentary, standing and stepping time. Specifically, where significant associations were observed in longitudinal regression analysis, the bi-directionality of these associations was further investigated via path analysis. For regression analysis, bootstrapping was applied to regression models to account for non-normally distributed data, with significance confirmed using 95% CIs. Where variables were normally distributed, parametric, non-bootstrapped statistics were also examined (significance confirmed via β coefficients, with P < 0.05) to ensure all plausible bi-directional associations were examined in path analysis. Results Longitudinal bootstrapped regression analysis indicated that from T1 to T2, change in pain, but not fatigue, was positively associated with change in sedentary time. In addition, change in pain and fatigue were negatively related to change in standing time. Longitudinal non-bootstrapped regression analysis demonstrated a significant positive association between change in fatigue with change in sedentary time. Path analysis supported the hypothesized bi-directionality of associations between change in pain and fatigue with change in sedentary time (pain, β = 0.38; fatigue, β = 0.44) and standing time (pain, β = –0.39; fatigue, β = –0.50). Conclusion Findings suggest pain and fatigue are longitudinally and bi-directionally associated with sedentary and standing time in RA.


1984 ◽  
Vol 75 ◽  
pp. 597
Author(s):  
E. Grün ◽  
G.E. Morfill ◽  
T.V. Johnson ◽  
G.H. Schwehm

ABSTRACTSaturn's broad E ring, the narrow G ring and the structured and apparently time variable F ring(s), contain many micron and sub-micron sized particles, which make up the “visible” component. These rings (or ring systems) are in direct contact with magnetospheric plasma. Fluctuations in the plasma density and/or mean energy, due to magnetospheric and solar wind processes, may induce stochastic charge variations on the dust particles, which in turn lead to an orbit perturbation and spatial diffusion. It is suggested that the extent of the E ring and the braided, kinky structure of certain portions of the F rings as well as possible time variations are a result of plasma induced electromagnetic perturbations and drag forces. The G ring, in this scenario, requires some form of shepherding and should be akin to the F ring in structure. Sputtering of micron-sized dust particles in the E ring by magnetospheric ions yields lifetimes of 102to 104years. This effect as well as the plasma induced transport processes require an active source for the E ring, probably Enceladus.


Author(s):  
A. Colin Cameron ◽  
Pravin K. Trivedi

Sign in / Sign up

Export Citation Format

Share Document