Explorations in statistics: correlation

2010 ◽  
Vol 34 (4) ◽  
pp. 186-191 ◽  
Author(s):  
Douglas Curran-Everett

Learning about statistics is a lot like learning about science: the learning is more meaningful if you can actively explore. This sixth installment of Explorations in Statistics explores correlation, a familiar technique that estimates the magnitude of a straight-line relationship between two variables. Correlation is meaningful only when the two variables are true random variables: for example, if we restrict in some way the variability of one variable, then the magnitude of the correlation will decrease. Correlation cannot help us decide if changes in one variable result in changes in the second variable, if changes in the second variable result in changes in the first variable, or if changes in a third variable result in concurrent changes in the first two variables. Correlation can help provide us with evidence that study of the nature of the relationship between x and y may be warranted in an actual experiment in which one of them is controlled.

Author(s):  
Weimin Cui ◽  
Wei Guo ◽  
Zhongchao Sun ◽  
Tianxiang Yu

In order to analyze the reason of failure and improve the reliability of the idler shaft, this paper studies the reliability and sensitivity for the idler shaft based on Kriging model and Variance Methods respectively. The finite element analysis (FEA) of idler shaft is studied in ABAQUS firstly. Then, combining the performance function and various random variables, the Kriging model of idler shaft is established and verified. Based on Kriging model which has been established, the relationship between random variables and the response value is studied, and the function reliability is calculated which explains why the failure of the idler shaft occurred frequently in service. Finally, the variance-based sensitivity method is used for sensitivity analysis of influence factors, the result shows that the reliability of idler shaft is sensitive to the inner diameter of body A and inner diameter of body B, which could contribute for the analysis and further improvement of idler shaft.


1976 ◽  
Vol 46 (6) ◽  
pp. 463-465 ◽  
Author(s):  
F. William Reuter ◽  
Geraldine E. Secor ◽  
Mendel Friedman

Bromine is measured in flame-resistant wool fabric by x-ray fluorescence spectrometry with a relative precision of 3 to 8% and relative accuracy of better than 10%. The method computes the bromine concentration from fluorescence measurements of the sample, and a thin film standard, and two measurements of attenuation by the sample. Deviations of 10 to 20% from a straight-line relationship of x-ray counts to bromine concentration are accounted for. X-ray fluorescence is generally useful for routine analyses of bromine in textiles and has advantages over wet chemical analysis.


2020 ◽  
Vol 24 ◽  
pp. 39-55
Author(s):  
Julyan Arbel ◽  
Olivier Marchal ◽  
Hien D. Nguyen

We investigate the sub-Gaussian property for almost surely bounded random variables. If sub-Gaussianity per se is de facto ensured by the bounded support of said random variables, then exciting research avenues remain open. Among these questions is how to characterize the optimal sub-Gaussian proxy variance? Another question is how to characterize strict sub-Gaussianity, defined by a proxy variance equal to the (standard) variance? We address the questions in proposing conditions based on the study of functions variations. A particular focus is given to the relationship between strict sub-Gaussianity and symmetry of the distribution. In particular, we demonstrate that symmetry is neither sufficient nor necessary for strict sub-Gaussianity. In contrast, simple necessary conditions on the one hand, and simple sufficient conditions on the other hand, for strict sub-Gaussianity are provided. These results are illustrated via various applications to a number of bounded random variables, including Bernoulli, beta, binomial, Kumaraswamy, triangular, and uniform distributions.


1984 ◽  
Vol 56 (3) ◽  
pp. 765-771 ◽  
Author(s):  
M. R. Pinsky

The relationship between right atrial pressure (Pra) and venous return describes a venous return curve. Because changes in venous return and right ventricular stroke volume (SVRV) are similar during small tidal volume (VT) breathing, we compared the relationship of SVRV and Pra during positive-pressure ventilation (VT less than 10 ml/kg) in 17 pentobarbital-anesthetized, closed-chest, canine preparations. The SVRV-Pra relationship describes a straight line with a negative slope and a positive mean systemic pressure (Pms)-zero flow intercept (instantaneous Pms). Instantaneous Pms is similar to ventricular fibrillation-induced (stop-flow) Pms (8.1 +/- 0.8 vs. 8.4 +/- 0.7 mmHg, mean +/- SE). With volume infusion, both instantaneous and stop-flow Pms increase to a similar degree (R = 0.9014, P less than 0.001). The effect of opening an arteriovenous fistula is time dependent and variable between dogs, but by 30 s it is associated with an increase in instantaneous Pms (5.2 +/- 3.2 mmHg). Vascular compliance determined by adding and removing blood from the intravascular space displays a curvilinear hysteresis. The instantaneous venous return curve is affected by intravascular blood volume, vasomotor tone, and resistance to venous return. The relationship between SVRV and Pra during small VT breathing define instantaneous venous return curves similar to those described using right-heart bypass preparations.


1979 ◽  
Vol 4 (2) ◽  
pp. 125-133
Author(s):  
Ramesh Gupta

The sharp rise in prices in recent years has stirred up a brisk debate on the meaning and measurement of business income. In computing income, costs of labour and material are generally reflected in current prices, but the provision made for depreciation is generally not sufficient to replace fixed assets. This paper investigates various approaches to the oroblem of providing sufficient depreciation allowance in measuring income. The relationship between written down value and the straight line inflation adjusted method of providing depreciation has been examined. Simulation technique has been used to illustrate the problem and pertinent policy implications of the use of various depreciation methods have been highlighted.


1965 ◽  
Vol 97 (2) ◽  
pp. 141-147 ◽  
Author(s):  
A. L. Turnbull

AbstractAgelenopsis potteri (Blackwall) spiders that were reared from egg to adult on live prey (Aedes aegypti L.) supplied at different daily rates varied in the rate that they were able to capture prey, grow, and attain maturity. Mortality varied inversely with feeding rates, but some spiders matured at each feeding rate. All spiders matured in seven stages regardless of the rate of feeding. Both sexes were mature following the sixth moult. The rate of prey capture declined sharply in the adults. Males matured about four days sooner than females. A straight-line relationship exists between the rate at which prey were captured and the dry weights of the adult spiders. A straight-line relationship was also found between the numbers of prey captured per day and the daily development of the spider.


Measurements of the burning velocities of methane, ethane, propane, butane, ethylene, carbon monoxide and cyanogen mixtures with air, in the range about 4 to 8 cm, are made by the flat-flame burner method with an accuracy of 2 to 3%. The results can be represented by a straight-line relationship between composition and burning velocity except for carbon monoxide which is sensitive to the percentage of water vapour present. Extrapolated values agree well with recent measurements of faster flames. Measurements are also made on binary mixtures with air of the gases, including hydrogen. The mixture law holds except with mixtures containing carbon monoxide. Limits of inflammability are also determined and the burning velocities at the limits average 3⋅6 cm/s. The mixtures obey the Le Chatelier rule accurately, except for carbon monoxide mixtures. The burning velocities of the hydrocarbons can be represented approximately by a straight-line relationship with the heat generated and with the maximum flame temperature, but correlation is best when thermal conductivity is introduced. At a given velocity the excess energy maintained by the flame appears to be constant for all the hydrocarbons investigated, except methane, which behaves slightly differently. The burning velocities of the hydrocarbons are controlled by a reaction which provides reasonable values of the activation energies and probably precedes the sudden development of chain branching.


2006 ◽  
Vol 1 (3) ◽  
pp. 270-283 ◽  
Author(s):  
Charilaos Papadopoulos ◽  
J. Andrew Doyle ◽  
Brian D. LaBudde

Purpose:The purpose of this study was to determine the relationship between various lactate-threshold (LT) definitions and the average running velocity during a 10-km and a 21.1-km time trial (TT).Methods:Thirteen well-trained runners completed an incremental maximal exercise test, a 10-km TT, and a 21.1-km TT on a motorized treadmill. Blood samples were collected through a venous catheter placed in an antecubital vein. Pearson's correlation coefficients were used to determine the relationship between the running velocity at the different LT definitions and the average running velocity during each TT. A dependent t test was used to determine statistical differences for the mean lactate response between the 2 running distances.Results:The LTDmax, the point on the regression curve that yielded the maximal perpendicular distance to the straight line formed by the 2 endpoints, was the LT definition with the highest correlation for both 10-km (r = .844) and 21.1-km TTs (r = .783). The velocity at the LTDmax was not, however, the velocity closest to the performance velocity for either distance. The mean running velocity at each LT was significantly different and tended to overestimate the mean TT performance velocities. The mean lactate concentration during the 10-km TT (3.52 ± 1.58 mmol) was significantly higher than during the 21.1-km TT (1.86 ± 0.90 mmol).Conclusion:These results indicate that a single LT point cannot be reliably associated with different running distances. Furthermore, these data suggest that a different methodology for estimating the LT that considers individual responses might be required for different running distances.


1956 ◽  
Vol 48 (1) ◽  
pp. 61-73 ◽  
Author(s):  
E. H. Callow ◽  
S. R. Searle

1. Carcasses of twenty-one fat cattle (from six cows, ten steers and five heifers, aged from 12 months to 12 years and representing seven breeds) have been dissected into nine joints, and the fatty and muscular tissue from each joint dissected and analysed for fat. The iodine number of each sample of fat has been determined.2. The fat in fatty tissues ranged from 28·3 to 96·7%, and in muscular tissues from 1·3 to 14·2%. Iodine numbers of fat from fatty tissues ranged from 34·0 to 69·0 and those of fat from muscular tissues from 45·9 to 73·2.3. By an analysis of covariance it has been possible to reduce these data to two families of regression lines—those for fatty tissue and those for muscular tissue.4. The equations for fatty tissue have the following generalized form:I.N. = AFT + CFT + JFT + b1F/FT,where AFT = 55·6 ± 1·7 and is a general constant,CFT is a series of constants for the various carcasses and ranges from +12·6 to −6·7,JFT is a series of constants for the various joints and ranges from +6·6 to −10·3,b1 is a general constant with the value −0·052(± 0·025),F/FT is the percentage of fat in any given sample of fatty tissue, and I.N. is its iodine number.5. A similar set of equations has been deduced for muscular tissue. Here, however, the relation between iodine number and percentage of fat is a rectangular hyperbola. Consequently, the inverse (Z) of the percentage of fat (in order to get a straight-line relationship) has been used.


Sign in / Sign up

Export Citation Format

Share Document