-
作者:GEARY, RC
摘要:If one of 2 measured variables is controlled, but subject to errors such as measuring or reading errors, and if the true relationship is linear, standard regression formulas and associated significance tests are exact. The second variable is assumed to be uncontrolled and also subject to error. In a similar situation if the true relationship is non-linear, standard methods are not appropriate. Methods are given for deriving correct formulas which are asymptotically exact. They are illustrated ...
-
作者:DURBIN, J
摘要:A method is given for incorporating into a multiple regression problem an extraneous estimate of one of the coefficients and its variance. It is shown to be more efficient than the usual procedure of using the estimated coefficient directly to correct observed values of the dependent variate. A numerical example is given.
-
作者:KATZ, L
摘要:The formulas given are good approximations for populations containing between 25 and 500 individuals. Tables already published give confidence intervals for populations outside this size range. Several examples are worked out.
-
作者:KING, EP
摘要:The author shows that the proper significance levels of several existing tests for rejecting outlying observations in a sample are actually about twice the levels previously given.
-
作者:PROSCHAN, F
摘要:A confidence interval is associated with a specified probability that the interval contains the true mean. A tolerance interval is associated with a specified probability that the interval contains a fixed proportion or more of the population values. The author gives a comprehensive discussion of the 2 types of intervals with several practical examples. Tables for computing interval factors are given and an interesting relationship between the 2 intervals is established.
-
作者:ANDERSON, RL
-
作者:EPSTEIN, B; SOBEL, M
摘要:In biological work life testing may refer to expts. in which a group of animals is exposed to some stimulus (irradiation, drugs, etc.) and the time required for each animal to attain a certain response (death, a physiological threshold, etc.) is recorded. The authors discuss methods for estimating mean response times in which response times for only r out of n animals in a group are required. They show that for certain distributions, estimates based on r out of n response times are just as acc...
-
作者:BROWNLEE, KA; HODGES, JL; ROSENBLATT, M
摘要:From new information concerning the behavior of the up-and-down method with small samples, the authors conclude that in many expts. it is superior to the probit method forestimating an L.D.50. Since animals must be tested sequentially or at best in sequential series, the up-and-down expt. will run from 2 to 5 times as long as a probit expt. In most other respects it is superior to the probit method. For estimates of approx. equal accuracy, the up-and-down method (as opposed to the probit metho...
-
作者:GARVY, G
-
作者:BURKHEAD, J