| Rice J. 1995, Mathematical Statistics and Data Analysis, Duxbury Press, Belmont. |
....vectors representing TCP sessions and n is the number of attributes measured. For our use, x represents a new TCP session and y is the cluster mean from an individual cluster of TCP sessions. Determining the natural bound of Euclidian distance values involves the use of Chebyshev s inequality [23]. The equation for this limit is, Pr x ks 1 k where x is a random variable with mean, and standard deviation, s. In this analysis, x is a single Euclidian distance, is the mean Euclidian distance values for a normal cluster and k is a multiplier that determines the significance ....
....determines the significance level. Chebyshev s inequality sets the natural bound on the variability of the points within a given cluster and is computed separately for each cluster. The value produced is a probability that the value, x, comes from a population with mean, and standard deviation, s [23]. The Euclidian distance between a given TCP session and a cluster can be compared to this bound to see if the distance is significant, i.e. outside the Chebyshev inequality bound. K can be set to approximate the typical significance of a known distribution. For example, if we set k = 4.47, then ....
[Article contains additional citation context not shown here]
J. A. Rice. Mathematical Statistics and Data Analysis. Wadsworth Publ. Co., 1995.
....distribution. 3. 1 Distribution of the Sample s Quantile Under assumption of independent sampling, the expected value of the p th sample s quantile is given by x p = F (p) The probability density distribution of the p th quantile of a random variable can then be derived as follows (see e.g. [14], section 3.7, p. 101) Let X (1) X (n) be the ordered observations from a i.i.d random variable. Let X (k) be the p th quantile where k = np if np is an integer, and k = bnp 1c if np is not an integer. The event x X (k) x dx occurs if k 1 observations are less than x, one observation ....
....can be written as M estimators. 4.1 Testing for Convergence to Normality We propose to (i) produce and (ii) analyze normal plots for increasing sample sizes n to test when and whether latency quantiles become convergent. We (i) apply the frequently used normal plot method (see e.g. Rice[14] p. 321 328) to a set of latency quantiles obtained from simulation runs with di erent seeds to the random number generator. This method only leads to qualitative results. We therefore (ii) enhance this method with a fully edged statistical test to obtain quantitative results (see [14] which can ....
[Article contains additional citation context not shown here]
J. Rice, Mathematical Statistics and Data Analysis, 2nd edition, Duxbury Press, 1995.
....of the flux from a given star in one hold. In principle, using the idealizations above, the probability of that measurement can be evaluated under the assumption of no occultation and under the assumption that an occultation has occurred. This allows construction of a likelihood ratio test ([Rice, 1995]) which would be optimal if the idealizations and assumptions held. For computational considerations we use a simple approximation to this test. For further discussion of a likelihood ratio test and tests based on multiple holds see [Liang, 2001] Consider first a single telescope. For a set of ....
Rice, J. (1995). Mathematical Statistics and Data Analysis. Duxbury Press.
No context found.
Rice J. 1995, Mathematical Statistics and Data Analysis, Duxbury Press, Belmont.
No context found.
John A. Rice, Mathematical Statistics and Data Analysis, Second Edition, Duxbury Press, 1995.
No context found.
J. A. Rice, Mathematical Statistics and Data Analysis. Duxbury Press, 1994, ch. 11, 12.
No context found.
J.A. Rice, Mathematical Statistics and Data Analysis, second ed., Duxbury Press, 1995.
No context found.
J. A. Rice, Mathematical Statistics and Data Analysis. Duxbury Press, 1994, ch. 11, 12.
No context found.
J. A. Rice, Mathematical Statistics and Data Analysis. Duxbury Press, 1994, ch. 11, 12.
No context found.
J. A. Rice. Mathematical statistics and data analysis. Duxbury Press, 1995. 2nd edition.
No context found.
J.A. Rice, Mathematical Statistics and Data Analysis, 2nd ed., Duxbury Press, 1995.
No context found.
J.A. Rice, Mathematical Statistics and Data Analysis, 2nd ed., Duxbury Press, 1995.
No context found.
J. Rice, Mathematical statistics and data analysis, 2 ed., Duxbury Press, 1995.
No context found.
John A. Rice. Mathematical Statistics and Data Analysis. Duxbury Press, 1995.
No context found.
J.A. Rice. Mathematical Statistics and Data Analysis. Duxbury Press, second edition, 1995.
No context found.
Rice, J.A., Mathematical Statistics and Data Analysis, 2nd ed., Duxbury Press, 1995.
No context found.
J. A. Rice, Mathematical statistics and data analysis, Duxbury Press, 1995.
No context found.
J.A. Rice, Mathematical Statistics and Data Analysis, 2nd ed., Duxbury Press, 1995.
No context found.
J. Rice, Mathematical Statistics and Data Analysis, Duxbury Press, Belmont, CA (1995).
No context found.
Rice, J.A., Mathematical Statistics and Data Analysis, 2nd ed., Duxbury Press, 1995.
No context found.
RICE, J. A. Mathematical Statistics and Data Analysis, second ed. Duxbury Press, 1994.
No context found.
J. A. Rice. Mathematical statistics and data analysis. Wadsworth, 1988.
No context found.
J.A. Rice, Mathematical Statistics and Data Analysis, Duxbury Press, 1995.
No context found.
J. A. Rice. Mathematical Statistics and Data Analysis. Duxbury Press, 1995.
No context found.
Rice, J. A. (1988). Mathematical Statistics and Data Analysis. Wadsworth and Brookes/Cole, Belmont, California.
First 50 documents Next 50
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC