Chambers, J. M., W. S. Cleveland, B. Kleiner & P. A. Tukey (1983). Graphical Methods for Data Analysis. Duxbury Press.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the Challenger s Disaster. 4.1 Example 1: Tasting Whisky Tasting whisky is a very complicated task, which is done principally with the nose, then by the tongue, etc. The taste of Whisky can be graded in 10 categories on the scale of 0 3 for each (3 being the highest) If you use a star plot [4] (also called a wheel) each category corresponds to a spoke of the wheel. When you finish the grading and join up the lines, a particular shape of wheel appears, which reflects the characteristics of the Whisky. Figure 1, shows two examples: on the left hand side the star plot of The Balevenie, ....

Chambers, J., Cleveland, W., Kleiner, B. & Tukey, P.: Graphical Methods for Data Analysis, Wadsworth, (1983).

....the visualization looks like a star. The user can arrange four up to about twenty axes within one instance of the Stardinates. Each axis represents one attribute of the data thus every data record is visualized by a line (record line) along the corresponding data points. In contrast to Star plots [4] this technique can display many records within one instance. Like Glyphs a number of instances of the Stardinates are displayed side by side. This enables the user to visualize highly structured and multidimensional data. We think that the principles of the Gestalt laws [1] are a good starting ....

Chambers, J., Cleveland, W., Kleiner, B., and Tukey P. Graphical Methods for Data Analysis. Wadworth.

....of the random variable time to target solution in five GRASP implementations. They showed that, given a target solution value, the time it takes GRASP to find a solution at least as good as the target fits a two parameter exponential distribution. Standard methodology for graphical analysis [11] was used to compute the empirical and theoretical distributions and to estimate the parameters of the distributions. We use the same methodology to study time to target value for GRASP and GP PR. Our objective is to show that these variants of GRASP have time to target value distributions that ....

....with the i th sorted running time (t i ) a probability p i = i ) 200, and plot the points z i = t i , p i ) for i = 1, 200. Tables 9 and 12 show the target values and the parameters estimated by the methodology for GRASP and GP PR, respectively. Following the methodology proposed in [11], we first draw the theoretical quantile quantile plot for the data to estimate the parameters of the two parameter exponential distribution. To describe Q Q plots, recall that the cumulative distribution function for the two parameter exponential distribution is given by F (t) 1 e (t ) # ....

[Article contains additional citation context not shown here]

J. M. Chambers, W. S. Cleveland, B. Kleiner, and P. A. Tukey. Graphical Methods for Data Analysis. Chapman & Hall, 1983.

....are potentially unbounded while flow rates are constrained by link bandwidths. A previous study of rate distributions at a web server suggested that the rate distributions were well described by a log normal distribution [2] To test that hypothesis, we use the quantile quantile plot (Q Q plot) [3] to compare the flow rate distribution with analytical models. The Q Q plot determines whether a data set has a particular theoretical distribution by plotting the quantiles of the data set against the quantiles of the theoretical distribution. If the data comes from a population with the given ....

J. Chambers, W. Cleveland, B. Kleiner, and P. Tukey, "Graphical Methods for Data Analysis," Wadsworth Int'l. Group, Belmont, CA, 1983.

....Problem Solving Methods. Of course, our earlier work only described problem paradigms as data analysis methods, but the idea is clearly more general than that. The problem domMn of exploratory data and pattern analysis contains many general Mgorithms that are applied to specific data [45] 13] [14] [26] and the develop ment was advanced by the abstractions existing in the field. There are many problem paradigms possible for a given analysis problem. They differ in viewpoint, assumptions, requirements, and behavior. Given an analysis question, it is usually possible to get more information ....

John M. Chambers, William S. Clevelmd, Beat Kleiner, Paul A. Tukey, Graphical Methods for Data Analysis, Wadsworth International (1983.

....of experimental algorithmics have come out recently [10, 8, 9, 6] In particular, 6] explains some of the rules presented here. However these papers have a wider range of topics and hence cannot be as detailed as the present paper. There are also entire books on presenting data graphically [5, 4, 17]. These books explain some of the rules considered here in a more general setting. The present paper is to summarize and specialize these rules for the types of diagrams most useful in algorithmics using a more domain specific terminology. However, the main emphasis of the above books is on ....

J. M. Chambers, W. S. Cleveland, b. Kleiner, and P. A. Tukey. Graphical Methods for Data Analysis. Duxbury Press, Boston, 1983.

....effect. On the other hand, the direction of the mean times has reversed for this experiment. This importantfindingwas investigated further using an inductive analysis. 6. 3 Inductive analysis and interpretation Figure 2 displays the spread of the collected times through the use of boxplots (see [12] for a full description) The first twoboxplots represent the times of the first experiment for the inheritance group (N = 27) and the flat group (N =23)#boxplots three and four represent the inheritance group (N =13) and the flat group (N = 14) times for the replication# finally,boxplots five ....

J. Chambers, W. Cleveland, B. Kleiner, and P.Tukey. Graphical methods for data analysis.Wadsworth International Group, first edition, 1983.

....upper and lower quartiles. The data s median is denoted by a bold line within the box. The dashed vertical lines attached to the box indicate the tails of the distribution; they extend to the standard range of the data (1. 5 times the interquartile range) All other detached points are outliers [5]. Finally, for expository convenience, we say that two distributions are different only if they are significantly different. 3.2 Data Reduction Data reduction is the manipulation of data after its collection. We have reduced our data in order to: 1) remove data that is not pertinent to our ....

J.M. Chambers, W.S. Cleveland, B. Kleiner, and P.A. Tukey, Graphical Methods for Data Analysis. Belmont, Calif.: Wadsworth Int'l Group, 1983.

....16K20 was found adequate to model the video sources. The results of the simulation are validated with the data by considering two sets of measures. The first is the matching of the distribution of the frame bit rates, as demonstrated by the visual agreement of the quantile quantile (QQ) plots [23]. The second measure used to validate the model is the losses occurring in a leaky bucket policing function. The policing is carried out for a discrete range of drain rates and bucket sizes. The drain rate is varied from the average rate to the peak rate of the VBR video sequence. The range of ....

J.M. Chambers, W.S. Cleveland, B. Kleiner and P.A. Tukey,"Graphical Methods for Data Analysis," The Wadsworth Statistics/Probability Series, Duxury Press, Boston, 1983.

....The accurate modeling of short term correlation structure is therefore a key ingredient in video traffic models. The other two measures we considered to validate our model are a visual agreement of the distributions of the frame bit rates as demonstrated by the quantile quantile (QQ) plots [28], and the expected value and variance of the number of bits of the video source over a given time interval. Representing the time interval in terms of the number of frames, let rs(m) be the sequence of values representing the aggregate number of bits in m successive frames. The sequence is ....

J.M. Chambers, W.S. Cleveland, B. Kleiner and P.A. Tukey,"Graphical Methods for Data Analysis," The Wadsworth Statistics/Probability Series, Duxury Press, Boston,

....of the six plots, the data (ordered values of the log(SIs) are shown on the vertical scale on the left, and the quantiles of the standard normal distribution are shown on the horizontal scale. A detailed account of the construction and interpretation of normal probability plots is provided by [5]. In our situation statistical theory indicates that the log(SIs) should be approximately normally distributed, and the large sample standard deviation should be about 0.28 if the coefficient of variation is 0.4. If the relation between the empirical and theoretical quantiles is linear, this ....

J.M. Chambers, W.S. Cleveland, B. Kleiner, and P.A. Tukey. Graphical Methods for Data Analysis. Duxbury Press, Boston, 1983. -

....values seems reasonable. Probabilities 3 2 1 0 1 2 0.0 0.2 0.4 0.6 0.8 1. 0 Quantiles Figure 4: A Legend Attempt Order statistics results (see Blom 1958, David 1972 and Hoaglin 1983) suggest a more plausible expression: P#x#=# i,a#=#i 1,2#a# for x = x #i# i = 1; nand0# a# :5 Chambers et al. (1983) suggest obtaining probability estimates for quantiles between the order statistics by linear interpolation and estimates beyond the sample extrema by extending the probabilities at the sample extrema. The estimates for the current graphics use this approach with a= 5. The software referenced by ....

Chambers, J. M. W. S. Cleveland, B. Kleiner, P. A. Tukey. 1983. Graphical Methods for Data Analysis, Wadsworth and Brooks/Cole, Pacific Grove, California.

....as is appropriate since this is a reasonable baseline for both Dose and Days, and the graph is deliberately not square since the two variables are on different scales. Since Days is integer valued, we have jittered it slightly that is, added a small amount of randomness to this variable (Chambers et al. 1983) to separate the points clearer without confusing their interpretation. The only data from the table that have been discarded in the graph are the ID numbers, but these do not appear to be relevant in the data or the subsequent analysis. For the graph, it might make more sense to plot Dose on ....

....over the graphical display so that small legible plots can be juxtaposed as necessary. The graphs in this paper are extremely simple, and serious exploratory data analysis benefits from many elaborations, including color, multiple linked (trellis) plots, and dynamic displays (see, for a start, Chambers et al. 1983, Cleveland, 1985, 1993) We have include only the simple plots here because these appear to be the most appropriate replacements for the tables we saw in the JASA articles. We expect that as statistical 11 researchers begin to think more seriously about displaying their results as comparisons, ....

Chambers, J. M., Cleveland, W. S., Kleiner, B., and Tukey, P. A. (1983). Graphical Methods for Data Analysis. Pacific Grove, Calif.: Wadsworth.

....by a variation of the other regressors. A partial influence is graphically demonstrated by a partial regression plot (also known as added variable plot ) where the x axis shows the regressor and the y axis shows the regressand, and both are adjusted for the remaining regressors (cf. 4.2. 6; Chambers 1983: 268ff) Since the supplementary coefficient is also the slope of a simple regression line, it can be depicted graphically, too. To achieve this, the x axis has to show the values of the adjusted and synchronized regressor and the y axis the original values of the regressand. Alternatively, a ....

Chambers, John M., [et al.] (1983): Graphical Methods for Data Analysis. Duxbury, Boston (Massachusetts).

....Series 209 Figure 6.53: Bar plot of the trading volume data in the dow time series. Dow Jones Industrial Average Sep 7 Sep 14 Sep 21 Sep 28 Oct 5 Oct 12 Oct 19 Oct 26 1987 150000 200000 250000 300000 350000 400000 450000 500000 550000 600000 Chapter 6 Menu Graphics 210 REFERENCES Chambers, J.M. Cleveland, W.S. Kleiner, B. Tukey, P.A. 1983) Graphical Methods for Data Analysis. Belmont, California: Wadsworth. Cleveland, W.S. 1979) Robust locally weighted regression and smoothing scatterplots. Journal of the American Statistical Association, 74: 829 836. Cleveland, W.S. 1985) The Elements of Graphing Data. Monterrey, ....

Chambers, J.M., Cleveland, W.S., Kleiner, B. & Tukey, P.A. (1983). Graphical Methods for Data Analysis. Belmont, California: Wadsworth.

No context found.

Chambers, J. M., W. S. Cleveland, B. Kleiner & P. A. Tukey (1983). Graphical Methods for Data Analysis. Duxbury Press.

No context found.

Chambers, J., Cleveland, W., Kleiner, B. & Tukey, P.: Graphical Methods for Data Analysis, Wadsworth, (1983).

No context found.

J. Chambers, W. Cleveland, B. Kleiner, and P. Tukey. Graphical Methods for Data Analysis. Chapman & Hall, 1983.

No context found.

J. M. Chambers, W. S. Cleveland, B. Kleiner, and P. A. Tukey. Graphical Methods for Data Analysis. Chapman & Hall, 1983.

No context found.

Chambers, J. M. et al., Graphical Methods for Data Analysis, Boston: Duxbury Press, 1983.

No context found.

J. M. Chambers et al. (1983) Graphical Methods for Data Analysis, Duxbury Press.

No context found.

J. M. Chambers et al. (1983) Graphical Methods for Data Analysis, Duxbury Press.

No context found.

Chambers, J. M., Cleveland, W. S., Kleiner, B., and Tukey, P. A. (1983), Graphical Methods for Data Analysis, Belmont, CA: Wadsworth. 66

No context found.

Chambers J. M., Cleveland W. S., Kleiner B. and Tukey P. A. (1983) Graphical Methods for Data Analysis. Wadsworth and Brooks/Cole, Pacific Grove.

No context found.

John M. Chambers, William S. Cleveland, B. Kleiner, and P. A. Tukey. Graphical Methods for Data Analysis. Wadsworth, Monterey, California, 1983.

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