List of Figures

Chapter 1. Prelude: understanding data with gnuplot

Figure 1.1. Number of finishers vs. time to complete (in minutes)

Figure 1.2. A DLA cluster of N=50,000 particles, drawn with gnuplot

Figure 1.3. Time required to grow a DLA cluster

Figure 1.4. Time required to grow a DLA cluster in a double-logarithmic plot, together with an approximate mathematical model

Chapter 2. Tutorial: essential gnuplot

Figure 2.1. Your first plot: plot sin(x)

Figure 2.2. An unsuitable default plot range: plot sin(x), x, x-(x**3)/6

Figure 2.3. Using explicit plot ranges: plot [][-2:2] sin(x), x, x-(x**3)/6

Figure 2.4. Plotting from a file: plot "prices" using 1:2, "prices" using 1:3

Figure 2.5. Introducing styles and the title keyword: plot "prices" using 1:2 title "PQR" with lines, "prices" using 1:3 title "XYZ" with linespoints

Figure 2.6. Any column can be used for either the x or y axis: plot "prices" using 2:3 with points.

Chapter 3. The heart of the matter: the plot command

Figure 3.1. Current athletic world-record running times as a function of the competition distance, for both men and women. The data is shown in listing 3.1.

Figure 3.2. Plotting a function for different values of a parameter: plot g(x,1), g(x,2), g(x,3) where g(x, s) = exp(-0.5*(x/s)**2)/s

Figure 3.3. The average speed of the current world record holder, as a function of the distance. The speed has been calculated from the distance and the finishing time as part of the plot command, using an inline transformation: plot "records" u 2:($1/$2) w lp. (See listing 3.1 for the data.)

Figure 3.4. Interpolating splines: both csplines and mcsplines interpolate the data points smoothly, but mcsplines doesn’t “overshoot” the data. In contrast to splines, which are locally defined, the Bézier curve constitutes a global approximation. plot "splines" u 1:2 w p pt 6, "" u 1:2 smooth csplines, "" u 1:2 smooth mcsplines,"" u 1:2 smooth bezier. (The directive pt 6 selects the shape of point symbols—see section 9.2.1.)

Figure 3.5. Influence of the weight parameter on approximate splines. As the weight increases, the curve approximates an interpolating spline; as the weight goes to zero, the curve approaches a straight line: plot "splines" u 1:2 w p pt 6, "" u 1:2:(10e6) s acs, "" u 1:2:(10) s acs, "" u 1:2:(1./10) s acs, "" u 1:2:(10e-6) s acs.

Figure 3.6. Visualizing point distributions using kernel density estimates and the cumulative distribution function: plot "measurements" u 1:(-0.1) w p pt 7, "" u 1:(1/12.) s kdens bandwidth .1, "" u 1:(1/12.) s kdens bandwidth 0.2, "" u 1:(1/12.) s cumul. The calculated default bandwidth is 0.58. See listing 3.4 for the data.

Chapter 4. Managing data sets and files

Figure 4.1. Different types of graphs, all based on a single data file. Clockwise from top left: a time-series plot of the number of orders per day, a histogram showing the distribution of the number of items per order, a parallel-coordinates plot of several order attributes (with ground-shipping orders highlighted), and a scatter plot of order value versus order weight.

Figure 4.2. Gnuplot treats missing values differently, depending on the value of the datafile missing option and the plot syntax. The data file is shown in listing 4.5.

Chapter 5. Practical matters: strings, loops, and history

Figure 5.1. Demonstrating string functions and the with labels plot style: plotting the Unix password file with gnuplot. See listing 5.1 for the data and listing 5.2 for the commands.

Chapter 6. A catalog of styles

Figure 6.1. Gnuplot automatically chooses a different line style for each curve.

Figure 6.2. The four core styles: with points, with lines, with linespoints, and with dots

Figure 6.3. Variants of the with linespoints style. Top to bottom: default behavior, using a (regular) character as symbol, omitting every second symbol, and using a character from an extended range of glyphs while omitting two out of every three points. Note that in the last case, the lines have been erased to make room for the characters by giving a negative argument to the pi keyword.

Figure 6.4. Different plot styles showing uncertainty in the data. From top to bottom: connected symbols using errorlines, unconnected symbols using errorbars, ranges indicated as boxes using boxxyerrorbars, and errors on top of a histogram using boxerrorbars.

Figure 6.5. Styles for time series: financebars and candlesticks. Filled bars indicate that the closing value is less than the opening value for the current record.

Figure 6.6. The three step styles. The same data set is shown three times (vertically shifted). Individual data points are represented by symbols; the three steps styles are shown with continuous lines. Note how different the same data set can appear, depending on the exact location of the vertical steps.

Figure 6.7. Box and impulse styles. The widths of boxes can be set globally or for each box individually. The second data set uses a fixed width (enclosed in parentheses in the using directive); the third one reads values for variable box widths from file.

Figure 6.8. Shading the area between two curves in a single data set: plot "data" u 1:2:3 w filledcurves. You can select only those areas where one of the curves is above or below the other curve. (See the text for details.)

Figure 6.9. Encoding additional information through symbol size or textual labels: pointsize variable and with labels. The corresponding data file is shown in listing 6.1.

Figure 6.10. Further capabilities of the with labels style: plotting a point symbol together with the text label, and changing the color, size, and orientation of the label

Figure 6.11. The data from listing 6.2, plotted using with vectors, with circles, and with ellipses. See listing 6.3 for details of the plot command.

Figure 6.12. All-in-one visualization of the data set in listing 6.4: temperature and dew point as solid lines, atmospheric pressure and its trend with symbols, visibility with grey bars; wind speed and direction across the top, cloud cover along the bottom. See listing 6.5 for the commands.

Chapter 7. Decorations: labels, arrows, and explanations

Figure 7.1. Providing minimal context for a plot using a title and axes labels. See listing 7.1.

Figure 7.2. The parts of a gnuplot graph: canvas, borders, and margins

Figure 7.3. Stacking order of layers, front to back. Within each layer, objects are drawn first, followed by labels and arrows. Curves are drawn in the order in which they appear in the plot command.

Figure 7.4. The parameters that control the shape of an arrowhead

Figure 7.5. Different arrow forms and the commands used to generate them. See listing 7.2 for the last one.

Figure 7.6. Using set key autotitle columnhead, key entries are taken from the first non-comment line in the data file. See listing 7.3 for the data file.

Figure 7.7. Explanations can be placed at the beginning or end of the curves rather than grouped together in a global key.

Figure 7.8. A complicated plot. See listing 7.4 for the commands used to generate it.

Chapter 8. All about axes

Figure 8.1. Using multiple axes on a plot to compare two different quantities side by side. (See listing 8.2 to find out how this plot was made.)

Figure 8.2. Two data sets showing daily temperatures. Note that the data sets use different units.

Figure 8.3. Linking two axes using link. See the text for details and listing 8.3 for the commands.

Figure 8.4. Nonstandard placement of coordinate axes and tic marks. See listing 8.6.

Figure 8.5. Allometric scaling: larger animals have a slower heart rate. Notice the customized tic labels along both axes. See the text for details and listing 8.7 for the commands.

Figure 8.6. Tic marks at multiples of π and Greek letters used for the tic labels. See listing 8.10.

Figure 8.7. A time-series plot, using set xmtics to use months for the labels along the horizontal axis. Also see listings 8.11 and 8.12. The data is shown in listing 8.9.

Figure 8.8. Men’s world records, together with the date when the record was set. The dates are read from the data file, reformatted, and placed onto the graph. See the text for details and listing 3.1 for the data.

Chapter

Color figure 1. Alpha shading and transparency. When a partially transparent color is added to an existing background, the added (foreground) color tends to prevail; adding it to a background of the same hue just increases the intensity. In the left panel, this effect is used to visualize point density in a dense data set. All data points are drawn with both blue disks and red rims. Because contributions from overlapping points add visually, regions of high point density show up as areas of high color intensity. Regions where the density is high enough for the rims to contribute significantly appear red. The right panel demonstrates what kinds of color mixtures you can expect when several colors are added together. (See chapter 9 for details.)

Color figure 2. Using color gradients to visualize data values. In the left panel, color is used to indicate the order in which particles were added to the cluster. The right panel shows a section of the complex plane, including part of the Mandelbrot set (black) and its fractal boundary. Color is used to visualize the number of iteration steps before the Mandelbrot iteration diverged. (See chapter 9 and appendix D.)

Color figure 3. The three built-in color sequences that can be selected using set colorsequence, and the custom sequence that was used for the color illustrations in this book. (See chapters 9 and 12 for details.)

Color figure 4. Several color gradients (or palettes) for visualization purposes, as discussed in appendix D. Each panel in the figure displays a different palette.

Color figure 5. Using stylesheets to change the appearance of a plot. The plot in this figure has been prepared with a stylesheet that uses relatively bright colors, but thin lines. (Compare color figure 6. Also see listing 12.7 in chapter 12 for details.)

Color figure 6. Using stylesheets to change the appearance of a plot. The plot in this figure has been prepared with a stylesheet that uses soft pastel colors, but relatively thick lines. (Compare color figure 5. Also see listing 12.8 in chapter 12 for details.)

Color figure 7. Using a combination of several graphical techniques to represent a complicated, multivariate data set. The figure shows a parallel-coordinates plot of the entire glass data set. Because of the relatively large number of records, all lines are drawn partially transparent so they don’t obscure each other. A subset of records has been highlighted in a different color: records with a Calcium (Ca) content between 9 and 10 have been selected for highlighting. (See chapter 14 for details.)

Color figure 8. A false-color plot prepared using a color gradient. The color of each square represents the number of defects produced by each machine on each day of the month. Missing squares indicate that the corresponding machine wasn’t used on that day. Many different aspects of the operation are visible in this graph. You should be able to recognize weekends and two clusters of machines that seem more error-prone than the others. This data set contains a handful of outliers (that is, isolated instances with an excessive number of defects), which are drawn in black. (See appendix D for details.)

Chapter 9. Color, style, and appearance

Figure 9.1. Alpha blending in action. The color that is added to an existing background tends to prevail; adding it to a background of the same hue just increases the intensity.

Figure 9.2. Using alpha blending to visualize point density in a dense data set. All data points are drawn both with blue disks and red rims. Because contributions from overlapping points add visually, regions of high point density show up as areas of high color intensity. Regions where the density is high enough for the rims to contribute significantly appear red.

Figure 9.3. A DLA cluster (see section 1.1.2) using the HSV color model to indicate the order in which particles were added to the cluster

Figure 9.4. Using indexed color to visualize discrete values or levels in a data set. For a set of current “supercars,” the date of presentation and the available horsepower are visualized through the location on the graph, and the number of cylinders is represented via color. Compare listing 9.2 for the data and listing 9.1 for the commands.

Figure 9.5. Gradient-mapped color. The control curve changes its color if the monitored quantity exceeds its safe value.

Figure 9.6. The three built-in color sequences that can be selected using set colorsequence, and the custom sequence used for the color illustrations in this book. (See section 12.6 for details.)

Figure 9.7. Available point symbols

Figure 9.8. Top: Dashed lines resulting from various combinations of the standard dash symbols or given by an explicit numeric tuple (as shown on the left—the pattern for the bottom line is given by an explicit numeric tuple). Bottom: Effect of the different line-cap styles. The line end and the added cap are shown in different colors.

Figure 9.9. Controlling overall image size and aspect ratio with set size. Clockwise from top left: default settings, aspect ratio 10/7; set size ratio 1, full plot size, aspect ratio 1:1; set size 0.5, reduced plot on a full-size canvas, default aspect ratio; set size ratio -1, full plot size, x unit with the same apparent length as the y unit.

Figure 9.10. The set border command takes as its first argument an integer that encodes the choice of borders to switch on in the form of a bit mask. The graph indicates which borders correspond to which bit (in decimal representation) and also gives the overall mask value for two common cases: borders on all sides and borders on the bottom and left.

Chapter 10. Terminals and output formats

Figure 10.1. The standard test image to demonstrate terminal capabilities, shown here for the pdfcairo terminal and the default color sequence set colorsequence default

Figure 10.2. Output produced when various free fonts are requested on my computer—your version may look different. The first part of the test string contains a selection of characteristic regular characters, and the second part consists of common mathematical symbols and Greek letters.

Figure 10.3. Enhanced text mode. See listing 10.2.

Figure 10.4. The final appearance of the LaTeX document shown in listing 10.3. Note the labels using enhanced text mode in the included gnuplot graph.

Figure 10.5. The final appearance of the LaTeX document shown in listing 10.5. Note the true LaTeX labels and tic marks on the graph. Compare to figure 10.4.

Figure 10.6. A graph drawn using the tikz terminal. See listing 10.6 for the commands.

Figure 10.7. The dumb terminal

Chapter 11. Automation, scripting, and animation

Figure 11.1. Graph paper: an easy application of gnuplot’s loops. See listing 11.1.

Figure 11.2. Successive approximations to sin(x) using the sum facility

Chapter 12. Beyond the defaults: workflow and styles

Figure 12.1. The example graph using the stylesheet from listing 12.7: thin lines, bright colors, small symbols

Figure 12.2. The example graph using the stylesheet from listing 12.8: thick lines, pastel colors. Compare figure 12.1.

Figure 12.3. The example graph using the stylesheet from listing 12.9: black-and-white, dashed hairlines, cross-shaped symbols

Figure 12.4. The example graph using the stylesheet from listing 12.10: as different from the gnuplot defaults as possible

Chapter 13. Basic techniques of graphical analysis

Figure 13.1. Curb weight versus price for 205 different cars. See listing 13.1.

Figure 13.2. Finishing times (in minutes) for the winner of a marathon (up to the year 1990), together with the best straight-line fit. Will women overtake men in the coming years?

Figure 13.3. The same data as in figure 13.2, together with a weighted-splines fit. The fit is based only on points prior to 1990, but the actual finishing times for the following years are also shown. The softer spline clearly reveals the leveling off of the women’s results well before 1990.

Figure 13.4. Traffic patterns for a website. Daily hit count versus day of the year. Note the extreme variation in traffic over time.

Figure 13.5. The same data as in figure 13.4, but on a semi-logarithmic scale. Note how the high-traffic outliers have been suppressed and the low-traffic background has been enhanced. In this presentation, data spanning two orders of magnitude can be compared easily.

Figure 13.6. Bottom panel: hits per day over time (as in figure 13.4); top panel: change in traffic between consecutive days, divided by the total traffic. Note how the relative change (top panel) doesn’t exhibit any seasonal pattern, indicating that the relative size of the variation is constant.

Figure 13.7. Allometric scaling: the duration of an average resting heartbeat as a function of the typical body mass for several mammals. Note how the data points seem to fall on a straight line with slope 1/4.

Figure 13.8. Three ways to represent a distribution of random points: jitter plot (bottom), histogram (with boxes), and cumulative distribution function (dashed line)

Figure 13.9. An alternative to histograms: kernel density estimates using smooth kdensity. Curves for three different bandwidths are shown. A bandwidth of 0.3 seems to give the best trade-off between smoothing action and retention of details. Note how it brings out the secondary cluster near x=3.5. Individual data points are represented through the rug plot along the bottom.

Figure 13.10. A rank-order plot. Because there’s no natural ordering in the independent variable (in this case, the country names), the data has been sorted by the dependent variable to emphasize the structure in the data.

Figure 13.11. A rank-order plot displaying a primary and a secondary data set for comparison. The country names are sorted according to the primary data set (the area); the points in the secondary data are connected by lines to make them easier to distinguish. Note the logarithmic scale for the horizontal axes.

Figure 13.12. A pie chart

Chapter 14. Topics in graphical analysis

Figure 14.1. Web traffic (hits per hour). Compare listing 14.2.

Figure 14.2. A noisy time series and a smoothed trend line. The trend was calculated using single exponential smoothing: s=a*x + (1-a)*s, with a smoothing parameter of a = 0.5.

Figure 14.3. A control chart showing three very different quantities simultaneously. What’s wrong with this picture?

Figure 14.4. A control chart showing normalized metrics. The data is the same as in figure 14.3.

Figure 14.5. Number of daily calls to a call center. The figure shows the raw data and the smoothed trend line.

Figure 14.6. Same data as in figure 14.5. In addition to the smoothed trend line, the region of normal fluctuations is indicated (shaded); outliers outside this region are indicated with symbols. (See the text for details.)

Figure 14.7. A control chart. The target value for the observed metric is 5. The horizontal lines indicate the permitted range around the setpoint.

Figure 14.8. Same data as in figure 14.7, but shown together with the cumulative deviation from the intended setpoint of 5. The sudden and persistent increase in the cumulative error indicates that the system is shifted away from its target value—even though this isn’t immediately visible in the raw data.

Figure 14.9. Four different ways to present changes in composition. Each line corresponds to the number of units produced for one of four different products. The top row shows absolute numbers of units produced; the bottom row shows fractions relative to total production. In the left column, each line is drawn separately; in the right column, the lines are stacked, giving a cumulative view into the data. See listing 14.4.

Figure 14.10. A stacked graph of absolute production numbers. The curves have been sorted in such a way that lines showing the smallest variation are lower in the stack, leading to a more stable baseline to which the subsequent products are added.

Figure 14.11. Number of data points for each type of glass. Compare listing 14.6.

Figure 14.12. Distribution of values for each attribute. Compare listing 14.7.

Figure 14.13. Distribution of values for each attribute, shown separately for each type of glass. The sequence of glass types is the same as in figure 14.11; type Vhcl2 is ignored because there are no records for it. Compare listing 14.8.

Figure 14.14. Distribution of values for attribute Si, shown separately for each type of glass. The width of the central box is proportional to the number of data points for each level. Compare listing 14.9.

Figure 14.15. Scatter-plot matrix of all attributes, ignoring the type of glass. The upper half of the matrix shows individual points, and the lower half shows smeared-out and partially transparent symbols to facilitate the recognition of clusters. Compare listing 14.10.

Figure 14.16. Parallel-coordinates plot of the entire glass data set. Because of the relatively large number of records, all lines are drawn partially transparent, so as not to obscure each other. To accommodate all attributes, this graph was composed using multiplot mode (see listing 14.11).

Figure 14.17. Same as figure 14.16, but with a subset of records highlighted. Records with a Ca content between 9 and 10 have been selected for highlighting. Compare listing 14.12.

Figure 14.18. Annual sunspot numbers for the years 1700 through 2000. What can you say about the shape of the curve in this representation?

Figure 14.19. A different representation of the sunspot data from figure 14.18. The aspect ratio of each segment in the cut-and-stack plot is such that it banks lines to 45 degrees.

Figure 14.20. Inflow to and outflow from a storage tank. What’s the net flow to the tank?

Figure 14.21. Net flow to the storage tank. This is the difference between the inflow and the outflow (see figure 14.20).

Figure 14.22. Two different views of the same data set. The outer graph includes zero in the vertical plot range, and the inner graph displays only that part of the vertical axis that corresponds to data points. Both views are valid: the outer graph shows that the relative change in the data is small compared to its absolute value; the inner graph shows that the variation increases dramatically from left to right.

Chapter 15. Coda: understanding data with graphs

Figure 15.1. Schematic outline of a fighter airplane. Areas where bullet holes were found on machines returning from combat missions are shaded. Where would you recommend placing additional armor?

Appendix C. Surface and contour plots

Figure C.1. Creating three-dimensional plots using the splot command: splot [-2:2][-2:2] exp(-(x**2 + y**2))*cos(x/4)*sin(y)*cos(2*(x**2+y**2))

Figure C.2. The same plot as in figure C.1, but without the opaque surface effect. Use set hidden3d to enable the drawing of an opaque, nontransparent surface.

Figure C.3. Comparing different values of isosamples and hidden3d trianglepattern. Except for the values of those two options, these graphs are identical to figure C.1.

Figure C.4. The isosamples option controls the number of nodes used to draw the surface. Here, we use the default value of set isosamples 10, whereas in figure C.1, we used a much finer grid of set isosamples 30.

Figure C.5. The angles used when setting the view point with set view

Figure C.6. Different view points. Compare to figure C.1, where the same function is shown with set view 45,50.

Figure C.7. Adding contour lines at the base and on the surface using set contour both and set cntrparam levels incremental -0.3,0.1,0.4. The function is the same as in figure C.1.

Figure C.8. The function from figure C.1, plotted in contour view. See listing C.1 for the commands.

Figure C.9. The data from listing C.2 plotted using splot "grid" u 1:2:3 w linesp

Figure C.10. The splot command requires all data points to be supplied on a regular grid, with none missing. This figure shows what happens when the last point (at x=3, y=1) has been replaced with a not-a-number marker or omitted entirely. See listing C.2 for the original data file. The splot command is the same as in figure C.9.

Figure C.11. The same data as in figure C.9, but plotted after turning on smoothing using set dgrid3d 30,30 gauss 0.6,0.6

Appendix D. Palettes and false-color plots

Figure D.1. Adding color from a continuous palette to a surface plot. Compare figure C.1.

Figure D.2. Two examples of false-color plots. The function is the same as in figure D.1. The graph on the left was drawn using a simple palette defined by set palette defined (-1 'white', 0 'blue', 0 'red', 1 'white' ); the graph on the right uses the palette from panel N in figure D.4.

Figure D.3. The result of the test palette command for a palette defined with set palette defined ( 0 'web-green', 0.4 'goldenrod', 1 'red' ) or, equivalently, set palette defined ( 0 "#00c000", 0.4 "#ffc020", 1 "#ff0000" )

Figure D.4. Several example palettes: each panel in the figure displays a different palette. See the text for details.

Figure D.5. Selecting colors from a palette (see listing D.1). In this figure, the color position in the palette is fixed, and the colors are therefore not data-dependent.

Figure D.6. Selecting colors from a palette (see listing D.2). In this figure, the values of the data points are used to select their colors from the palette; the plot uses data-dependent coloring.

Figure D.7. Showing the number of defects per machine and day in a false-color plot. This plot uses the with points style. See listing D.4.

Figure D.8. The with pm3d style can create false-color plots even when the grid is distorted. Every tile is spanned by the data points at its four corners. See listing D.5 for the data.

Figure D.9. A false-color plot for a regular grid, created using the with image style. Notice that each tile is centered on a data point. The data is shown in listing C.2.

Figure D.10. A section of the complex plane, showing part of the Mandelbrot set and its fractal boundary. See listing D.6.

Appendix E. Special plots

Figure E.1. Choosing the layout direction in multiplot mode: rowsfirst downwards (left; this is the default) and columnsfirst upwards (right)

Figure E.2. A regular array of small plots created in multiplot mode. See listing E.1.

Figure E.3. A regular array of small plots. All individual plots are of the same size and therefore line up, although some of them have marginal notes whereas others don’t. See listing E.2.

Figure E.4. Showing two plots side by side using multiplot mode (see listing E.3)

Figure E.5. A larger graph with insets showing ancillary information. See listing E.4.

Figure E.6. A box-and-whisker plot, when drawn using only default appearance options

Figure E.7. Construction of a box-and-whisker plot from the data set. (The individual data points are shown on the left.)

Figure E.8. A serial box-and-whisker plot. The distribution of sepal lengths is shown separately for each type of flower.

Figure E.9. An array of serial box-and-whisker plots. The distribution of all four observed quantities is displayed for all flower types separately. (See listing E.6.)

Figure E.10. Parallel-coordinates plot for a single record

Figure E.11. Parallel-coordinates plot for the entire Iris data set

Figure E.12. A parallel-coordinates plot for the entire Iris data set. Records are color-coded based on the flower type.

Figure E.13. Election results as a time series. The data file is shown in listing E.9.

Figure E.14. Election results using set style histogram clustered. This is the same data set as in figure E.13.

Figure E.15. Election results using set style histogram rowstacked. This is the same data set yet again. Note the effect of the xtic() function, which is used to read x axis labels directly from the data file.

Figure E.16. Several histograms can be combined in a single graph using newhistogram.

Appendix F. Higher math

Figure F.1. A simple Lissajous figure, drawn using parametric mode. See listing F.1.

Figure F.2. A more complicated figure drawn with parametric mode. Plotted are the functions sin(0.99*t)-0.7*cos(3.01*t) for the horizontal coordinate and cos(1.01*t)+0.1*sin(15.03*t) for the vertical coordinate.

Figure F.3. A graph in polar mode. See listing F.2.

Figure F.4. Using a spherical coordinate system together with splot. On the left, the data is plotted as a regular two-dimensional plot using plot; on the right, it’s plotted (together with a grid) using set mapping spherical and splot. (See listing F.3.)

Figure F.5. A vector field, plotted using the with vectors style

Figure F.6. Same as figure F.5, but using data-dependent coloring to represent additional information in the graph

Figure F.7. The body whose coordinates were given in listing F.4. Load listing F.5 into gnuplot, using an interactive terminal, and then grab this figure with the mouse and try to rotate it!

Figure F.8. The Mandelbrot set, rendered and calculated in gnuplot. See listing F.6 for the complete implementation.

Figure F.9. Normal probability plots for two data sets: one that follows a Gaussian distribution, the other showing non-Gaussian behavior

Figure F.10. A normal probability plot, with secondary tic marks and grid lines added. (See listing F.8.)

Figure F.11. Data from a scattering experiment. How many peaks are there, where are they located, and what’s their width?

Figure F.12. The data from figure F.11 when fitting two peaks using the fit command. The agreement is quite good, but is there a third peak, centered near x=65?