Data Loading...
Syntax Flipbook PDF
truncate(#) truncates the data to this value and displays the plot in these units. ... The stem-and-leaf display provide
109 Views
73 Downloads
FLIP PDF 103.76KB
Title
stata.com stem — Stem-and-leaf displays
Syntax Remarks and examples
Menu Stored results
Description References
Options Also see
Syntax stem varname if in , options Description
options Main
prune round(#) truncate(#) digits(#) lines(#) width(#)
do not print stems that have no leaves round data to this value; default is round(1) truncate data to this value digits per leaf; default is digits(1) number of stems per interval of 10digits stem width; equal to 10digits /width
by is allowed; see [D] by.
Menu Statistics
>
Summaries, tables, and tests
>
Distributional plots and tests
>
Stem-and-leaf display
Description stem displays stem-and-leaf plots.
Options
Main
prune prevents printing any stems that have no leaves. round(#) rounds the data to this value and displays the plot in these units. If round() is not specified, noninteger data will be rounded automatically. truncate(#) truncates the data to this value and displays the plot in these units. digits(#) sets the number of digits per leaf. The default is 1. lines(#) sets the number of stems per every data interval of 10digits . The value of lines() must divide 10digits ; that is, if digits(1) is specified, then lines() must divide 10. If digits(2) is specified, then lines() must divide 100, etc. Only one of lines() or width() may be specified. If neither is specified, an appropriate value will be set automatically. 1
2
stem — Stem-and-leaf displays
width(#) sets the width of a stem. lines() is equal to 10digits /width, and this option is merely an alternative way of setting lines(). The value of width() must divide 10digits . Only one of width() or lines() may be specified. If neither is specified, an appropriate value will be set automatically. Note: If lines() or width() is not specified, digits() may be decreased in some circumstances to make a better-looking plot. If lines() or width() is set, the user-specified value of digits() will not be altered.
Remarks and examples
stata.com
Example 1 Stem-and-leaf displays are a compact way to present considerable information about a batch of data. For instance, using our automobile data (described in [U] 1.2.2 Example datasets): . use http://www.stata-press.com/data/r13/auto (1978 Automobile Data) . stem mpg Stem-and-leaf plot for mpg (Mileage (mpg)) 1t 22 44444455 1f 1s 66667777 1. 88888888899999999 00011111 2* 2t 22222333 444455555 2f 2s 666 2. 8889 3* 001 3t 3f 455 3s 3. 4* 1
The stem-and-leaf display provides a way to list our data. The expression to the left of the vertical bar is called the stem; the digits to the right are called the leaves. All the stems that begin with the same digit and the corresponding leaves, written beside each other, reconstruct an observation of the data. Thus, if we look at the four stems that begin with the digit 1 and their corresponding leaves, we see that we have two cars rated at 12 mpg, 6 cars at 14, 2 at 15, and so on. The car with the highest mileage rating in our data is rated at 41 mpg. The above plot is a five-line plot with lines() equal to 5 (five lines per interval of 10) and width() equal to 2 (two leaves per stem). Instead, we could specify lines(2): . stem mpg, lines(2) Stem-and-leaf plot for mpg (Mileage (mpg)) 1* 22444444 1. 556666777788888888899999999 2* 00011111222223334444 2. 555556668889 3* 0014 3. 55 4* 1
stem mpg, width(5) would produce the same plot as above.
stem — Stem-and-leaf displays
3
The stem-and-leaf display provides a crude histogram of our data, one not so pretty as that produced by histogram (see [R] histogram), but one that is nonetheless informative.
Example 2 The miles per gallon rating fits easily into a stem-and-leaf display because, in our data, it has two digits. However, stem does not require two digits. . stem price, lines(1) digits(3) Stem-and-leaf plot for price (Price) 3*** 291,299,667,748,798,799,829,895,955,984,995 010,060,082,099,172,181,187,195,296,389,424,425,453,482,499, ... (26) 4*** 5*** 079,104,172,189,222,379,397,705,719,788,798,799,886,899 6*** 165,229,295,303,342,486,850 140,827 7*** 8*** 129,814 9*** 690,735 10*** 371,372 11*** 385,497,995 12*** 990 13*** 466,594 500 14*** 15*** 906
The (26) at the right of the second stem shows that there were 26 leaves on this stem — too many to display on one line. We can make a more compact stem-and-leaf plot by rounding. To display stem in units of 100, we could type . stem price, round(100) Stem-and-leaf plot for price (Price) price rounded to nearest multiple of 100 plot in units of 100 33778889 3* 00001112222344455555667777899 4* 5* 11222447788899 6* 2233359 7* 18 18 8* 9* 77 10* 44 11* 45 0 12* 13* 056 5 14* 9 15*
price, in our data, has four or five digits. stem presented the display in terms of units of 100, so a car that cost $3,291 was treated for display purposes as $3,300.
4
stem — Stem-and-leaf displays
Technical note Stem-and-leaf diagrams have been used in Japanese railway timetables, as shown in Tufte (1990, 46–47).
Stored results stem stores the following in r(): Scalars r(width) r(digits) Macros r(round) r(truncate)
width of a stem number of digits per leaf; default is 1 number specified in round() number specified in truncate()
References Cox, N. J. 2007. Speaking Stata: Turning over a new leaf. Stata Journal 7: 413–433. Emerson, J. D., and D. C. Hoaglin. 1983. Stem-and-leaf displays. In Understanding Robust and Exploratory Data Analysis, ed. D. C. Hoaglin, C. F. Mosteller, and J. W. Tukey, 7–32. New York: Wiley. Tufte, E. R. 1990. Envisioning Information. Cheshire, CT: Graphics Press. Tukey, J. W. 1972. Some graphic and semigraphic displays. In Statistical Papers in Honor of George W. Snedecor, ed. T. A. Bancroft and S. A. Brown, 293–316. Ames, IA: Iowa State University Press. . 1977. Exploratory Data Analysis. Reading, MA: Addison–Wesley.
Also see [R] histogram — Histograms for continuous and categorical variables [R] lv — Letter-value displays