home *** CD-ROM | disk | FTP | other *** search
- ***************** Multiple regression data file format ************************
- * (Read in Monaco 9 with wide margins)
- *
- * Lines that start with an asterisk at the extreme left are comments,
- * and are ignored by the program.
- *
- * The first non-comment line contains the names of variables,
- * separated by commas (no spaces). There may be up to 20 variables.
- * Each name may be of any length, but the program will display only
- * the first 14 characters with the regression results.
- *
- * The non-comment lines that follow the variable-names line consist
- * of numeric data. The first number is the value of the first variable
- * for the first observation; the second number is the value of the second
- * variable for the first observation; and the last number on the last line
- * is the value of the last variable for the last observation. In other
- * words, the numbers are arranged in groups where each group contains the
- * data for one observation of all the variables; and within each group, the
- * data are ordered to correspond with the variable names. Although the data
- * doesn't have to be arranged in neat columns, it'll be easier to maintain it
- * that way. Missing values are not supported.
- *
- * Each number may be in normal (176.58) or in scientific (-1.7658E2)
- * notation. The number of digits before and after the decimal point, and
- * after the "E" or "e" (if any), can vary. Each number must be separated,
- * from the next by a comma and/or any number of spaces, tabs, or carriage
- * returns. There must be a separator (e.g., a carriage return) after the
- * last number on the last data line.
- *
- * The program imposes no limit on the number of observations.
- *
- * The data that follows is from a paper by J. W. Longley that appeared in the
- * Journal of the American Statistical Association, Sept. 1967, Vol. 62 No. 319.
- * It was used in a study of the the accuracy of multiple regression programs.
- * The regression coefficients for the data were hand-calculated to 12 digits.
- * The thus-derived coefficient for the "Price" variable, with "Dependent" as
- * the dependent variable, was +1.5061872271373E1.
- *
- * Here is the data:
- Price,GNP,Unemployment,Army,Nins,Year,Dependent
- 83.0 234289 2356 1590 107608 1947 60323
- 88.5 259426 2325 1456 108632 1948 61122
- 88.2 258054 3682 1616 109773 1949 60171
- 89.5 284599 3351 1650 110929 1950 61187
- 96.2 328975 2099 3099 112075 1951 63221
- 98.1 346999 1932 3594 113270 1952 63639
- 99.0 365385 1870 3547 115094 1953 64989
- 100.0 363112 3578 3350 116219 1954 63761
- 101.2 397469 2904 3048 117388 1955 66019
- 104.6 419180 2822 2857 118734 1956 67857
- 108.4 442769 2936 2798 120445 1957 68169
- 110.8 444546 4681 2637 121950 1958 66513
- 112.6 482704 3813 2552 123366 1959 68655
- * For the last 3 observations we illustrate some format variations:
- 114.2,502601, 39.31E02,2514 125368 1960 69564
- 115.7 518173 4806 2572
- 127852 1961 69331
-
- .001169E005 554894 4007, 2827 130081 1962000e-3 70551