home *** CD-ROM | disk | FTP | other *** search
- -The data editor is designed for efficient entry of data for
- further analysis.
-
- -Data will be in the form of variables, with a column for each
- variable.
-
- -Commands are issued either through the menu or from the keyboard
- on the command line.
-
- -The first set of commands deals with columns. These allow you to
- insert a column, blank a column, or delete columns. If you choose
- the "Del Many" option you will be asked for the number of columns
- to delete. That number of columns, starting at the column in which
- the cursor resides and moving to the right, will be removed.
- Blanking a column does not remove it but just erases the data. -
- You may either load or save an individual column, as well. This
- allows you to save a data variable from one data set and bring it
- into another.
-
- -The actions available for rows are the same as those for columns,
- except that you cannot save or load a row. One additional row
- operation is available. The "delete Marked" option checks the
- value of each row in the column in which the cursor resides. If
- the value is other than zero then the row is removed. In no case
- will the program reduce the number of rows below 1.
-
- -The decimals selections are fairly straight forward. They simply
- affect how the data will be displayed. If the "Global" option is
- checked then any action will apply to all columns. If not, then
- the action affects only the column in which the cursor currently
- resides.
-
- -The "Item" menu allows for the insertion or deletion of individual
- data items. This is useful if you find after typing in a whole
- series of data points that you forgot to enter the third one or
- that you entered the third one twice. It is also useful for
- creating lagged variables to be used elsewhere in the program.
-
- -The "Miscellaneous" section has naturally a variety of options.
- By selecting "Mean", "Std Dev", "Largest", "Smallest" or quartiles
- you will get the corresponding statistic for the variable in which
- the cursor resides.
-
- -The Auto right selection if chosen changes the default movement
- of the cursor to a move right when the RETURN key is pressed
- rather than down.
-
- -The "Help" option will bring up this help file (as you can see).
-
- -The "RELABEL" option allows you to change the labels which are at
- the left hand side of the screen.
-
- -The remaining options allow you to move to other menus.
-
- -Various operations can be performed on the variables by entering
- commands on the command line.
-
- The first group affect the variable in which the cursor is
- located. Assume that variable Y contains the cursor.
- The operations are:
-
- a+b Column Y is the sum of column a and column b
- a+x Column Y is the sum of column a and a constant "x"
- a-b Column Y is column a minus column b
- a-x Column Y is column a minus a constant x
- x-a Column Y is x minus column a
- a*b Column Y is the product of column a and column b
- a*x Column Y is the product of column a and a constant "x"
- a/b Column Y is column a divided by column b
- a/x Column Y is column a divided by a constant "x"
- x/a Column Y is x divided by column a
- a^b Column Y is column a to the power of column b
- a^x Column Y is a to power of constant "x"
- a<b Column Y is 1 if column a is less than column b,
- otherwise it is 0
- a<x Column Y is 1 if column a is less than a constant "x",
- otherwise it is 0
- a>b Column Y is 1 if column a is greater than column b,
- otherwise it is 0
- a>x Column Y is 1 if column a is less than a constant "x",
- otherwise it is 0
- a\b Column Y is the minimum of column a and column b
- a\x Column Y is the minimum of a and a constant "x"
- a|b Column Y is the maximum of column a and column b
- a|x Column Y is the maximum of a and a constant "x"
- a!b Swaps column a and b. The column where the cursor
- resides is not affected unless it is in column a or b
- a!x Column a is set equal to the constant "x"
- a@x Column Y is the accumulation of the values in column
- a, where interest is asumed to be at "x" percent. The
- first payment is assumed to be at the end of the
- period
- a@b Similar to a@x, except that the percent interest rates
- for each period are in column b
- a&x Similar to a@x, except that the payments are at the
- start of the period
- a&b Similar to a@b, except that the payments are at the
- start of the period
- a$x Column Y is the present value of all remaining periods
- in a, taken at interest rate x%, Payments are at the
- end of the period
- a$b Similar to a$x, except that the interest rates are
- given in column b
- a#x Similar to a$x, except that payments are at the start
- of the period
- a#b Similar to a$b, except that payments are at the start
- of the period
- a%x Calculates the rate of interest, such that the present
- value of the values in column a equal the constant
- "x". Payments in a are assumed to be at the end of
- each period
- a%b Column Y is set equal to a series of rates, such that
- the present value of the items in a are equal to the
- values in b
-
- -In addition, there are commands that operate on one column only:
- COUNTERxx Creates a variable in column Y which has the natural
- numbers from 1 to xx in it
- RANDOMxx Creates a variable in column Y with xx random numbers
- EXPONENTIALxx Creates a variable with xx exponentially distributed
- deviates.
- NORMALxx Creates a variable with xx normally distributed
- deviates
- GAMMAxx,kk Creates a variable with xx random deviates
- distributed with a gamma distribution of order kk
- POISSONxx,nn Creates a variable with xx random deviates
- distributed with a poisson distribution with mean nn
- BINOMIALxx,nn,pp Creates a cariable with xx random deviates
- distributed binomially with nn trials and pp
- probability of success.
- deg(a) Column Y is a conversion of column a from radians to
- degrees
- rad(a) Column Y is a conversion of column a from degrees to
- radians
- mult(a) Column Y is the running product of column a
- rot(a) Column Y is the reverse order of column a
- fact(a) Column Y is the factorial of the values in a
- abs(a) Column Y is the absolute value of a
- rnd(a) Column Y is a random real number from 0 to the value
- in column a
- sqr(a) Column Y is the square root of column a
- sin(a) Column Y is the sine of column a. Radians are assumed
- cos(a) Column Y is the cosine of column a
- tan(a) Column Y is the tangent of column a
- asin(a) Column Y is the arc sine of column a
- acos(a) Column Y is the arc cos of column a
- asec(a) Column Y is the arc secant of column a
- atn(a) Column Y is the arc tangent of column a
- log(a) Column Y is the natural log of column a
- log10(a) Column Y is the log to base 10 of column a
- exp(a) Column Y is "e" to the power of a
- exp10(a) Column Y is 10 to the power of column a
- int(a) Column Y is the next lower integer of a
- sinh(a) Column Y is the hyperbolic sine of a
- cosh(a) Column Y is the hyperbolic cosine of a
- tanh(a) Column Y is the hyperbolic tangent of a
- asinh(a) Column Y is the arc hyperbolic sine of a
- acosh(a) Column Y is the arc hyperbolic cos of a
- atanh(a) Column Y is the arc hyperbolic tan of a
- ndist(a) Column Y is height of normal for value in a
- cundist(a) Column Y is cumulative probability of value in a
- arccundist(a) Inverse of cundist(a)
- standardize(a) Column Y is formed from "a" by creation of variable
- with mean of zero and standard deviation of 1
- percent(a) Column Y is percent that value in "a" is of total in
- "a". Negatives are ignored.
- percentsum(a) Column Y is the total percent that values in "a"
- up to the current represent of the total in "a".
-
- amort(amount,term,conv,payts,intr) amortizes a loan of amount
- "amount" over a period of "term" years.
- interest is convertible "conv" times per year
- there are "payts" payments per year (ie monthly is 12)
- the interest rate is specified in percent by "intr"
-
- These commands are relatively quick in execution since they are
- not actually parsed as a line of code would be. As you have
- probably noticed, they are all simple operations. B/STAT also has
- a general parser. If the regular commands cannot be used, the
- program reverts to the parser section. The parser section allows
- many, but not all, of the commands above. It allows certain
- modifiers, as well, which are not possible in the simple commands
- above. The mathematical operations supported are:
- + - plus and minus
- ^ exponentiation
- * / multiply and divide
- < > = logical less than greater than and equals
- \ minimum
- | maximum
-
- Three other logical operators are:
- OR logical or
- AND logical and
- XOR exclusive or
-
- The functions may be combined in any way. Operator hierarchy is
- standard algebraic. When in doubt use brackets. An example of the
- use of these functions might be the following:
-
- a*(( b<6) and (c>12))
-
- This would create a variable with the same value as variable a
- only for the points where b is less than 6 and c is greater than
- 12
-
- Another example is
-
- (exp(a)-exp(-a))/(exp(a)+exp(a))
-
- This formula is actually the formula for TANH(a) which would
- accomplish the same thing.
-
- OTHER Commands
-
- >axx Causes the cursor to go to column "a" row "xx"
-
- flip Swaps the rows and columns of the data matrix.
-
- Cursor movements can be made using the arrow keys. Holding down
- the "Control" or "Shift" key and pressing the arrow keys causes
- movement of a full page. If you hold down both the "Control" and
- "Shift" keys simultaneously and press an arrow key then you will
- go to the end of the data in that direction.
-
- sort(a) Sorts column a in ascending order. Y is unused
-
- sortall(a,b,c) Sorts whole data set with keys a,b,c. Not all keys
- are needed, but at least 1 must be used
-
- blank Column Y is blanked out
-
- shrink For any occurences of the same value in column Y
- the rows are added together. There will thus be 1
- row for each unique value in column Y.
-
- label The values in column Y become the labels in the
- label column.
-
- find(x) Searches from the current cursor position to the
- next occurrence of the value "x"
-
- outprod(a,b,*) creates a matrix of data with the number of rows as
- given by the number of values in column a and the
- number of columns by the number of values in b.
- The operator is the third element.
- Available operators are
- +,-,/,*,^,<,>,|,\
-