home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
SPACE 2
/
SPACE - Library 2 - Volume 1.iso
/
apps
/
559
/
spsheet.doc
< prev
next >
Wrap
Text File
|
1991-06-17
|
11KB
|
235 lines
-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
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
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
+,-,/,*,^,<,>,|,\