I am trying to develop a function that will count the number of runs within a range with one run being defined as when the value encounters a sign change (i.e. from positive to negative or negative to positive). Consider the following range: starting in cell A1 and continuing down to cell A10 I have these numbers: 125, 200, 175, -50, -75, 135, -65, -225, -90, and 115. The first three values (125, 200, 175) are positive and are considered as one run (a winning streak). The next two values are negative (-50, -75) thus are considered another run (a losing streak). The next value (135) is positive and is considered another run. The next three values (-65, -225, -90) are negative and thus is considered another run. Finally, the last value in the range (115) is positive, thus is considered another run. Therefore, for this example range the total number of runs 5. I have developed functions to calculate the longest winning and losing streaks, but I am really hitting a wall developing a function to calculate the total number of runs within a range. Any suggestions on how to approach this problem would be greatly appreciated. Thanks.

adj.cf name freq cum.freq(%) rank
0 MARY 2.629 2.629 1
.02629 PATRICIA 1.073 3.702 2
.03702 LINDA 1.035 4.736 3
.04736 BARBARA 0.980 5.716 4
.05716 ELIZABETH 0.937 6.653 5

=VLOOKUP(RAND(), $A$2:$B$6, 2)
or if you use the actual data
=VLOOKUP(RAND(), $A$2:$B$4275, 2)

http://www.excelforum.com/showthread.php?t=628930
shows similar question with Letters of alphabet

A1 is one cell above the input list
[n] is the upper bound for the number of input values + 1
[tweakable constant] controls the decay in probabilities for the exponential function. 0.1 ‘works’ over 10,000 cells, 0.9 can be seen to ‘work’ for a shorter list. 0.1 is very ‘slow’ decay in probability.

Poor Mary ain’t what she used to be!

From these charts, you can easily pinpoint when I was born.

BTW, here’s where I got the names from:

http://www.census.gov/genealogy/names/names_files.html

And each name has a frequency.

It’s making a lot more sense now.

I must also admit I found it surprising Mary was at the top of the example list. The 2005 data for the USA listed at:
http://www.socialsecurity.gov/OACT/babynames/

Top ten for the 2000s
http://www.socialsecurity.gov/OACT/babynames/decades/names2000s.html
Emily 149,420 1.2525%
Madison 123,729 1.0372%
Hannah 110,081 0.9228%
Emma 106,428 0.8921%
Ashley 91,644 0.7682%
Abigail 89,848 0.7532%
Alexis 89,512 0.7503%
Olivia 88,971 0.7458%
Samantha 88,669 0.7433%
Sarah 85,747 0.7188%

If the list is in, say, cells A1 to A10, then how about making your selections with a formula like this:

=OFFSET($A$1,TRUNC(RAND()^2*COUNTA($A$1:$A$10),0),0)

The COUNTA is for scalability of the list. Set the power of 2 higher to skew the selection more to the top of the list, or between 0 and 1 to skew the selection to the bottom of the list.

Cheers,

XLurker

I should add, if you want a working copy of the spreadsheet I’d gladly mail it.

M