A P P E N D I X
1
@Risk Crib Sheet
1337
#Simulations
Leave this at 1 unless you have a
RISKSIMTABLE functon in the spread-
sheet. In this case, choose #Simulations to equal the number of values in SIMTABLE. For
example, if we have the formula
RISKSIMTABLE({100,150,200,250,300}) in cell A1,
set #Simulations to 5. The first simulation will place 100 in A1, the second simulation
will place 150 in A1, and the fifth simulation will place 300 in A1. #Iterations will be run
for each simulation.
Pause on Error
Checking this box causes @Risk to pause if an error occurs in any cell
during the simulation. @Risk will highlight the cells where the error occurs.
Update Display
Checking this box causes @Risk to show the results of each iteration on
the screen. This is nice, but it slows things down.
See Figure 2. The Sampling tab options are as follows.
Sampling Type
While a little slower, Latin Hypercube sampling is much more accurate
than Monte Carlo sampling. To illustrate, Latin Hypercube guarantees for a given cell that
5% of observations will come from the bottom 5th percentile of the actual random vari-
able, 5% will come from the top 5th percentile of the actual random variable, etc. If we
choose Monte Carlo sampling, 8% of our observations may come from the bottom 5% of
the actual distribution, when in reality only 5% of observations should do so. When sim-
ulating financial derivatives, it is crucial to use Latin Hypercube.
Standard Recalc
If you choose Expected Value, you obtain the expected value of the ran-
dom variable unless the random variable is discrete. Then you obtain the possible value
of the random variable that is closest to the random variable’s expected value. For in-
stance, for a statement
RISKDISCRETE({1,2,},{.6,.4})
the expected value is 1(.6)
2(.4)
1.4, so Expected Value enters a 1.
If you choose the Monte Carlo option,
when you hit F9, all the random cells will re-
calculate. This makes it much easier to understand and debug the spreadsheet.
Thus, with
Monte Carlo selected,
RISKDISCRETE({1,2,},{.6,.4})
will return a 1 60% of the time and a 2 40% of the time.