SIPmath™ Modeler Tools

Getting Started Guide *

Rev. May 2, 2016

SIPmath™ Modeler Tools Version 3.1.45

* For more details please consult the SIPmath Modeler Tools Reference
 The SIPmath Modeler Tools operate in two primary modes:

Generate Mode
Random inputs are generated either with the RAND() function or the seed-able generator from Hubbard
Decision Research. This can quickly create interactive Monte Carlo simulations in Excel, in which thousands
of trials are run before the user’s finger leaves the <Enter> key.

SIP Library Mode

Stochastic library mode utilizes pre-computed Monte Carlo trials (SIPs) to represent the uncertainty in the
model inputs as auditable arrays of data. This allows results to be aggregated across applications. A SIP
Library may be either built into the model workbook or linked to as an external file, for example in the
cloud, so that many users share a common set of trials.

Both modes produce models that run in native Excel. The Tools merely assist in creating such models but
are not required to run them. Thus models may be easily distributed and shared across a very wide
audience.

Main Ribbon in
Windows and Mac
(Note Mac does not currently
include Import)

NOTE Excel Calculate Mode must be Automatic for Interactive Simulation

The Tools turn on Automatic Calculation by default, but this may be controlled under Sittings
Main Ribbon
The Initialize tool is where the modeler identifies the Stochastic Library
containing the Input SIPs for use in SIP Library mode and specifies the
number of trials to run if creating a model to run in Generate mode.
The Library Inputs tool is where the modeler identifies the model’s input
cells and links them to the desired input SIPs. This tool is not required in
Generate Mode.
The Generate Input tool allows the modeler to define input cells whose
values are computed by Excel formulas for 13 different standard
distributions. Rand() or seed-able HDR modes supported.
The Define Outputs tool is where the modeler identifies the model’s
output cells and links them to the data table to create the Output SIPs,
which are created in either mode.
The Clear tool allows the modeler to remove inputs and outputs from the
model.
The Graphs tool provides an easy way to create histograms and
cumulative distributions from the output SIPs in either mode. It may also
be used to graph pre-existing data arrays.
Windows
Main Ribbon continued
The Get Stats tool facilitates the use of Excel’s native statistical formulas
with the output SIPs. It is useful for formulas such as AVERAGE() and
STDEV() etc. Click button on a blank cell for instructions.
The Trial Info tool provides a simple way to step through the input SIPs or
HDR variates one trial at a time or view optional meta data in SIP Library
mode only.

The Import tool allows the modeler to import SIP Library files stored in
XML or CSV format.

The Save Results tool allows the modeler to convert a model’s simulated
data to an Excel format SIP Library.

The Settings tool specifies Automatic Calculation options, sets the default
number of Bins for charts, and activates Crystal Ball or @Risk interface if
they are installed,.
The About tool displays version and copyright information.
Generate Mode
1. Initialize SIPmath in generate mode
3. Specify model outputs

Specify number
of trials
2. Insert random numbers into model:

4. Graph and explore the output SIP

Choose HDR Seed-able or RAND() generator

Generating Random Inputs
“Anyone who attempts to generate random numbers by deterministic means is, of course,
living in a state of sin.” - John von Neumann (Mathematician and father of the modern computer)
The SIPmath Modeler Tools may either read Inputs as data in SIP Library Mode, or generate its
own results as demonstrated here. For any of the distributions listed in the Generate Input
dialog you may use three types of pseudo random number generators as described below.

1. The Excel RAND() function. This produces a different

simulation every time you press the calculate key. The
results are not repeatable or auditable.

2. The HDR generator. This is an experimental generator

from Hubbard Decision Research. It produces
repeatable results with native Excel formulas, without
needing to store entire SIPs of data, that is, it is seed-
able by changing the Start Variable ID.
It has undergone some testing, but is not
pedigreed. However, this is the recommended
generator for quick modeling. As newer and
more proven generators are discovered, they
will be added to the SIPmath Tools.

Select External Random cell.

3. You may generate SIP libraries
from your favorite pseudo Then point to the cell where you will place
random number generator or the input for your own random number
sources such as Random.org. using Library Input Mode.
SIP Library Mode
1. Initialize in the Library mode (current
or external workbook)

2. Designate input cells 4. Specify outputs

3. Build a model using

uncertain inputs

model
Get Stats
When you specify outputs, the results appear in the PM_table sheet in ranges with
the assigned names. You may apply any of Excel’s native statistical formulas to
these ranges.
For example, suppose you have
created four outputs in cells D9
through G9, named Portfolio1
through Portfolio4. To calculate
the variance of each portfolio
you could create the formula
=VAR(Portfolio1) in cell D10. But
this formula cannot be copied
because the range names are an
absolute reference.

The Get Stats button allows you

to copy statistical formulas as
shown on the right.
SIP Library Sheet Format
Libraries are free-format, with information conveyed in
Excel variable names & range names. Optional Indexed Meta Data Values
may be appended to the ends of the
Other Formats include XML and CSV. SIPs. The labels and index positions
must be stored in arrays with special
Number of realizations is stored in a names. In this case, for example, the
cell named: PM_Trials. NOTE Do NOT “Average” is stored one position
count the Meta Data, just the Trials beyond the last trial in the SIP

Range name:
PM_Meta

Library Provenance is stored in a Range name:

cell named: PM_Lib_provenance PM_Meta_INDEX

SIPs may be stored as Rows or

SIP Libraries created with SIPmath Columns, but must be named.
Tools group the rows containing SIP for example, the SIP in D10:D19
data for easy expanding and hiding is named Uniform
data. Use the + and – buttons in the
margins to show and hide data.

If a SIP includes Meta Data, then the SIP

together with its Meta Data should be
named with a name ending in .MD. For
example, the SIP and Meta Data in
D10:D21 is named Uniform.MD
 Stay in Touch

Questions?
Come visit ProbabilityManagement.org for the newer software distributions and examples

Check SIPmath Modeler Tools Reference Manual for more information

The SIPmath Modeler Tools Reference contains tutorials and a more detailed description

Submit suggestions & bug reports

Send an e-mail to Dave@probabilitymanagement.org with OS version, Excel version, SIPmath
Modeler version and exact sequence of events that need an improvement

Contribute for code & standards development

ProbabilityManagement.org is a non-profit organization depending on sponsor contributions
and volunteer work. If interested, please contact us at Melissa@probabilitymanagement.org

Copyright © 2012-2016 ProbabilityManagement.org