Data Tables & Monte Carlo Simulations in Excel – A Comprehensive Guide
This is a Guest Post by Hui, an Excel Ninja and One of the Moderators of our Forums. Please note that this post is unusually large by Chandoo.org standards.
============================================================
If anybody asks me what is the best function in excel I am drawn between Sumproduct and Data Tables, Both make handling large amounts of data a breeze, the only thing missing is the Spandex Pants and Red Cape!
How often have you thought of or been asked “I’d like to know what our profit would be for a number of values of an input variable” or “Can I have a graph of Profit vs Cost vs …”
This post is going to detail the use of the Data Table function within Excel, which can help you answer those questions and then so so much more.
 Introduction
 1 Way Tables
 2 Way Tables
 Monitor Multiple Variables
 Multiway Tables
 Monte Carlo Analysis
 Iterative Functions and Fractals
 Download Example Workbooks
INTRODUCTION
How often have you thought “I’d like to know what our profit would be for a +/ 10, 20 and 30 % variance in the costs” ?
This post is going to detail the use of the Data Table function within Excel, which can help you answer that question.
The Data Table function is a function that allows a table of what if questions to be posed and answered simply, and is useful in simple what if questions, sensitivity analysis, variance analysis and even Monte Carlo (Stochastic) analysis of real life model within Excel.
The Data Table function should not be confused with the Insert Table function.
DATA TABLE BASICS
The Data Table function is hidden away in different locations within different versions of Excel but apart from the menu location the functionality is the same throughout.
Where is the Data Table Function
Excel 2007/10
In Excel 2007 & 2010 go to the Data Tab, What If Analysis panel and select Data Table
Excel 9703
In Excel up to 2003 go to the Data Menu and select Table…
Both Excel 9703 and 2007/10 then bring up the same Data Table dialog box.
… and this simple dialog box is all empowering ?
Yes !
Blue Sky Mine Co
For demonstration of the Data Table function I am going to use a simple profit model of a Gold Mine, “The Blue Sky Mine Co”. This is a fictitious mine but provides a simple model which we can use the data Table function to analyse.
It consists of 6 input variables and a simple cost and revenue model to produce a profit.
In our Blue Sky Gold Mine Co model, we can see that if we mine and treat 1,000,000 t of gold ore containing 1.68 g/t gold, we will make A$ 5.452M profit. But what if the inputs change ?
1 WAY DATA TABLES
Lets make a 1 Way Table with our Blue Sky Gold Mine Co example.
This is shown in the attached Excel Workbook on the “1 Way” Tab or 1 Way Example
In our Blue Sky Gold Mine profit calculation example, we can see that if we mine and treat 1,000,000 t of gold ore containing 1.68 g/t gold, we will make A$ 5.452M profit. But what if the grade is more or less than that value of 1.68 g/t ? After all it is only a geological estimate.
This is what the Data Table function is made for.
Next to the model add a couple of columns as shown in blue
Note: Throughout this post you will see the use of 1E6 in formulas which is simpler to write than 1,000,000.
The first column is a list of values that will be applied to each iteration of the Column Input Cell
The Top Cell of the second Column contains a formula which will retrieve the answer you want to watch, in this case Profit. It will be displayed as M$.
Now select the entire Blue Area and Select Data Table
This is the Data Table input screen.
The tricky/confusing part here is that in our example we are changing the input value to our Gold Mine Profit model using a Column of Numbers, so enter $C$6 in the Column Input Cell, Leave the Row Input Cell blank.
Click Ok
You can now see a Table of Profit Values for each Grade Value.
The variance in the Profit can easily be graphed against the Gold Grade and we can now see that if the Gold Grade is below about 1.55 g/t Au we will not make a profit and conversely if it is above 2.0 g/t Au we will make a large profit.
Before we move onto 2 Way Data Tables it is worth exploring small variations on One Way Tables.
What if my Data is in Rows?
Had our input data been arranged horizontally in Rows, we could have used a Row Input Cell to process the data.
What if I want to vary the inputs by a certain Percentage ?
Another Scenario is often where you want to vary an input by a Fixed Percentage.
This is easily done using Data Tables
Setup the input cells with the percentage variations you want to examine, noting that the values don’t have to be evenly spread.
Setup a Temporary Input Cell, This will hold the Percentage Variance briefly whilst calculations are happening. Set a default value of 0 (zero)
Change your Main Input Cell, Gold Grade in our case, to Multiply the fixed answer by 1+ the temp Input Cell.
Run the Data Table with a Column Input Cell, which will refer to the Temp Input Cell.
2 WAY DATA TABLES
So the Boss comes in and asks, what Happens if the Gold Grade changes as well as the A$/U$ Exchange Rate?
You guessed it, Two Way tables to the rescue.
This is shown in the attached Excel Workbook on the “2 Way” Tab or 2 Way Example
Two way data Tables work the same as One Way Data Tables except that you can vary 2 parameters at once.
With Two Way Data Tables you need to setup a Column of data for one Input and a Row of data for the second Input. The answer is returned at the intersection of the Row and Column.
Here we have setup a Column of Gold Grades ranging from 1.5 to 2.1 g/t Au and a Row of Exchange rates =varying from 0.70 to 1.00 A$/U$
Note at the intersection of the Row and Column there is a Reference to the variable you want to monitor in this case profit.
You can now see the variance in Profit for variations in Gold Grade and Exchange Rate.
What about varying by Percentages?
Once again we can rearrange the input variables to examine percentage changes in the inputs via a Temporary Input Cell.
MONITORING MULTIPLE VARIABLES
So you have a complex model and want to monitor a number of input and output variables at once. No problems, Data Tables to the rescue.
In this example we are varying one input variable but monitoring 3 Output variables, 2 input variables and then doing a calculation all as part of the Data Table.
This is shown in the attached Excel Workbook on the “Monitor Multi variables” Tab or Monitor Multi Variables Example
The first 3 columns, Total Cost, Revenue and Profit are output variables even though Total Cost doesn’t change, we can still monitor it to make sure our model is working correctly
The next 2 columns, Gold Grade and Gold Price are input variables even though only Gold Grade is being varied.
The last column Cost per Oz is not calculated as part of the model (ok sometimes we forget don’t we), but it can be calculated on the fly as part of the Data Table.
The result is:
MULTIWAY DATA TABLES
But I hear you thinking, “If Data Tables are so good why can I only Change 2 variables at Once? I want to change more! “.
No Problems
Data Tables in fact allow you to Change any Number of input variables at once and monitor any number of input and output variables. It does however require a slight of hand.
This is shown in the attached Excel Workbook on the “Multi Way Tables” Tab or Multiway Table Example
First things first,
Setup a table of what scenarios you want to examine:
Setup the Data Table area to monitor Inputs, Outputs and Calculated Fields
Note that the Input Data Column will be used to select the Scenario No.
Also note that we have setup F2 to retrieve the Scenarios Name.
And in H6 we will put the Scenario name into the Data Table, who said Data Tables were only for Numbers!
Next Link the Model to the scenario
And run the Data Table
Note how the Description Column is populated with the Scenario’s Name (Text values)
So now when your boss asks you what effect the price of … has on the budget, you know where to turn.
MONTE CARLO SIMULATIONS IN EXCEL
Monte Carlo simulation (or analysis) as its name suggests puts an element of gambling into the scenarios, or more correctly allows you to measure the effect of variability on input parameters.
This is done by running scenarios against your model hundreds or thousands of times and changing the inputs each time and then measuring the effects at the end of the runs.
And Data Tables can do that? Absolutely!
First some statistics.
Everything in life has variability, from the size of Zebra’s Strips, The height of people and the Arrival times of trains, the time that people read this post, the time that it takes people to read this post.
Most things are variable around a central or mean (average) value. The spread of variability is commonly known as the distribution.
Distributions can have many names and shapes, but common ones are
 Normal: Bell shaped around a mean
 Uniform: All values have an even chance of selection
 Exponential: Low or High values have a much higher probability that the other values
In life most distributions are Normal in nature indicating that the distribution is Bell shaped around a mean with a known method of describing the variability around this.
Excel has 2 functions that produce Random numbers, Rand() and Randbetween(). These 2 functions both have a Uniform Distribution, that is any value between the minimum and maximum values will have the same probability of being chosen.
We can convert a uniform distribution to a Normal distribution by some simple maths (simple to do, not simple to explain).
=norminv(rand(),mean,standard_dev)
Example =NORMINV(rand(),100,10)
Will generate a distribution of random numbers centred on 100 with a spread having a bell shaped curve with a standard deviation of 10. This means that the function will produce a number with a 99.7% probability of being between 70 and 130 and on average will have a mean of 100.
Monte Carlo simulations
So how can I use this and Data Tables to do Monte Carlo simulations.
Before we go any further the author wants to explicitly state that he is not suggesting that the use of Normal Distributions for the variables modeled below is appropriate, except for the purpose of demonstration of the principles behind Monte Carlo Modelling.
As with all models you need to have a good understanding of the distribution of inputs before you start playing with simulations or of which Monte Carlo is but one type. ie: Rubbish In Rubbish Out.
We can model an input vaiable, in this case Exchange rate with a distribution instead of a fixed value and then run the model a number of times and see what impact the variation has on the output.
This is shown in the attached Excel Workbook on the “Monte Carlo (Simple)” Tab or Monte Carlo (Simple) Example
The formula =NORMINV(RAND(),0.92,0.02), will generate a Random Exchange Rate with a distribution based on a mean on 0.92 A$/U$ and a spread of approximately 6 cents each way ie: there will be a 99.7% probability of the exchange rate being between 0.86 and 0.98 A$/U$.
Copying the formula down from H6 to H1005 will allow our data table to generate 1000 iterations of the model each with a randomly generated Exchange Rate.
In the model above, you can see that for a Base case exchange rate of 0.92 the profit is $M 5.452, however after running 1000 simulations the profit is actually $M 5.7134. More important is that you can now run statistics on the model to tell what is the probability of the profit being greater than 0.00 based on variance in the exchange rate etc.
Note 1: You will note that in the above data table that the Input Column (darker blue) has the formula for calculating a random input grade from a distribution. =NORMINV(RAND(),0.92,0.02)
This is a Volatile Formula , ie: It recalculates every time the worksheet changes.
What this means for the worksheet is that when the Data Table goes to Calculate Row 2 of the Data Table it will recalculate the Input value for Row 1.
On Calculation of Row 2, It doesn’t change the Table Values for Row 1, just the Input Column value.
So after 1,000 calculations of the Data Table, the Input Column values will have no relationship to the data from the original Calculations stored in the Data Table body area.
To make up for this we also add an Input variable to the Data Table.
Doing this allows the Data Table to capture and store both the Input variable and corresponding Output variable in the Data Table’s Body.
Note 2: Always run at least 1000 iterations of Monte Carlo models. This is to ensure that you have a statistical chance of getting sufficient outliers (extreme values) to make the variance analysis meaningful. This is important because as the number of iterations increases the variance of the average output decreases.
Press F9 a few times and watch the average H6:H1005 change.
Try changing the Data table from 1,000 rows to 10, 20 or 100,0000 rows. As the number of iterations increases the variance in the average of the output decreases.
Advanced Monte Carlo Simulations
We can now put our knowledge of Data Tables and Monte Carlo Simulation to the test by varying 4 input variables at the same time.
This is shown in the attached Excel Workbook on the “Monte Carlo (Advanced)” Tab or Monte Carlo (Adv) Example
In the example below we have inserted distributions for 4 input variables.
Ore Tonnes  Mean 1,000,000 tonnes  Standard Deviation of 100,000 tonnes 
Gold Grade  Mean 1.68 g/t Au  Standard Deviation of 0.1 g/t Au 
Gold Price  Mean 1,200 U$/Oz  Standard Deviation of 100 U$/Oz 
Exchange rate  Mean 0.92 A$/U$  Standard Deviation of 0.02 A$/U$ 
And setup a data Table for the 4 Input Variables and main output variable, Profit.
Note: When this model is run through the Data Table, note that the Row or Column input cells can be set to anywhere. The Model is not using the value of the Input Cell (Row or Column) and isn’t even using the Run No (Column F) for the model, the data table is simply being used to run lots of iterations of the model, with the variability coming from the Random Numbers in the 4 input cells.
ITERATED FUNCTIONS INCLUDING FRACTALS
At a meeting in early 2005, the company I was working for was looking at an integrated Scheduling & Budgeting system.
The salesman gave a great demo except that the system would take approx. 30 mins to calculate our budget as opposed to a half a second in Excel.
Complaining I mentioned that our current, Excel based, system could do the job in seconds.
And he returned stating that “the system was doing a lot of things Excel couldn’t do”.
I responded “but Excel can do anything”
and he immediately shot back that “Excel can’t do a Mandelbrot”
To which I responded “Yes it can”
And he responded “Not without VB Code”
Without too much thinking I responded that I would accept the Challenge.
The attached file, which is described below is my response.
Excel Mandelbrot
The attached file is an implementation of the classic Mandelbrot implemented in Excel without the use of VBA code.
A Mandelbrot is a graphical display of the simple equation Z_{n+1} = Z_{n}^{2} + c, where Z is a complex number (x +iy).
Which is described at http://en.wikipedia.org/wiki/Mandelbrot_set
This can be solved in the real XY domain using:
Xnew = Xold^{2} – Yold^{2} + X_Orig and
Ynew = 2 * Yold * Xold+ Y_Orig
Study of iterated functions reveals that these functions will either converge on an answer or diverge once a boundary has been breached
In the case of the Mandelbrot, this function diverges after the function Z^{2 } > 4
So to construct a Mandelbrot a program needs simply to
 Loop from Xmin to Xmax in small steps and
 Loop from Ymin to Ymax in small steps and
 For every X, Y Point in the above 2 loops, solve the above equations until the answer is > 4
 Color the screen according to how many iterations it took to diverge or not
Simple…
Except that Excel doesn’t have any looping functions unless you use VBA Code
The calculation of the solution for any X, Y starting point is simple enough using a series of Rows and Columns where each Cells is the starting iteration of the solution for each various X, Y coordinate.
This is shown in the Calculations page in the Xnew, Ynew, Xold, Yold, Rsq and Count columns.
The iterations are simply done in the Xnew and Ynew columns
For each iteration we check that the Z^{2} value hasn’t diverged (not > 4) (Xnew^{2} + Ynew^{2})
And keep track of how many iterations it took to diverge, the Count Column
The above 5 lines I refer to below as the Calculator.
The trick to working out how to do this for a XY Grid was the use of the Table Function to send the starting positions to the Calculator and return the Count for that location.
This is the large Yellow Area.
The Large yellow area (Data Table Area) is flanked on the Top and Left by the X and Y coordinates for a grid encompassing the area which we want to plot.
The Table Function extracts the Top and Left values and puts them in the X Orig and Y Orig positions of the calculator.
The Calculator returns the Count of the Divergence of the Calculator to the H2 position (Top Left corner of the Grid) and that value is stored at the Grid Location.
The Data Table repeats this for each position in the XY Grid.
An Excel Surface Chart can then Chart the Large Yellow area in effect creating a Traditional Mandelbrot plot by joining up adjacent areas of equal value (Contouring).
The Chart can also be displayed as a 3DSurface rather than a Contour Chart for a dramatic effect.
Zooming In can be added by adding code that allows the user to say Right click in the Large Yellow area and the code will then take the Coordinates and Zoom in by a fixed factor
Zooming Out can be added by adding code that allows the user to say Double click in the Large Yellow area and the code will then take the Coordinates and Zoom out by a fixed factor
DOWNLOAD EXAMPLE WORKBOOKS
Download the complete example workbooks described above and practice data tables on your own.
 Click here to download Gold Mine Monte Carlo Simulations & Data Tables workbook. [XL 2003 version here]
 Click here to download Excel Mandelbrot workbook.
Note: A few people have said the above files either Hang or Freeze there PC’s. This is probably because they have a number of large Data Tables within them.
I have uploaded each Tab as a separate Excel 2007 file, see below:
1. 1 Way.xlsx
2. 2 Way.xlsx
3. Monitor Multi Variables.xlsx
4. Multiway Table.xlsx
5. Monte Carlo Simple (updated)
6. Monte Carlo (Adv).xlsx
In the Example Files some of the Data Tables have been removed and there are instructions on how to reinstate them included in the file.
FINAL THOUGHTS
Speed
If you start adding a number of Data Tables to Complex Models you will rapidly cause even the fastest machines to grind to a halt.
VBA
The best way around the above speed issue is to setup a number of Data Tables for whatever analysis you wish to undertake. Then as you run each analysis copy the Data Table Data Area, The area between the Rows and Columns and paste it as values over itself. Then move onto the next data table and run it.
This allows the Data Tables to be quickly recalculated if required.
This process can be automated via 3 lines of VBA code for each Data Table.
‘Calculate Data Table in F5:H18, using Column Input cell C9
Range(“F5:H18″).Table ColumnInput:=Range(“C9″)
‘Copy Data Area as Values
Range(“G6:H18″).Copy
Range(“G6:H18″).PasteSpecial Paste:=xlPasteValues
‘Repeat Above for each Data Table
‘Deselect Current Range
Application.CutCopyMode = False
Cell Contents
If you look at a cell in a Data Table you will see something like:
 {=TABLE(,E5)}: for a Column Input Cell
 {=TABLE(E4,)}: for a Row Input Cell
 {=TABLE(E4,E5)}: for a Row and Column Input Cell
Although these appear like Array Formula, they cannot be manually set.
So setting up a data table and typing =TABLE(,E5) CtrlShiftEnter, only produces an error message.
Further Reading & References
 http://www.exceluser.com/explore/statsnormal.htm
 http://www.vertex42.com/ExcelArticles/mc/GeneratingRandomInputs.html
 http://www.itl.nist.gov/div898/handbook/eda/section3/eda366.htm
 http://en.wikipedia.org/wiki/Mandelbrot_set
 http://chandoo.org/wp/2011/06/20/analysedatalikeasuperhero/
Added by Chandoo
This post is by far one of the most comprehensive posts on Chandoo.org. And each of the 3100+ words in it show the passion and knowledge that Hui has. Thank you so much Hui for sharing this wealth of knowledge with our members.I have learned a lot of interesting and useful things from this article.
If you have enjoyed this article, please say thanks to Hui.
 
 

Leave a Reply
Change Data Labels in Charts to Whatever you want [Quick Tip]  Important Update: Dilbert is gone! 
171 Responses to “Data Tables & Monte Carlo Simulations in Excel – A Comprehensive Guide”
Hi
Looks like a grate post, I’ll need to give some time over to reading it fully, thanks Hui,
Ross
Hats off to Hui,
Can’t say i’ve ever used that in Excel before, but thanks for the introduction
Hui,
definitively, another post to print for the travel back home…Without all your knowledge, I am a convinced that Excel can do everything, but in my case it’s more a question of faith, rather than knowledge…
Thanks for sharing, and thanks Chandoo for giving him the space !
Rgds,
Martin
Excellent post describing a powerful yet unknown/underused feature of Excel.
1st of all thanks to Hui!
And great bost, maybe the most usefull i’ve seen here so far. I already knew of excels capabilities in this area but often couldn’t find it well described before…
Thanks for the crytal clear explanation!
Very thorough use of the tables functions! I wonder though, wouldn’t creating a Pivot table after the fact make rearranging these various scenarios a lot easier?
Also, have you tried using the Analysis ToolPack for the Monte Carlo simulations? I’m curious if it has any similar functionality (it’s disabled for some bizarre reason at my place of work).
Nice post, funny – there was a very similar discussion about monte carlo and mandelbrot a few days back, don’t know if that’s just coincidence?
See: http://www.excelhero.com/blog/2010/03/montecarlopi.html
The linked workbook is in hires so takes a little while, but is worth the wait – maybe someone can help speed it up? (eg using IF(ISNA(…) in the iterations runs about 1/3 quicker)
As mentioned there, to improve performance of Monte Carlo, you need to find ways to reduce the variation. In fact, since convergence is order root n, every additional digit of precision requires 100 times more computations! Lori
@ All, Thanx for the words of appreciation.
@ LHM, although the Monte Carlo system is in use in both examples, Excel Hero is using the weight of numbers to zero in on an answer and yes the more iterations you do the more accurate it is, as it is a definaite formula that gets you closer to the answer the more you feed into it.
In Monte Carlo analysis of variability you never get the right answer, regardless of how many iterations you do. What you do need to do is do enough iterations that the results are statistically accurate. Generally for a Normal Distribution you need a minimum of 30 points to define a statistical population which will be representative of the real population.
Because you are generally interested on the probability of the results of the simulation being above/below a certain range, you are more interested in the outliers, than the mean. This means that you need enough outliers to be presented to be a representative population of outliers. For a Normal distribution where +/ 3 Standard Deviations is 99.7% of the population if you then want 30 samples in the remainder 0.3% you need 30 x 100/0.3 or about 10,000 iterations to guarantee a sufficient sample of outliers. This is starting to become a large number and most people run in the 110,000 iteration range and then check that sufficient outliers have been presented.
Hui & Chandoo,
This was a very instructive post learned a great deal from it.
I split out the High and Low in the scenario description and creted a high and low field to hold a percentage and then linked the scenario data to these fileds so that I could change what High and Low meant. If you wanted to test for +10% for high you would put 10% in the High field, or if you wanted to test for a low of 8% you could put 8% in the low field.
The possibilities are endless.
Thanks once again for the post
Hui et. al.,
I dont’ really see the advantage of doing this in data tables and not running the iterations in vba? Certainly it isn’t any faster.
Seriously speaking MC Valuation need a few thousand iterations, i don’t think you’ll like to do that within the sheet. I’d prefer looping it in VBA for smaller projects, for larger ones you need to switch to oracles crystal ball or sth similar.
Never the less, i really liked the application of advanced excel methods.
Thy
@Chris
I ran the Multi Way Tables example with 4 inputs and 5 output variables
with 1,000,000 iterations in the Data Table in about 8 seconds
Using VBA to do 100,000 iterations took about 230 seconds
I Deleted all the other sheets to speed it up in both cases
Excel easily handles 100,000+ iterations on modestly complex models with ease
I try and use the native functions of Excel as opposed to VBA as much as possible, they are just so much faster.
The low iteration count in the example models were for demonstration purposes only.
just one word for this “AWESOME”
Hui, great post, you and Chandoo really qualify as an =”Excel ” & POWER(Ninja,99) – someday I will get there.
I have to say that I agree with your “Excel can do almost anything” statement, I prove it everyday in my workplace.
In all my years of working with Excel I´ve learned an important lesson, the very first and ultimate principle that determines the approach to take with an Excel problem: Whatever you can do with the built in Excel functions would run faster and better than if you do the same thing with macros. The problem is that the common mortal doesn´t know what tools he can use, or how to use them.
I have to admit that I dont use the Data Table feature but I like´d your way to do the fractal iterations!
There is another way to make iterations in Excel… you can use one of the most powerful but concealed feature (and often considered only as an error warning): circular references. Of course, you have to make more complex formulas, you have to be careful with the physical placement of your variables and calculations, but you can write really complex models with it!
Great work, keep the good posts coming!!!!
P.D.: Sorry about my english, It´s not my native language!
[...] He elegantly guides you through the wonderful process of whatif. [...]
Stunning!
Hui, great article with a lot of insight. I will need more time to get everything out of it, but I have learned a great deal already. Thanks.
All of the inputs in my workbook are on a different sheet. When I try to create a data table I get the error message ” Input Cell Reference Is Not Valid “. Is there a workaround?
Kwesi
Setup an input area on the same sheet as where the data Tables will go
Link the real inputs to these values
Run the data tables on these values
It is a shame that you can’t use named ranges here to get around this issue, like you can elsewhere where intersheet formulas can’t be used.
Hi..
Montecarlo simulation using excel is amazing. I really appreciate you for sharing the knowledge.
I have this question for you.
Let us say I am trying to figure out the variation in profit for variable A. Using the method suggested I can run the montecarlo simulation. I want to draw a graph at each iteration (X axis Different units of variable A), (Y axis – Different profit levels which corresponds to different variable units.)
So once the iterations are over a chart will be drawn where this could be used as an efficient frontier.
Please let me know whether it will be possible. If this can be done we all have to agree that Excel can do wonders.
Many thanks in advance for your support
Dilishan
You have 2 easy options here:
1. Scatter Chart – As you have a table of Input Values and results next to it in the actual Data Table, it is a few clicks to chart the data as a scatter chart. You should see more symbols near the mean value and less as you get towards the outliers.
2. Cummulative Column Chart – The most common method of Charting Distributions is a Column Chart and associated Cummulative Line Chart (like http://i26.tinypic.com/34yo20k.png ) .
This will require you to setup an area where you have a list of result containers, ie: If your results go from 0 to 100, you may want to setup containers every 5, 04.99, 59.99,1014.99 etc and then use a sumifs or sumproduct to count how many times these results are in the Solution part of the Data Table.
i wanted to knw the procedure of sum the numeric value in one sentance
Ankit
You can do something like:
=Sum(1,2,4,10) which will give you the answer of 17
or
=Sum(A1, A3, A5, A10) which will add up the values in those cells
or
=Sum(A1:A5) will add up all the values in the Range A1:A4
or a combination of the above
=Sum(2,4,A5, A7) which will add 6 (2 + 4 ) to the values in A5 + A7
Fantastic article, just helped me save hours of work!
Great article. Microsoft should be paying royalties!
Thanks.
[...] http://chandoo.org/wp/2010/05/06/datatablesmontecarlosimulationsinexcelacomprehensiveguide/ [...]
Excellent post. You should check out http://tukhi.com. One of the examples in the Mosteller workbook plays the game of craps!
this is pretty cool.
anyone know how to make a data table of a data table?
i.e. i want to run a model for N period where each period is dependent upon the previous period while varying multiple input variables.
then i want an outer loop running the above model for M iterations to find the optimal settings for the input variables based on summary statistics found for each iteration
when i try a data table referencing another data table the referenced data table does not change, so i only get iterations for 1 set of input variables
One of the best articles I have ever read on Excel.
Thank you very much Hui.
Hello, please can I know how you created the Large yellow area (Data Table Area)?
Thanks
Rut
Is there any way to unsubscribe from comment posts? I think my last comment was a year ago….
@Bill.. you can unsubscribe from the email alert you get. There should be a link to managing your profile in that mail.
@Rutgerius
The large yellow area is the Data Table area
You only need to populate the Top Row and Left Column
Then Use the Data Table command to fill it.
Select Cells D2:DE103 and goto Data Table
Select cells for the Row and Column Inputs as described in the text above
hi,i need some help with the data table.Could please explain how you did the yellow data table for the Mandelbrot set in much details as possible and stepwise..thanks
Hi
I tried it.. but it aint working
@Schizophrenia, Rutgerius
.
For a start thanx for taking the interest in this post.
.
1. Download the Mandelbrot Spreadsheet from: http://cidb663e096d6c08c74.skydrive.live.com/self.aspx/Public/excelmandelbrot.xls
Don’t click on the icon and open it in the web browser.
It is an .xls file and will work in all versions of Excel.
.
2. Goto the Calculations Page and select I3
3. Press Ctrl Shift Right, Ctrl Shift Down, Delete
This will select and delete the range I3:DE103
4. Go back to H2
5. Press Ctrl Shift Right, Ctrl Shift Down,
This will select the range H2:DE103
6. With your mouse got the Data Tab, What If Button and select Data Table
7. In the Data Table dialog enter
Row input cell: E5
Column input cell: E6
.
Voila
thank you much..it works!:)..it was of a great help for my project at university..
@hui
any idea for my earlier posted problem?
thanks.
@James
Do you have an example you can email or post for me to see what your trying to do?
@hui
couldn’t find your email, trying google docs
https://docs.google.com/leaf?id=0B00vs4RO2HKmN2I0OTRkNGEtMWYxNy00ODIzLWJhMjEtYmMyNTUwZDJjYjZi&hl=en&authkey=CIXlFQ
the top middle portion show 1 iteration with 4 periods, the right data table is 1000 iterations of it
then, as you can see the bottom 2way data table is referencing the data table on the right, but the different prices and supplies are not being used.
supposing that works, next i want to be able to change the prices from period to period for each iteration, (i.e. 60455570) [all possible permutations of priceperiod combinations]
Thanks!
Why not just use the first Table and run for 20 50,000 iterations
Change
F3: =CHOOSE(RANDBETWEEN(1,6),45,50,55,60,65,70)
C20: =CHOOSE(RANDBETWEEN(1,6),1500,2000,2500,3000,3500,4000)
Add 2 columns to the table
to show Price and Fleet size
and feed results into a pivot tabl
as per: http://rapidshare.com/files/432543642/data_table_referencing_another_data_table_Hui_ver.xls
thanks! that’s an interesting approach, never used pivot tables before,
if i’m understanding correctly from the file, this means the number of 45′s vs 50′s vs 55′s prices may differ, so I can’t compare across supplies for averages computed upon the same sample sizes
same goes for the number of iterations that use 1500, or 2000′s — they will differ.
anyway to have a fixed amount of all of them? ( i guess you could have 6 * 6 * 1000 iterations…)
That’s why you do a large number of iterations
You can also do a count in the pivot table to see how many iterations fall into each category
say i need exactly 1000 45′s and 1000 50′s and 1000 55′s though?
is the way to private msg on this board without revealing email?
My email is at the bottom of
http://chandoo.org/wp/abouthui/
Great post, I would just comment that the formula you are using to calculate percentage change in the one way table should be =1.68*(1+E2)
Hi Hui
I have a question about using a data table to do scenario analysis on a model I have built. The model has an INPUT tab/worksheet where I have placed my data table. However all the inputs pull onto a second tab (Profit and Loss) to calculate the output value which I want to use in my scenario data table. The data table is not producing the correct values. So my question is: for a data table to work, do all inputs AND calculations need to be on the same tab/worksheet?
@Dionne
Yes, All inputs/ouputs have to be on the same sheet as the Data Table
But this can easily be done with an input/output area, which will be linked to other parts of your model as appropriate
Calculations can go across any number of sheets
Hi Chandoo,
Congratulations for the great work!
I have a couple of questions regarding excel.
1) How can I generate auto correlated numbers in excel – Suppose I have fixed the correlation to be 0.7 ( Corr(X,Y) =0.7) then how can I generate random X,Y for it? Is there any way to fix Intra class correlation and generate X,Y for it?
2) How a Monte Carlo simulation can be done to test Krippendorff’s Alpha ( statistical parameter) against Intra class Correlation?
It would be really thankful for your reply!
Thanks!
[...] Related Tip: Learn how to work with Data Tables & Montecarlo Simulations in Excel [...]
Thank you sooooo much Hui!
A life changing post for me!
Happy Valentines too
Thank Hui, your contribution is excellent, and use to calculate the value of the premiums for helth insurance.
@ Natalia & Carlos
Thanx for your kind words
Hui…
Hui,
Thank you so much for this valuable information! I have a question regarding the Monte Carlo Simulation (Simple.) You state, “The formula =NORMINV(RAND(),0.92,0.02), will generate a Random Exchange Rate with a distribution based on a mean on 0.92 A$/U$ and a spread of approximately 6 cents each way ie: there will be a 99.7% probability of the exchange rate being between 0.86 and 0.98 A$/U$.” My question is: how did you determine the 0.02? I realize that 3 standard deviations will give us the range between 0.86 and 0.98, but how did you determine the 0.02 in the first place? Why not 0.01 or 0.03 or some other number?
Thanks!
@Kathleen
The Norminv function accepts 3 Parameters
=NORMINV(probability,mean,standard_dev)
.
The 0.92 and 0.02 are the Mean and Standard Deviation that the distribution I used had.
.
In this case the 2 numbers are made up for Demonstration ourposes only.
Had I used a SD of 0.02 or 0.2 or 0.3 the Ranges Distribution would change, but it will still have a 99.7% probability of being within the Mean +/ 3 SD’s.
.
In reality you will take a number of measurements of your data and then work out what the actual distribution (Mean and Standard Deviation) of the data is.
[...] One area where Random numbers is used is in Monte Carlo Simulation. This has been discussed at Chandoo.org at Data Tables and MonteCarlo Simulations in Excel a Comprehensive Guide [...]
Hi Hui
Very good work. I am just not too sure one one particular point. Why is it that the Exchange rate you generated using “=NORMINV(probability,mean,standard_dev)” be different to what you have in the “Ex Rate” column? Shouldn’t they be the same?
Thanks
Sam
Hi Chandoo,
I downloaded both xls and xlsx versions of the file “goldminemontecarloanalysis” and attempted to open them twice. On both occasions, MS Excel simply hung up and I had restart it. Is there something wrong with my machine? I think it’s unlikely there can be something wrong with your two files…pls advise. Cheers!
@Nitin
Both files are still Ok
.
Click on the link and then select either
1. The Download Link which will open the Download as Dialog
or
2. Click on the large Spreadsheet Icon in the middle of the screen which will open the files in the browser, then File, Save a copy
good information is presented. Thanks
Chandoo, I need a help from you. Could you send the procedure of how to copy the values automatically into other cells using a macros? After completing the data table functions, I need to copy the STDEV of the generated numbers. I want to select the min STDEV. Kindly help me out of this.
Hi Hui,
Thanks for responding. As mentioned earlier, I have complete faith in Chandoo.org. However I regret to state that in spite of following the process you mentioned, I continue to face the same problem – both files simply take too long to load and in the meantime, MS Excel hangs – this in spite of a machine with C2D processor and 2GB RAM. Apologies for bothering you time and again, but could you suggest some other course of action? Thanks!
[...] few people have told me that the example files in Data Tables & Monte Carlo Simulations in Excel – A Comprehensive Guide either Hangs or Freezes there [...]
All good ,Top stuff
Keep the good work
Wonderful work Hui & thanks for sharing :o)
I find the Madlebrot set fascinating, although my PC didn’t like the chart.
To get around this I applied xl 07 Graded Colour Scale Conditional formatting to the, once yellow, area. Zooming out this gives a very similar effect & less resource heavy.
Off to see if I can apply different distributions to the Monte Carlo exercise.
+ 1e6 Internets
=kick(“ass”,”everybody’s”)
anyone know why i am getting the same output for every cell in the table?
@Kim
You have probably put the Data Table link in the Row Input Cell instead of the Column Input Cell or vise versa
[...] http://chandoo.org/wp/2010/05/06/datatablesmontecarlosimulationsinexcelacomprehensiveguide/ [...]
[...] Hui and Chandoo have been speaking a lot about the data tables. They are your best friends when it comes to doing the donkey work about changing the variables and noting the scenario results. One post that I love is here. [...]
Hui,
first of all, thanks a lot for this post. it is great and what I have been looking for to understand how data tables work.
I have a question about 1 way (or 2 way) data tables using percentage variation of inputs. you recommend using formula for input cell (e.g. gold grade) as =INPUT*1+Temp.Input.Cell. So, using this logic, 50% as in your spreadsheet, gives 15.5 M$ profit which corresponds to 1.18 (i.e. 1.680.5). However, for some people, to me at least, 50% more logically means half of the original value, i.e. 1.6850%=0.84. In this case the formula for the input cell should be =INPUT*(1+Temp.Input.Cell).
thanks again for the great help
@Elkhan
I’m glad that you liked the post.
I nearly use Data tables as much as Sumproduct, it is such a powerful function.
.
Great pickup on the formula
You are totally correct that the formula should be
=INPUT*(1+Temp.Input.Cell)
.
I will severely scold the proof reader…
Hui
thank you very much for this wonderful post.
Regards
Anup
Excellent article Hui – thanks for unearthing such a beautiful and potent gem from the Excel vault!
[...] myself an expert by any measure. There were areas where I was very competent (Sumproduct & Data Tables) and I do consider myself very good at systems analysis, but I was no way an excel [...]
[...] more details on data table in Excel, see http://chandoo.org/wp/2010/05/06/datatablesmontecarlosimulationsinexcelacomprehensiveguide/ Filed under Teaching You can leave a comment, or trackback from your own [...]
In the first 1Way Table (http://chandoo.org/wp/2010/05/06/datatablesmontecarlosimulationsinexcelacomprehensiveguide/#1waytables) I tried using NormInv to vary the Gold Price using formula =NORMINV(RAND(),1.68,0.1). This produces profit results that are completely inconsistent with the input values.
The problem is easy to see if the input value is rounded to produce a few discrete values: =ROUND(NORMINV(RAND(),1.68,0.1),1). Here is a partial sample output from the table using rounded inputs:
5.45
1.70 2.10
1.80 10.48
1.80 6.29
1.70 2.10
1.80 6.29
1.80 10.48
…
How is it that the same input value can produce different outputs? I’m guessing that somehow the call to volatile Rand() causes the Table to evaluate a different value than what is shown. If you look at the output values, they are discrete as well (as expected), but it seems they are mismatched with the inputs.
As a test, I tried a different randomizing function for Gold Price =INT(RAND()*3)+168. The same problem occurs: the same input can yield different output in different trials.
I guess the moral of the story is not to use Rand() in the input variable formula. Has anyone else noticed this behavior?
@Andy
Thank you for pointing this out
.
There are two things that could be going on here.
.
1. Excel Version
In Excel versions up to XP Norminv returns errors from time to time
Refer: http://support.microsoft.com/kb/827358
.
2. Rand()
Your suggestion as to the Volatile Rand is correct
The Data Table recalculates after every iteration of each cell in the Input Row/Column.
So when the second row calculates, it forces the input cell in the first row to recalculate.
So in the example the Input Column is recalculated 9 times after the first iteration scrambling the input grades.
The Profit figures in that example are still correct as they haven’t been recalculated, its just the input column that is incorrect.
.
Workaround
Luckily there is an easy work around for this:
When setting up the data table at another column that will link to the Input Cell (C10)
in this example
Gold Grade Profit Grade
5.5 1.68
1.70 $6.3 1.70
1.80 $2.1 1.60
1.70 $2.1 1.60
1.70 $10.5 1.80
1.50 $6.3 1.70
1.70 $2.1 1.50
1.60 $2.1 1.60
1.80 $6.3 1.70
1.60 $10.5 1.80
1.60 $2.1 1.50
.
You can see above that the Profit is now consistent with the Grade Column, but not the input grade column.
@Hui
Great explanation, and the workaround is so simple I couldn’t see it. BTW, this is a great article! I’ve never seen the Excel Data Table function explained so well anywhere else.
@Andy
Thank you so much for the information
In the first 1Way Table (http://chandoo.org/wp/2010/05/06/datatablesmontecarlosimulationsinexcelacomprehensiveguide/#1waytables),
I am trying out the second example of varying the input by percentage. I couldn’t quite follow and it doesn’t produce the result as you have shown.
I have set up the temp cell with a value zero.
Changed the Main Input Cell multiplied by 1+ the temp Input Cell.
created the data table by giving column input as $c$6. I am not seeing the result as you have shown above. Can you please explain me further? you can email me separately if this is too basic for others in this thread. Thanks.
Sorry, got it. Don’t know how, but tried again and now it works. Thanks. Now I can move on to the rest of this article. Thanks.
@Premalatha
Glad you fixed the issue
Questions are never “Too Basic” for others
I assume that only 1 in 10 people actually ask if they have a problem and so all questions are valid
I hope you enjoy the rest of the post
[...] Introduction to Data Tables & Monte Carlo Simulations [...]
Hi Hui,
thanks for your reply. Very pleased to see you replying back. I have started liking data tables after reading this post. I haven’t completed reading it, but I have already started using it. I am very pleased to see the results.
Thanks. Will come back again and again.
Cheers
Premalatha
Hui,
Thanks for an excellent tutorial with great examples. I have been using data tables a lot lately to produce an output page that shows the results from the multiyear model when certain inputs change. Ordinarily, we run the model in real time, demonstrating what happens when we change input #1, etc. Occasionally, we must rely on printouts, and we like a summary sheet of the six scenarios we usually run. Since this is a multiyear model, I generally use a onevariable table and include the profit/loss value for each year as “monitoring multiple variables” as you describe. However, I have one scenario where I change two values at once. Since this only happens once, and since I have only one alternate value for each variable, I would love to have a twoway table with monitoring multiple variables. I don’t want to use the scenario method you described, because I want to be able to change assumptions in real time — i.e., I don’t want to replace those cells with VLOOKUP formulae. Since my final scenario is a combination of two earlier scenarios, I tried what I would call a nested data table — i.e., the output cells are not my profit/loss row, but the results of an earlier data table. In other words, data table #1 changes revenue growth by 5%. I would like data table # 2 to point to that output and change salary expense to 5% and have the results of data table #2 be a combination of the two.
Thoughts?
Thomas
@Thomas
Unfortunately Data Tables cannot be nested or dependent.
Although they look like Array Formula, they do not recalculate when a subsequent Data Table is dependent on the output of a previous table.
.
I have assisted other posters here in redesigning there models to get around this.
Happy to have a look if you want to email me.
Would you mind if I use your examples or part of your datasheets to teach Montecarlo Simulation using Excel? Yours have been the most comprehensive I’ve found and it’s not because I’m lazy or anything… just pressed for time.
Really well explained.
Thank you.
Thanks Hui, help me a lot ! Very fancy function!
Hui,
I’m struggling to get my data tables to work properly; I’ve recreated the structure that you provide in the post above (and the same offered by Office.com and other sites) and I’ve also downloaded your files and tried recreating the data table using the exact structure you used, but I keep running up against the same problem.
To take your 1Way file as my example. If I select the two columns necessary to create the table (G5:H16) and then select C6 as my column as the Column Input Cell, it populates the cells (H6:H16) with the same value as in H5. The cell contents show {=TABLE(,C6)} but it comes back with$5.5 in each and every cell.
Even if I change the Column Input to something else, like C5 – it stays with the same values.
Any chance I’ve tweaked some setting long ago that is interfering with the Data Table?
Otherwise I can’t make sense of why I can’t get it to work.
Followup – it only seems to be happening with some of my files.
They all carry out calcs on various other sheets and the I bring everything back to the same sheet (both inputs and outputs) for the data table, but for some reason it doesn’t work.
The only way to get it to work is if I close that entire excel window, start a new session with a blank file (or your example file) or a file that I know will work) and then do it that way.
Still not sure why its giving the error, but at least I know how to get around it.
FWIW, the scnerario manager works just fine as a standin.
@Bilsko
You may want to look at the Calculation settings
You have enabled Automatic Calculations except Data Tables!
So change it back to Automatic.
Hui,
Thanks for the suggestion – I had checked the options/settings to make sure everything was set properly.
It seems to have been a weird fluke (read: user error)
Got everything working properly now – both for single and double parameter Data Tables.
Thanks again for the walkthrough
I find this post as useful as I have ever found on Internet!
Thanks for this great job!!
This is a really great post. I must however point out a slight typo in the formulas used in the mandelbrot iteration :
the correct formula to calculate Ynew should be 2* Xold * Yold + Y_orig and not 2*Xold^2 * Yold^2 + y_orig (just remove the square)
I am truly amazed at the power of excel you unravel before us.
Thibaut
Great pickup after nearly 3 years of this post being live.
I have corrected it above.
Thanx
Hui…
by the way, the Formula used in the text in error (now corrected), still gives a rather nice Fractal pattern when used in the workbook.
Hui,
Didn’t someone say that “posts grow old but never die”.
A simple modification of the formula gives you another interesting outcome :
Y new = 2* abs(Xold * Yold) + Y_orig
You get the burning ship fractal. (for better visualisation, place the series in reverse order, and use a Yrange of 2 starting at 2.5
Hi all – I have added two multivariable tables to my file and although both work, they have slowed down my file signficantly. I added the VBA above, however, I keep getting a popup when I try to run it which highlights (“C9″),
Range(“F5:H18″)).Table ColumnInput:=Range(“C9″) and says “Invalid Outside Procedure.” Any idea why that might be occuring?
@Damien
Are you still looking for a solution ?
Hui…
If we were in jail together I would protect you in the showers….
Bestest, most helpful excel Tute I have come across!!!
~Kieran
This is truly outstanding, Hui. For some reason i’d missed this post until now.
THere is a great article at http://support.microsoft.com/kb/282851 about combining Data Tables with Advanced Filters and excel’s Database formulas (e.g. DSUM etc) . You can use datatables to ‘swap out’ the criteria in a DSUM or any other database function. Very very powerful indeed.
So you can set up amazing filtering run directly off the spreadsheet in a way that users can easily interact with, and do stuff that would otherwise require SQL. Blows pivots out of the water, on account that pivots only allow additive filtering. (Of course, you could always feed pivots with disconnected recordsets and generate the requisitie SQL on the fly, but that could get messy)
Again, hats off on this amazinf post
Thanx for the feed back Jeff
The web is a big place, it is difficult to checkout every corner
Is it possible to change Data label in this way?
If the cell data is 145 , then the chart bar will extend till 145. But I want to display above this bar is “100%”. (I mean the label should display “100%”).
[...] A comprehensive guide [...]
Chandoo,
Very impressive!
I knew data tables with two variables. You have made it a swiss knife by literallyusing only one input variable and getting what all one wants.
Thanks a lot.
Regards,
Vivek Sahasrabuddhey
Hui (Pronounced you) are a LEGEND!!!!!!!!!!!!!!!!
Hui, Great post on data tables the last section on Mandlebrot was way over my head as i am not mathematician and it the first time i heard that word..lol..but the rest of the article is realy useful for me.
Ques: is there a simple way to generate any other distribution funcs eg a skewed func.
Hui, this is a very good post, easy to understand…
Q: exept where you have under the Monte Carlo when you run the data table and select $E$5 and this has a value of 2, I don’t understand where this 2 comes from… I am trying to run a contingency calculation with risk value and contingency wih a % on both and I don’t know what to input in the cell that will be what is your $E$2
Thank you very much!
@Claudia
Firstly, I’m glad you liked the post and I hope it helps you in its use.
Cell E5 is just a place marker or holding cell
Nothing uses it or is linked to it.
WTF?
The 4 Green cells C5:C9 all have formulas that will calculate Random value, based on its own distribution, for each run of the Data Table
The Top row of the Data Table retrieves the values from these cells as wellas the solutions cells
The Left Row of the Data Table has a simple run number 1 to 1000
So when you select the Data table and put a value into either the Row or Column Link cell it doesn’t matter which you choose, lets assume you used a Row Link, what happens is that the DT will place the value of the Row’s value into cell E5 and then recalculate the Worksheet
That will produce new values for the Inputs C5:C9 as well as new results
The Data Table puts all these values into the table next to the Run No in that row and then moves to the next Row/Run No. and does it again
So after 1000 iterations you will have a table of 1000 unique runs including 1000 uniques Input combinations of the 4 inputs C5:C9 and associated solutions.
It is now up to you to analyse this and see what it all means
Hope that helps in some small way.
for some reason after following the formulas, mine doen’t work,
I have Risk Value =NORMINV(RAND(),4300000,8%) (4,300,000)
Contingency =NORMINV(RAND(),430000,5%) (430,000)
then i created a chart :
4,300,000 430,000 (=+C8 and =+C9 respectively)
1
2
3
…
then I selected the table and open the data table and entered under the column imput cell a linked cell wich is empty?
what am I doing wrong… I have never used excel for this kinds of calculations… any help will be greatly appreciated!
@Claudia
Can you email it to me?
Click on Hui above
email at bottom of page
Hui, I just got it… turns out that for my calculation I cannot use a % under =NORMINV(RAND(),4300000,8%) I have to multiply the % then give the value for the standard deviation… then I get resonable numbers…
thank you so much for ur help
Well done
I was going to suggest they were small’ish
Hi Hui,
I’m working on a sheet and using this Data Tables with 2 variables, I’ve kept the formulas setting to ‘Automatic’ and following every step, but still all the cells show the same figure.
Can you please give a solution?
@Avadhut
Can you email me the file:
Click on Hui… above, email at Bottom of page
Unbelievably helpful, thank you!!
Hi Hui,
I was attempting to recreate the 1 way data table where the value increments by a certain percentage. I notice that the formula in C6 is
1.68*1 + F6
when I think it should be
1.68 + F6 * 1.68
–> This way the value of the cold grade is the varying percentage of the initial value (1.68).
Thanks for your post. It has helped me to understand whatif analysis.
@Abi
In fact we are both wrong
It should be =1.68*(1+F6)
That way when the value in the Column Input is 25%
the value will be 75% of 1.68
When it is 25% it will be 125% of 1.68
Just a small thing: Is the A$/U$ backwards in the equation for revenue that you’re using? Wouldn’t dividing by these exchange rate – as you do in the spreadsheets – convert to U$ when what we want is dollars?
I meant when what we want is A$*
I just answered my own question:
http://www.tradingquarter.com/currencypairs/currencypairsaudusd/
Great site!
Hi, This is wonderful work. Just one concern:
in the Very by Percentage section, the formula in the grade cell should use parentheses ie (1+E2) or (1+F6)
@Anil
Correct
Please see Abi’s post 2 posts above this
Hui…
Daydreaming on a lazy friday evening at work, thought of building a mandelbrot in excel. Found this site, on googling! Thank God, I am not the only crazy guy.. and thank god, for internet, google and people like you! Stay awesome!
Hui,
Thanks for the post. Is there a YouTube video of this? :)
Thanks!
@Scott
no there isn’t a YouTube video
But it is on my list of things to do
Hui…
Thank you very much for the detailed and thoughprovoking guide.
I have a “what if” data table I am using to calculate marginal tax rates, using a column of amounts incrementing by, usually, $1000 and using as the what if column variable, different elements of the complete model such as deductions, capital gains, ordinary income, etc.
As far as I can see, the cell that was selected to be the column variable is not shown anywhere after the data table is run, or created. Is it possible for my spreadsheet to show what that variable was? More importantly, I would like some of the calculations I am making to be different (using if formulas) depending on whether the variable that is being incremented is ordinary income, capital gains etc. and to do that I think I need for the if formula to be able to refer to somehow to the cell that was the “seed” for the resulting data table.
Thanks for any help you might offer. Dave
@Dave
Thanks for the kind comments
You can add any cells as links to be collected/returned as part of the Data Table
If you look at the MONITORING MULTIPLE VARIABLES section you will see formulas for various measures across the top of the DT
or in the Monte Carlo section once again you will see formulas that link to individual cells
Any of these formulas can link to the cell that is the Column Link cell or any other cell or formula
The second part of your question is about setting up your model how you want to run it.
I would try and avoid changing too many variables at once, but if you decide that that is what you want to do, I would setup a DT that collects enough input and output variables so that you can analyse the results and make some sense out of them.
The other way is to have a variable that is also used to define what type of model it is eg: Model_Type: with values of 1=ordinary income, 2=capital gains etc
Then record that as part of the DT.
The MULTIWAY DATA TABLES shows you a methodology for doing that.
Amazing Post Hui…Loved every bit of it..reminded me of a project i built during my Uni using Monte Carlo Simulation [Excel] for Determining the Number of Counter to be assigned based on service time and arrival of customer in a department store. I just used the rand() function there, i wish i cud have read the post earlier so that i could use the Normal Distribution Trick
You made it look so easy
On a serious note a must mention that you are hell of a teacher Respect.
Note: Please do make a video tutorial of it…the society will be extremely benefited out of it..
@Sami
Thanx for the kind words
A video is on my To Do list
Learning something new everyday… great post!! Thx!
One question… is it possible to make the Data Table for the WhatIf Analysis dynamically size? Reason being, I am running a lot scenarios and due to the complexity of the model takes a sufficient amount of time to run each update. To speed things up, I would like to par down the list of scenarios to run based on certain toggles in the model. Of course, this would only be useful, if the DataTable used for the WhatIf Analysis can shrink down accordingly. I have already tried turning off the output variables using an if statement if a certain scenario was not applicable but it is still quite slow.
Your thoughts are appreciated. Thx.
@Aero_Engr
Thanx for the positive feedback
It isn’t possible to make Data Tables resize automatically
I nearly always run Data Tables via macros to set them up and then copy them as values when complete
This saves recalculation time in complex models
It is possible via VBA to resize the Data tables if required.
@ Aero_Engr: “is it possible to make the Data Table for the WhatIf Analysis dynamically size…I would like to par down the list of scenarios to run based on certain toggles in the model. ”
Ever had exposure to SQL? I think using SQL to do the number crunching is a good option here, as the result set is dynamic. And SQL is fast. But this can be tricky if you’re not quite sure where to start.
Thanks for your comments Jeff. Unfortunately, I’m not very familiar with SQL and, as you suggest, wouldn’t know where to start.
Hi there
This is a valuable post.
I have a quick question – any inputs or suggestions would be valuable!
I am using a simple data table (1 variable). The variable has a low, base and high values, which are dependent on a user choice in another sheet. So I have used an index/match function to retrieve low/base/high for the variables. When I use either an index/match, index or offset functions to retrieve these values, the data tables does not update. is there a way to use volatile excel functions and still get the data tables to work?
thanks!
@Sreenix
Can you email me the file:
Email at bottom of: http://chandoo.org/wp/abouthui/
Hi,
Thank you very much for your post. I’ve been trying to figure out how to do a combination of your multiway tables with the advanced monte carlo simulation. What I would like is a scenario multiway datatable that returns aggregate functions of a monte carlo simulation datatable that runs 1000 or more iterations of said scenario. Can you link two data tables in such a way? I’ve been having trouble doing so.
Thanks
@Chase
Unfortunately, Data Tables can’t be daisy chained like that
I think it is a deficiency in Excel and is probably made like that to stop Excel freezing on large models
Even if they would allow 2 levels it would be useful in lots of circumstances.
They easiest way around it is to setup a quick look in VBA to Setup the Data Table
Copy the value/calculation off to the side
recalculate the Data table
Copy the value/calculation off to the side
etc
If your unfamiliar with VBA I’d be happy to do that for you
Click on Hui… above, Email is at bottom of page
Hui,
This is just toooo good. I want to do operations which imitate the functionality of Crystall ball/risk solver, those kind of tools. Is it possible to do them with data tables and monte carlo analysis ??
Pls guide me how to proceed
Thank you
@Abhinandan
Thanx for the kind words
I’m not familiar with Crystal Ball but most of the @Risk modelling functionality can be replicated by using then techniques described in the post and then sending the output into Pivot Tables or other methods for further/statistical analysis.
However if you are into serious modelling these techniques can negatively affect spreadsheet performance.
Addin’s like @Risk etc are highly optimised and can often do these functions a lot faster than the techniques I have described.
Note that I have a section dealing with performance in the post above and describe one way of mitigating poor performance.
I haven’t used @Risk since learning these techniques over a decade ago
I use these techniques due to portability, I know I can send a model to anyone and it will work as it requires no addins
Thanks for your help.I was able to come close to a model like how i wanted. I used a plug in which provides distributions for other functions as well.
I wanted to know if data table can be controlled by VBA. Instead of pressing a F9 everytime with rand() function being used in data tables.
Can VBA coding be used ?
@Abhinandan
Yes, Data Tables can be controlled by VBA
There are notes at the bottom of the post above about how to do that
hi Hui,
i’ve tried input with another figure of gold grade.
then i did the normal procedure, block the entire blue cells, and go to data – what if analysis – data table, click the gold grade on column input cell, click ok.
but the profit column figure won’t change.
could you help me on this.
thank you
@Nopian
Can you email me a copy of your model
Click Hui… above
my email is at the bottom of the page
Hui…
Thanks for the this thorough example Chandoo. Please keep them coming!
Hi Hui,
Thanks for providing the great and thorough post above! I am fairly new to Excel in terms of these functions. I have a problem I’m working on and I’m unsure where to start due to the number of variables; I think data tables used in Monte Carlo simulation would be appropriate but I’d really appreciate your help.
I am trying to estimate/predict the effect of changing customer demand for each terminal of an aiport on the number of air traffic movements for each respective airport terminal, based on five years of previous data. Eventually, the aim is to see how the number of traffic movements translates into revenue and whether this is likely to increase in the future. I’d like to take into account the time value of money (NPV), both for the historical and future figures.
There is also a skew on data for one year due to a special event which I am unsure of how to account for.
In terms of generating the particular values for each variable based on past data, would the NORMINV(Rand()mu,sigma) command be appropriate, once a mean and standard deviation are calculated for the 5 years, or is this too generalised?
I would really appreciate any advice you could provide, even if it’s just where to start! Many thanks.
Hi Sammy
Once you collate your data the next thing to do is to look at the natural distribution for it.
Then construct a model that uses the date to model what is happening now
That can be validated against the natural data
Then you modify the model to reflect what you want to change
Then check/truth those results
Sounds like an interesting project
[…] Originally Posted by shg Whuchoo talkin' about, CC? Hey shg – talkin' bout multiway data tables. Here's are some examples, described in detail by the great man Hui himself: Data Tables & Monte Carlo Simulations in Excel – A Comprehensive Guide  Chandoo.org – Lea… […]
Hi All
Visit My Blog http://exceljoins.blogspot.in/to see all kinds Joins that can be performed on Excel Data. It has Inner Join, Left Outer Join, Right Outer Join and Full Outer Join Solutions
Pros
1) No Formuales writing
2) No Excel Macros preparation
3) Simply Download the Files, Input the Data, Run the Script and Check the Result
Cons
1) At present i have limited the Joins only for 2 Tables
Regards
Vijay
Edit (in 3 minu
Hi
I have a question on how to use nested IF functions instead of data tables where you may have more than 2 variables. For eg. say you are working in cell C9, you have inputs in cells C1:C3, C4 being the calculation. C8, B9 and D6 being the input assumptions used for the sensitivity.
I have seen formula like the below used before, but when I try and make one myself I cannot get it to refresh, it just shows 0.
=IF(AND($C$1=C$8,$C$2=$B9,$C$3=$D$6),$C$4,C9)
Anyone encountered these formulas before?
@MJ
You can use any formula’s you like but in this case you will only ever get two results, ie: When all conditions are True you will get the result for C4 and if any are false you will get a value using C9
The issue is that you have to manually change a cell to get it to recalculate and then record the answer somewhere
If calculation is set to Manual you will also have to force Calculation using F9
Data tables get around this by being able to calculate your model for multiple cells with multiple inputs as well as recording the output of multiple cells all at the same time.
[…] commenters seem to like http://chandoo.org/wp/2010/05/06/datatablesmontecarlosimulationsinexcelacomprehensiveguide/ […]
Wow Hui, this is incredible.
I’ve been looking all over the internet for how to do this multiway data table, very very useful! I will make sure that your posts are the first place I look in future, I have a lot to still learn from you.
Thank you for being prepared to share your hard work and to teach others.
@Roy
Thanx for the comments
Hui…
[…] […]
The article was really helpful
[…] In this example, taken from a blog post on the very interesting and useful Excel web site: chandoo.org, you can see that the MCS is not an intimidating as you may have first thought. This example simply […]
[…] In this example, taken from a blog post on the very interesting and useful Excel web site: chandoo.org, you can see that the MCS is not an intimidating as you may have first thought. This example simply […]
Chandoo…..
you’re awesome. Mandelbrot on excel…. plain wow.
@Amit
Thanx
Hui…
What if I want to vary the inputs by a certain Percentage
The above article not worked for me as you said formula is C6 1.68*1+E2, but it was not provided the answer as it your website shown. And u said in your answer F6 in formula instead of E2, explain?
Can you send me better example at my above email address
Thanks
@Faisal
I think that should say: C6 1.68*(1+E2) where E2 will contain the percentage you want to var it by
eg: if E2 has 10%
it will evaluate to C6 1.68*(1.1)
which is correct
Hui,
Thanks for this fantastic post. I have been using multiway data tables technique since 2001, and frankly, I thought I’d invented it as I hadn’t been able to find any evidence of other people doing it.
I use it in stress analysis of aircraft structures taking large numbers of finite element analysis results of aircraft structures and performing standard detail stress calculations for the many scenarios of internal loads applied to skin panels and beam members, etc.
I have explained it to people I work with, but generally kept this gem to myself.
It can save many manmonths of work in a professional context, and stops people writing clunky and fragile macros to attempt the same thing with lots of recorded copy and pastes.
When people see excel sweep through tens of thousands of complex stress analyses with dozens of inputs each and producing dozens of results in only a few seconds in one workbook instead of hundreds of uniquely named workbooks, their eyes nearly pop.
Your concise explanation of the method is excellent. I have explained it to people and they find it hard to get their head around, but you have something that I can show to people and trust that they will get it. Great for getting a team of people to understand not only how to do the method, but to appreciate how to collect the problem inputs into a structured list of scenarios.
Thanks,
Don
@Don
I Appreciate the feedback
It is strange that more isn’t written about use of Data Tables as they are hugely powerful.
Can any body the share the example of monte carlo simulations for software engineering example with perspective of high maturity area of CMMI L5
Thanks for the needfull
Regards,
shalini
@Shalini
I am unfamiliar with Software Engineering or high maturity area of CMMI L5
However the techniques in this post can be added to any system/model
If you would like to explain or post a sample file with what you want we can see if we can help you.
Hi Hui,
Very interesting article. But could you please explain me why the values of exchange rate on column 1 and 2 on one input Monte Carlo simulations are with different meaning?
Thanks
Batu
@Batu
The Formula in Column F of the Data Table Column is: =NORMINV(RAND(),0.92,0.02)
This calculates a new random value for every iteration of the data table
The value that was used in the calculation of each Profit is stored in Column G with the profit in Column H
It is because it is being recalculated every iteration, that we need to store it in a Data Table column like Column G
The Final values in Column F has no relationship to the stored Values in Columns G & H