ibrahimdaas
New Member
- Joined
- Sep 1, 2014
- Messages
- 8
- Office Version
- 365
- Platform
- Windows
Hi
I have a project with many types of cost line items in Excel. Each cost needs to be spread over a number of months in something approximating a bell curve where I can set a parameter for the skew and "flatness" of the curve. i.e.: are costs front-loaded or end-loaded and how much greater is the most expensive month from the least expensive month in my time series.
The distribution "shape" depends on the kind of cost. Obviously this is easy to evenly distribute (total cost / number of months for each month). My formula can also "normally" distribute these costs in a typical bell-curve shaped distribution. In construction this called an "s-curve" for obvious reasons. By adjusting the Standard Deviation for a cost, I can tell the formula how "steep" or "flat" I want my curve. My formula derives the % under the curve for a time period and multiplies by the total cost to get a normally-distributed cost curve.
My problem is that most of my costs aren't normally distributed (bell-curve) or flat. They are very often front-loaded or rear loaded in a skewed bell-curve. Excel doesn't appear to support any skew for a normal distribution. What are some alternative approaches?
This is my S-Curve ..
http://store2.up-00.com/2016-03/145877201776591.png
I hope to find some help
I have a project with many types of cost line items in Excel. Each cost needs to be spread over a number of months in something approximating a bell curve where I can set a parameter for the skew and "flatness" of the curve. i.e.: are costs front-loaded or end-loaded and how much greater is the most expensive month from the least expensive month in my time series.
The distribution "shape" depends on the kind of cost. Obviously this is easy to evenly distribute (total cost / number of months for each month). My formula can also "normally" distribute these costs in a typical bell-curve shaped distribution. In construction this called an "s-curve" for obvious reasons. By adjusting the Standard Deviation for a cost, I can tell the formula how "steep" or "flat" I want my curve. My formula derives the % under the curve for a time period and multiplies by the total cost to get a normally-distributed cost curve.
My problem is that most of my costs aren't normally distributed (bell-curve) or flat. They are very often front-loaded or rear loaded in a skewed bell-curve. Excel doesn't appear to support any skew for a normal distribution. What are some alternative approaches?
This is my S-Curve ..
http://store2.up-00.com/2016-03/145877201776591.png
I hope to find some help
Last edited:
