4 ms·

the best i've seen is a woman who did corporate financial planning and analysis. i could ask a question and she would have a pivot table spit out the answer as

the best i've seen is a woman who did corporate financial planning and analysis. i could ask a question and she would have a pivot table spit out the answer as i finished the question. i consider myself a good* (but not great) excel user, but she was amazing! * having done complex financial modeling incorporating monte carlo simulation for predictive what-if analysis

paulryanrogers 8y ago
How did she handle the precision problem?
- clairity 8y ago
  not sure which precision problem you're referring to, but in general, we understood that excel's precision issues are probably overwhelmed by the imprecision of the model. but i'd be more weary of using excel if we were doing scientific research (R or matlab would be better there).
intended 8y ago
Modeling is very different from pivots. It’s practically an entirely different skill set. But man are they fun to use once you get the hang of them. Hmmm. This helps me realize one of the many things excel-should-be-replaced by code proponents miss. There will be far more excel masters over time, than language masters. Simply because excel remains stable, and is not prone to flavor of the year issues.
julienfr112 8y ago
I didn't now you could do monte carlo (and what-if) with excel. Can you tell us more about that ?
- Jabbermonkey 8y ago
  https://support.office.com/en-us/article/introduction-to-monte-carlo-simulation-in-excel-64c0ba99-752a-4fa8-bbd3-4450d8db16f1 https://support.office.com/en-us/article/introduction-to-mon... https://support.office.com/en-us/article/introduction-to-what-if-analysis-22bffa5f-e891-4acc-bf7a-e4645c446fb4 https://support.office.com/en-us/article/introduction-to-wha...
  - clairity 8y ago
    those are great resources. i've also used the crystal ball plugin for excel, which was bought out by oracle a few years ago: https://www.oracle.com/applications/crystalball/ https://www.oracle.com/applications/crystalball/ scenario modeling is useful when you have a set of uncertain inputs and you want to estimate the range of possible output values. the monte carlo part is providing random sets of inputs that conform to the distributions for each input (often estimated as just a normal distribution for each). you run thousands (or more, depending on the margin of error you wish to achieve) of these random input sets to generate the mean, variance and estimated error of the output variable. in financial modeling, the inputs are typically estimates of future revenue, costs, cost of capital, etc., and NPV (net present value) of the enterprise as the output. sometimes you even include environmental or regulatory uncertainty (as a binomial value) in the model.

paulryanrogers 8y ago

How did she handle the precision problem?

clairity 8y ago
not sure which precision problem you're referring to, but in general, we understood that excel's precision issues are probably overwhelmed by the imprecision of the model. but i'd be more weary of using excel if we were doing scientific research (R or matlab would be better there).

intended 8y ago

Modeling is very different from pivots. It’s practically an entirely different skill set. But man are they fun to use once you get the hang of them. Hmmm. This helps me realize one of the many things excel-should-be-replaced by code proponents miss. There will be far more excel masters over time, than language masters. Simply because excel remains stable, and is not prone to flavor of the year issues.

julienfr112 8y ago

I didn't now you could do monte carlo (and what-if) with excel. Can you tell us more about that ?

Jabbermonkey 8y ago
https://support.office.com/en-us/article/introduction-to-monte-carlo-simulation-in-excel-64c0ba99-752a-4fa8-bbd3-4450d8db16f1 https://support.office.com/en-us/article/introduction-to-mon... https://support.office.com/en-us/article/introduction-to-what-if-analysis-22bffa5f-e891-4acc-bf7a-e4645c446fb4 https://support.office.com/en-us/article/introduction-to-wha...
- clairity 8y ago
  those are great resources. i've also used the crystal ball plugin for excel, which was bought out by oracle a few years ago: https://www.oracle.com/applications/crystalball/ https://www.oracle.com/applications/crystalball/ scenario modeling is useful when you have a set of uncertain inputs and you want to estimate the range of possible output values. the monte carlo part is providing random sets of inputs that conform to the distributions for each input (often estimated as just a normal distribution for each). you run thousands (or more, depending on the margin of error you wish to achieve) of these random input sets to generate the mean, variance and estimated error of the output variable. in financial modeling, the inputs are typically estimates of future revenue, costs, cost of capital, etc., and NPV (net present value) of the enterprise as the output. sometimes you even include environmental or regulatory uncertainty (as a binomial value) in the model.