GUESTIMATE methods are a combination of mathematics, statistics and common sense of getting meaningful results for decision making.
The most famous approach was suggested by Enrico Fermi. Mr. Fermi was convinced that an error of order of magnitude (up to ten times) can be accepted for certain problems.
There are many useful results can be obtained by using probability estimation.
Expected values are largely used in economics, business, finance, and opportunity choice.
Old fashion rule of thumb is a widely used approach among non-mathematicians.
Certain results can be obtained if we consider either limits or average values of variables.
This is a brief introduction to GUESTIMATE methods, which are a combination of mathematics, statistics and I would say common sense of getting meaningful results for decision making.
Guesstimate is an informal English combination of guess and estimate, for the first time, It was used by American statisticians in 1934. It is defined as an estimate made without using complete information, or, as an estimate arrived at by guesswork.
However, we always balance between simplification and usefulness of the results. The purpose of guesstimate is insight and not receiving numbers.
This is a list which is based mostly on examples.
There are extremely large number of applications which use guesstimate methods.
The most famous approach was suggested by Enrico Fermi. Mr. Fermi was convinced that an error of order of magnitude (up to ten times) can be accepted for certain problems.
There are many useful results can be obtained by using probability estimation.
Expected values are largely used in economics, business, finance, and opportunity choice.
Old fashion rule of thumb is a widely used approach among non-mathematicians.
Certain results can be obtained if we consider either limits or average values of variables.
Fermi problems typically involve making justified guesses about quantities and their variance or lower and upper bounds. The classic Fermi problem, generally attributed to Fermi, is "How many piano tuners are there in Chicago?"
Enrico Fermi estimated the strength of the first atomic bomb based on the distance traveled by pieces of paper he dropped from his hand during the blast.
Another example
How much time does it take to read all books on “Data Science” in English?
First we need to estimate how many books and page are available.
I choose as a source Amazon and find that twelve hundred books are available.
Giving reasonable estimation for the number of pages and my capacity , I conclude that It takes about 30 years.
Probably, new books will be published soon.
Probability can be involved when a dealer sets prices for his cars.
Expected is what we expect to get in the future. Expected values are use for choosing among different options.
The rule of thumb is a principle with broad application that is not intended to be strictly accurate or reliable for every situation. The Rule of Thumb method is known for four hundred year with this name and probably It was born before the Math was invented.
If we set boundaries, for example for the Bermuda triangle, the notion of Atlantic Ocean becomes smaller and more clear.
After choosing the boundaries, you can continue with classical mathematical methods.
The more time you spend on elaborating of the model and data mining, the more precise and useful result you get. For certain problems there are limits, where further development does not improve the accuracy.
There is no much information in the example. However a simple solution exists.
If there are N errors in total, then John’s performance is 2 divided by N and Mary’s performance is 3 divided by N. Common detected error is considered as a dependant event, so the common performance of both testers is the product of the individual rates: Now we can calculate the expected number of errors N = 6 and the number of undetected errors which is 2.
These are typical guestimate questions. Try to solve them without looking into the answers at the website A+Click.
As a conclusion, there are 2 most important points for me .
First : never give up – continue elaborating different models, search for data until you reach useful results.
Second: Perfect practice makes perfect. The more you practice to solve the problems, the more agile and skillful you will be. Simple.
If you wish to go deeper into the subject and see more example, there is an useful book on guesstimation with dozens more elaborated examples.
All answers to the mentioned here problems you find the A+Click website.
Thank you watching and good luck in Your guesstimates.