Primer Home /
Estimation / Complextimation
Complextimation
Topic: Estimation
by Manal, 2019 Cohort
Problems are often complex largely due to their sheer magnitudes. Consider, for example, the problem of poverty. Helping a particular individual to rise out of it may be a difficult task, but addressing the problem for a whole city has a completely different level of complexity. Such problems are however, commonplace and a plan to tackle them needs to be created.
The added challenge with these problems of scale is partly due to the fact that there is a large disconnect with the ground level. For example, a first step in an attempt at alleviating poverty might be to gather a count of the number of people affected by it. However, such statistics are hard to find, and oftentimes never present.
On the good side, exact values for these statistics rarely matter and thus they need not be perfectly accurate. Obtaining the right magnitude is generally acceptable most times. This means that instead of using rigorous data collection techniques to gather the statistics, one might use the technique of estimation to derive a ballpark figure for the data in a matter of minutes.
Estimation Techniques#
Estimation of data usually only involves simple arithmetic. The standard procedure for estimation involves calculating the required quantity using data that is either known, guessed based on reasonable assumptions or estimated themselves.
Let us consider the problem “How much money is spent by people in Australia on pizzas on a typical Friday night?” The answer to this problem is unknown as such statistics are either unavailable or never published. We can however, estimate the number using different techniques.
One method to approach this problem may be a top-down technique. For instance, we can assume that of the 25 million people in Australia, 1% of the people order pizza on a Friday night. This percentage is a hunch based on the author’s observations of the real world. A better method for arriving at the figure may consider the various segments of the population, and arrive at a more appropriate figure. We can also assume that the average money that is spent by each person is $10. Thus, one can calculate the required amount as
While this is unlikely to be the “correct” answer to the problem, it certainly provides a reasonable estimate for it. Interestingly, this is not the only way to estimate the answer to the problem. A business perspective may involve estimating the number of pizza chains and the average earnings of each store. The optimality of the methods is mainly dependent on the domain knowledge of the person performing the estimation.
However, calculating the amount spent on pizzas is only one of the problems that estimation can simplify. In an information rich world where numbers are ubiquitous, the technique of estimation can be helpful in unravelling the complexity of almost any large problem whether it be as simple as finding the optimal location for opening a new restaurant or as hard as alleviating poverty in a city.
Explore this topic further#
Return to
Estimation in the Primer
Disclaimer#
This content has been contributed by a student as part of a learning activity.
If there are inaccuracies, or opportunities for significant improvement on this topic, feedback is welcome on how to improve the resource.
You can improve articles on this topic as a student in "Unravelling Complexity", or by including the amendments in an email to: Chris.Browne@anu.edu.au
Problems are often complex largely due to their sheer magnitudes. Consider, for example, the problem of poverty. Helping a particular individual to rise out of it may be a difficult task, but addressing the problem for a whole city has a completely different level of complexity. Such problems are however, commonplace and a plan to tackle them needs to be created.
The added challenge with these problems of scale is partly due to the fact that there is a large disconnect with the ground level. For example, a first step in an attempt at alleviating poverty might be to gather a count of the number of people affected by it. However, such statistics are hard to find, and oftentimes never present.
On the good side, exact values for these statistics rarely matter and thus they need not be perfectly accurate. Obtaining the right magnitude is generally acceptable most times. This means that instead of using rigorous data collection techniques to gather the statistics, one might use the technique of estimation to derive a ballpark figure for the data in a matter of minutes.
Estimation Techniques#
Estimation of data usually only involves simple arithmetic. The standard procedure for estimation involves calculating the required quantity using data that is either known, guessed based on reasonable assumptions or estimated themselves.
Let us consider the problem “How much money is spent by people in Australia on pizzas on a typical Friday night?” The answer to this problem is unknown as such statistics are either unavailable or never published. We can however, estimate the number using different techniques.
One method to approach this problem may be a top-down technique. For instance, we can assume that of the 25 million people in Australia, 1% of the people order pizza on a Friday night. This percentage is a hunch based on the author’s observations of the real world. A better method for arriving at the figure may consider the various segments of the population, and arrive at a more appropriate figure. We can also assume that the average money that is spent by each person is $10. Thus, one can calculate the required amount as
While this is unlikely to be the “correct” answer to the problem, it certainly provides a reasonable estimate for it. Interestingly, this is not the only way to estimate the answer to the problem. A business perspective may involve estimating the number of pizza chains and the average earnings of each store. The optimality of the methods is mainly dependent on the domain knowledge of the person performing the estimation.
However, calculating the amount spent on pizzas is only one of the problems that estimation can simplify. In an information rich world where numbers are ubiquitous, the technique of estimation can be helpful in unravelling the complexity of almost any large problem whether it be as simple as finding the optimal location for opening a new restaurant or as hard as alleviating poverty in a city.