Standard Statistical Methods for Quantitative Data

Business intelligence has many tools which deal with quantitative data, in the aim making better decision while using these data. Firstly, the data must be collected and gather from any related data sources to the organization. Once, the data collected, it must be mining and analysed , to determine which data going to be useful and which are not to the decision. Then, many business intelligence standard statistical methods can apply for this day.
There are many examples of standard statistical methods, like decision trees, predictive models, sensitivity analysis, and many others. However, not all of them can be used for all type of decisions, it depends about what data you have and what decision you are going to make, you must know these two things, then you can determine which tool will help to make the decision.
For example, a decision tree is an analysis method for structured decisions; appropriate for when actions are to be performed in a sequential order. An advantage is that it can be easily interpreted by a layman. However, a disadvantage is that it cannot be used for all scenarios, especially ones that are more complex.
Sensitivity analysis is a tool that enables a company to engage in forecasting based on historical data. This can be an advantage in that it allows company to predict likely outcomes and prepare for them. However, serious disadvantages include the fact that a sensitivity analysis system is only as good as it is programmed to be. Secondly, if the poor quality data is input, it is likely that the wrong outputs will be given as well.
Another example of statistical methods is the analytic hierarchy process. AHP, Processing allows for decision makers to create criteria whereby all the desired attributes are listed and how desired those attributed are. The all desirable attribute are given a weighting that add up to one, with the most favoured attribute given the biggest value. The advantage of this is that decisions makers can follow logical steps to reach a conclusion. However, a disadvantage is that the desirability of these attributes is very subjective; there is no universal measurement to determine how useful something is. Therefore, attribute A might be very desirable to person A but not desirable to person B.
Therefore, from the above three examples, it’s clear that not all tools can be used for any decision, it depends why do you want to use the tools, and what decision do you want to make. For instance, if you want to make decision based on forecasting the future, you are not going to use a decision tree, and instead you can use a sensitivity analysis, or what if analysis.
In addition, it is important to have clear, related, up to date data to make any decision based on those types of Business intelligence methods. Since if the input was not right input or related to the decision, then the decision is going to be based on this data which the company used. Therefore, they will have result which might cause many problems to them, especially if the company used the tool to predict something in the future, and pay money to prepare for it, and at the end what they were expecting does not happened.

Advertisement