Thursday 27 October 2016

Modeling in business and academic fields

Nowadays the society, especially the intelligence group, loves numbers and numerical evidence as numbers are considered to be biased and easier to state opinions. Modelling as part of the statistics is widely used in business researches and academic researches. We can compare the social science academic work from centuries ago and the work now, for example, when we compare Adam Smith's 'Wealth of Nation' and many current economic work, Adam Smith does not include as much numerical evidence and statistical analysis as that is included in nowadays work. Moreover, once a new theory is first introduced, the theory developer is more likely to use simple numbers to describe their theories. Sometimes they may also use data from the real world, but once the theory is recognised and developing further, more real world data will be tested to approve the theory and further improve the theory.

In addition, in many business companies, modeling has been widely and regularly used. Modeling is used to analyze customers' behavior, employees' productivity, marketing and etc. In the financial sector, modeling is even more widely and frequently used in order to forecast the market movement. However, in the business, modeling is used to project the market future performances and a rough trend sometimes is good enough. Moreover, in the business, heteroskedasticity issue is often ignored and some other statistical issues are also ignored. Because of the limited resources and the information privacy, in the business field, they can merely use the available resources and data for modelling. In addition, modeling is a very complicated job and ability, not every employee can develop their own models and even if they can, the models they develop do not have the same quality; therefore, within one company, employees tend to use same or similar modelling tools that have already been developed by the company experts. There is one problem here that the modelling tool can be outdated and if it is not accurate, the entire company researches based on this modelling tool will be imprecise.

However, in the academic field, people love to use the models they develop by themselves, and even if they borrow from others, they tend to make some improvement, which means they are technically not using the same. Therefore, in the academic, when people use different models, they can have different results, once more researches on one particular topic, the topic will be studied and tested by many different models; therefore, the results tend to be more accurate than the researches done by the business companies.

No comments:

Post a Comment