Decision Modeling & Analysis Final Paper

Decision Modeling & Analysis Final Paper

BUS 461 Decision Modeling & Analysis

Introduction

Data analyzation is essential to growth of an organization, as doing so greatly assists them in making sound, intelligent and rational decisions. Throughout the course of this class, though this was one of my more challenging classes I have had the pleasure of learning the elements of analysis and decision modeling. For this paper, I will discuss the following elements of decision modeling and analysis: Probability, Distribution, Uncertainty, Sampling, Statistical Inference, Regression Analysis, Time Series, Forecasting Methods, Optimization and Decision Tree Modeling. Furthermore, in providing an explanation of those elements to my best understanding, I will also provide scenarios and examples of the same.

Probability

According to our text, Probability is, “a number between 0 and 1 that measures the likelihood that some type of event will occur, where the event with a probability of 0 cannot occur while the probability of 1 is more than certain to occur,” (Albright & Winston, 2017). Using probability enables organizations to obtain information and carefully assess the data to make the correct decision. With probability, no matter the amount of data collected the model itself is deterministic, that is, they predict exactly what will happen under certain circumstances (Kinney, 2015). For example, when the NFL conducts a coin toss to determine which team is the receiver first, with probability there’s just as much of a fair chance that the result would be tails just as much as it would be heads.

Distribution

Now that the data has been collected, the next key point is distribution, which is described as the uncertainty of a numerical outcome, where all possible outcomes and results or corresponding probabilities are listed according to every possible outcome (Albright & Winston, 2017). In other words, once the data is obtained it must be distributed in a manner that is both understandable and presentable. Only then can the data be analyzed and presented. In doing so, this allows for personnel who receives the information to view it with a clear understanding so they can know which option is the best way forward for their organization.

Uncertainty

When conducting an analysis, there is always a bit of uncertainty, meaning there is a situation where full knowledge of the nature of or consequences of an event cannot be accurately predicted, and the probabilities can’t be assigned in a proper manner. “Uncertainty often amounts to no more than the mere insufficiency of the usual safety valves,” (Bonatti, 1984). In other words, uncertainty deals with a lack of information to be considered to make sound decisions, however when presenting the uncertainty in analysis, analysts can use the data as a means of comparing data with different scenarios. When attempting to venture into the unknown, uncertainty prompts people to be prepared for unexpected events to occur. For example, when hiking or hunting its best to be prepared for the unknown such as cloudy weather which has the potential to generate hazardous weather not suitable to hunt or hike.

Sampling

Data collection varies with every initiation of analyzing and obtaining the same. In some cases, the data collected may be small and easy enough to interpret, while in other cases the data received may be so grand that its virtually impossible to obtain all of the data of a population (Wienclaw, 2019). To combat this, researchers use samples from certain populations. This is known as a sample, where researchers take information from a part of a population or category to use for their analysis. There are several categories of sampling that are used when conducting analysis. For example, with cluster sampling “the population is separated into clusters, such as city blocks, and then a random sample of the clusters in selected,” (Albright & Winston, 2017). A researcher takes the population and divides hem into small clusters based on shared characteristics within the population. Then, random samples can be taken to obtain data for the purpose of analysis for the same (Albright & Winston, 2017). A great example of when to take a cluster from a population would be to determine the type of vehicles people chose to purchase. This can be divided between states and cities such as the example in the text states, because the vehicle choice varies from person to person, city to city and state to state. Corporations can utilize this information to determine the number of vehicles to mass produce and where to send them and specifically which locations require more than others.

Statistical Inference

Once data has been collected, sorted and ready for review, analysts can make conclusions and provide recommendations based on the sample selected from a population. The process of conducting this is known as statistical inference. For example, Pew Research Center conducts many surveys and tests based on population. If a specific state was being analyzed, a sample of a town or city would be selected. So, if Louisiana was the state being surveyed, then popular cities such as New Orleans, Baton Rouge or Lake Charles would be the samples used. Furthermore, you could also break down the sample by wards or parishes (in other states, these terms would be known as towns and counties, respectively) for those cities to obtain more accurate data.

Regression Analysis

Regression analysis is a process of analyzing and creating mathematical models to predict values. “Although regression analysis is widely used in business, it makes several assumptions included that the model is correct and that the data is good, but real-world data tend to be messy and in result these assumptions are rarely true,” (Wienclaw, 2019). Meaning the models are not always accurate and when conducting interpretation analysts must be careful to while looking into the results. The independent variable is known as x while the dependent variable is known as y and is displayed using a scatterplot; “which purpose is to make a relationship apparent or show the lack of it,” (Albright & Winston, 2017). In further, the scatterplot is a graph which shows the values from the data which has been collected. The linear regression line represents trends and common occurrences of the data, which allows for a prediction of correlation to be made.

Time Series

Time series data is closely used in concert with forecasting, as a means of analyzing patterns, trends and cycles from previous events to predict events in the future (Wienclaw, 2019). It is used by many organizations to assist with future decisions by looking into previous events by gathering data collected during a certain timeframe. In this manner, organizations such as clothing lines can use previous years data to determine what’s in style or what to produce more of for next season. Furthermore, in the same case organizations can use this data to determine what decisions to make to avoid past business decisions that resulted in failures.

Forecasting Methods

Similar to time series, forecasting is a cost-effective management process of predicting future events. The mantra of forecasting declares: “a business succeeds or fails in proportion to its ability to forecast the future trend of the influences determining the relation of supply and demand of their business,” (Brookmire, 1913). Forecasting is a pivotal tool to be used by organizations as it provides what is necessary for an organization to continue their success by identifying the needs and demands of its customers. However, with forecasting organizations should utilize care and caution, as there are several external and internal factors that may affect the outcome of forecasting.

Optimization

Analysts use optimization models to provide organizations the ability to gather options while reducing costs.

Decision Tree Modeling

The decision tree model is a model used to depict outcomes from various scenarios to allow for organizations the ability to obtain a visual picture of choices available to the company. Using this model reduces risks, and greatly increases the chances of making a sound business decision.

Conclusion

This course was highly challenging yet provided a variety of information regarding decision models and analyzing. I have analyzed each of the elements presented throughout the duration of the course to my best understanding. Having a sound understanding of the elements and methods used to analyze data is essential to practical application in the business world. As a student who has not yet been introduced into the business world, I hope to gain a deeper understanding of the terms and hope to utilize them in the future.

References

Albright, S. C, & Winston, W. L. (2017). Business Analytics: Data Analysis and Decision

Making (6th ed.). Retrieved from https://redshelf.com/

Kinney, J. J. (2015). Probability : an introduction with statistical applications. Hoboken, New

Jersey : John Wiley & Sons, Inc., [2015]. Retrieved from http://search.ebscohost.com.proxy-library.ashford.edu/login.aspx?direct=true&db=cat02191a&AN=aul.10992239&site=eds-live&scope=site

Wienclaw, R. A. (2019). Sampling. Salem Press Encyclopedia. Retrieved from

http://search.ebscohost.com.proxy-library.ashford.edu/login.aspx?direct=true&db=ers&AN=89185687&site=eds-live&scope=site

Bonatti, L. (1984). Uncertainty. [electronic resource] : studies in philosophy, economics, and

socio-political theory. Amsterdam : B.R. Grüner, 1984. Retrieved from http://search.ebscohost.com.proxy-library.ashford.edu/login.aspx?direct=true&db=cat02191a&AN=aul.10511289&site=eds-live&scope=site

James H. Brookmire. 1913. “Methods of Business Forecasting Based on Fundamental

Statistics.” The American Economic Review 3 (1): 43. http://search.ebscohost.com.proxy-library.ashford.edu/login.aspx?direct=true&db=edsjsr&AN=edsjsr.1828258&site=eds-live&scope=site