Business Analytics and the Decision Making Process

Probability 

Probability is a branch of statistic which analyses the numbers of outcome in ratio form in an exhaustive set of an equally and likely outcome. The probability outcome should produce a given event to the total number of the possible result of that event. Probability statistics results are very critical in decision making especially in the field of business. The primary focus in defining probability lies on the formula and statistical calculations, however, primary concepts that are the determinants of an extent to which event interactions affect probability lies underneath. More importantly, people are given chances even in times of uncertainty by probability concepts as well as mathematical calculations to make good business decision. 

The concepts of probability provide ideas to use while identifying an extent to which the company is facing a risk. Alternatively, it determines the degree to which benchmark expectation differs from the potential outcome. This can be done by using the random or full data sample to base probability calculation. For instance, random sampling from the target market population commonly utilizes data from the consumer demand forecasts. However, when cost is used to make a purchasing decision, the determination of item which is closer to matching price expectation is based on the cost of each item ( Albright & Winston, 2017). 

Distribution 

The statistical distribution data is best based on the distribution fitting. In distribution fitting, data set generated by some random process that is suited by bets statistical distribution is selected. In other words, distribution fitting is applied where one is interested in finding random data availability. In such circumstances, individuals are usually interested in what particular distribution can such data be described. Why should we then consider data distribution? It is an unchallengeable fact that all areas of our life are affected by Random factors. Alternatively, a tool which can be used to deal with risk and uncertainty is much needed by businesses as they strive to succeed in the dynamic business environment. The best tool which helps business in the dynamic environment is probability distributions which are a scientific way of making informed business decisions based on uncertainty calculations ( Albright & Winston, 2017). 

For instance, many fields in the world apply the distributive data based on the probability in most of their mathematical operations and decision making. These areas are inclusive of reliability engineering, chemical engineering, actuarial science, image processing, risk analysis, mining, physics, medicine, sociology, demography, investment, market research customer support, hydrology, and many others ( Albright & Winston, 2017). 

Uncertainty 

Decision making under uncertainty is applied while solving many pressing problems. In other words, it is used when choosing actions with unknown outcomes and most often imperfect observations are involved. For instance, in automated decision support systems, various sources of uncertainty should be considered by the designers while balancing the multiple objectives of the system ( Black, 2012). 

Assumption has to be made by an individual that there's a "real value" that range of the measurements encompasses whenever one makes a measurement while collecting data. For this reason, one needs to find the best estimate of his/her measurement when calculating the uncertainty of the measurements. In a case where one has to add or subtract the measurement of risk, consideration of the results is very paramount. 

Today’s business environment is very dynamic and full of uncertainty. The managers must find ways of delivering in their respective position under the dynamic and uncertain condition the business face. For this reason, mustering uncertainty application based on the statistics is very critical. There are many instances where risk measurement data is paramount. For instances, factors such as customer demand changes, shifts in economic, government legislation, technical variation and other factors beyond the reaction of competitors. These affect issues such as a new product launching, marketing strategy change or first branch opening by the company ( Black, 2012). 

Sampling 

Sampling is a tool that is used to indicate the quantity and frequency of data collection with the intention of defining the samples to be taken to quantify a problem, system, process and many others. For example, consider a situation where one needs to find how good an orange is. Is it necessary to eat the whole orange to know? In this situation, the answerer is no. It is necessary only to taste a sample of the orange to make a judgment about the entire orange. This is the best scenario where sampling is used to make a decision. Improvement as well as monitoring sampling is where small samples over time are frequently taken. This is very paramount in today’s business. 

The sample is applied in many industries and business today in situations where decision-making between two or more alternatives is made. For instance, when collecting information or gathering data to use when faced with healthcare issue, the data to use is based the randomly sampled areas which are relevant to the information or decision to be made. 

Statistical Inference 

Statistics of Inference is employed in a situation where one is interested in measuring and estimating as well as drawing conclusions about a parameter. Often, many ways of measuring an object are possessed by scientists. One of this ways is to use statistical inference. For instance, where they are interested to know the mass of an electron, the scientist uses the statistical inference. Estimation which incorporates reasonable expectations or prior judgments which is also known as Bayesian estimation is one main approach to statistical inference is employed by the scientist when making a decision. It also involves new observations or experimental results. There are other methods of statistical inference which scientist prefers most. Likelihood approach is a good example. Under this approach, prior probabilities for calculating a value of the parameter are eschewed. The observed distribution of experimental outcomes is most likely to be produce by these values ( Sharma et al . 2014). 

A particular mathematical form in parametric inference that affects distribution function is assumed. On the other hand, this assumption is avoided by the nonparametric inference. Most often in a situation where a value of an unknown distribution is to be estimated by an individual where he/she has an unknown functional form, a nonparametric inference is utilized. 

Statistical inferences used when making a decision to carry out scientific research in an environment. For instance, when experiments are done in well-controlled environments 

Inference based on hypothesis testing is advisable. Alternatively, Type II error is of little consequence when addressing fundamental science questions ( Sharma et al. 2014). 

Regression analysis 

Regression analysis is a method that scientist and statisticians use to measure the phenomena’s link. Regression is a very paramount tool for decision making particularly when statistics are involved today. For instance, imagine a situation where one decides to analyze or only know the link between the prices and the actual size of the house. A regression presents some critical charts which will enable the person to link the relationship between these two variables and in so doing; he/ she will be enabled to pinpointing an average causal effect of the connection. 

More precisely, take bivariate regression which is linear in nature and where there is description relationship between two unchanging phenomena as an example. In case one would like to know if there is a connection between the times he/she spent to do homework and received a grade, he/she can plot the data as points on a graph. In this case, the exam overall scores are represented by the y-axis, and the average number of hours weekly is represented by the x-axis. The points representing data will be scattered. Single line is created in regression which all distribution points are summarized ( Sharma et al. 2014). 

Time Series 

A time series is a point in time order which is indexed, listed or graphed a series of data. More importantly, a time series sequence is based on spaced points in time which are taken successively as well as equally. This show that time series is a data sequence of discrete-time One major example of time series is tides’ heights in ocean ( Albright & Winston, 2017). 

Time series analysis is an important scientific and statistical tool where methods for analyzing data from time series is comprised. The data analysis using time series is done with the intention of getting relevant statistics. The model application in using the previously observed values to predict future values in time series analysis is known as time series forecasting. Time series is applied widely in the context of data mining where it is used for clustering, classification and many other activities ( Albright & Winston, 2017). 

Forecasting methods 

The application of data during the future trends’ direction determination in business or any other field is forecasting. In a bid to determine how to allocate their budgets or plan for anticipated expenses for an upcoming period, companies utilize forecasting. They based the focus data on two major issues that are the projected services they offer and demand the goods ( Albright & Winston, 2017). 

The appropriate forecasting method is dependent largely on what data are available to be used by the statistician or the organization. For instance, qualitative forecasting methods must be utilized in a case where there are no data available. Alternatively, the same methods of forecasting can also be employed when the data available are not relevant to the forecasts. The user of the qualitative method should bear in mind that the methods are not purely guesswork. For this reason, they should be well developed based on structured approaches to obtaining good forecasts with less usage of historical data. According to Albright and Winston (2017), q uantitative data forecasting is used where there is empirical statistics to be analyzed. An example is a financial statement which most often used in large business or profit oriented organizations. 

Optimization 

Optimization is one of the most important statistical tools used in both statistic and the field of mathematics a well as an engineering field. Alternatively, Optimization in conjunction with the Decision Theory research is very critical on research-based decision making. The two mainly involve the development of algorithms based on linear, dynamic, nonlinear, integer, vibrant and stochastic programming techniques which have a variety of importance in the modern error engineering industries. When applied to major problems encountered in real life, optimal algorithms need a frequent as well as excessive computation of resources and time. It is very effective to develop heuristic algorithms through exploiting the problem structure in the case where it is applied in life. For this reason, both classical heuristics and metaheuristics research are employed in this Group. This method has increased in popularity among many industries in the world ( Holsapple, et al. 2014). 

Optimization or decision theoretical models are of primary concern in much scientific and engineering discipline. Some examples where optimizationis used for decision making includes financial systems, telecommunication systems, manufacturing plants, engineering design, biological and logistics networks. For instance, to solve a significant design problem in structural mechanics, an engineer should find the optimal solution of the corresponding equilibrium problem. More importantly, costs in the big business can significantly be reduced through applying optimization techniques to operate their supply chains or design them ( Holsapple, et al. 2014). 

Decision Tree Modeling 

One of the better-known decision-making procedures for a scientist is the decision making a tree. Its popularity is attributed to its integrated set of choices, the associated uncertainties, and outcomes as well as the ease at which it visually communicates decisions. They have simple structure which enables them to be applied in various places. Many people prefer the decision tree since it can help quickly outline or communicate the decision making critical elements simply by drawing it. Alternatively, it has an easy and orderly structure which makes it easily applicable in tricky scenarios. Also due to its simplicity, it can be used to solve problems with the aid of computers as asserted by Albright and Winston (2017). 

The natural "if ... then ... else ...” possessed by the decision tree makes it fit comfortably into a programmatic structure. Some good example of the application of decision tree includes where the most likely buyers of a product in the organization it to be determined to enable targeting of limited advertisement budget by using demographic data ( Black, 2012). 

References 

Albright, S. C, & Winston, W. L. (2017).  Business Analytics: Data Analysis and Decision Making  (6th ed.). Boston, MA: Cengage Learning 

Black, K. 2012. Business Statistics for Contemporary Decision Making (7th ed.). 

Holsapple, C., Lee-Post, A., &Pakath, R. (2014).A unified foundation for business analytics. Decision Support Systems , 64 , 130-141. 

Sharma, R., Mithas, S., &Kankanhalli, A. (2014). Transforming decision-making processes: a research agenda for understanding the impact of business analytics on organizations. European Journal of Information Systems , 23 (4), 433-441.