Application of Multiple Regression Analysis in Supply Chain and Logistics Management Decision - Coursework Example

Quantitative Concepts Name Course Lecturer Date Executive Summary There are as many contexts in which multiple regression analysis can be used in organisation; it can be used in various processes and functions. The first part of this report indicates that multiple regression analysis is the single most useful method in respect of analysis and decision making or problem solving in supply chain and logistics management. It indicates that this analysis takes in to account numerous variables and hence it is accurate. It also indicates that it uses research and turns past data in to actionable information. In the second part, the report identifies several specific supply chain and logistics management decision and problem solving functions in which multiple regression analysis is applied. It demonstrates how it is applied directly in decision making or problem solving as well as how its application helps the decision makers to reach to a more informed decision in the context of the identified issues. It identifies functions such as marketing, manufacturing, human resource, production, sales and advertising. The last part discusses the limitations of application of multiple regression analysis. The report identifies limitations such as limited use if there is no straight line, differentiating between correlation and causation variables and it does not indicate the causal agents. It outlines what can be done to minimise the negative consequences of the limitations in the specific decisions. Likewise, the report indicates that inclusion of many variables can help to increase accuracy, identifying causal agents, and to consider other factors that may affect variables. Table of Contents 1.0 Introduction 1 2.0 Why Multiple Regression Analysis Is the Single Most Useful Method In Respect Of Problem Solving and Decision Making In Supply Chain and Logistics Management 3 3.0 Application of Multiple Regression Analysis in Supply Chain and Logistics Management Decision 5 4.0 Limitations of Using Multiple Regressions 10 5.0 References 11 1.0 Introduction The later stages of the 20th century have been marked by swift developments of research methods in real problem solving and decision making. There have been important institutional and structural changes as well as rapid progress of information technology; they have formed new scenery of the economic and corporate environments towards harnessing decision making and real problem solving as Johnson & Christensen (2010) describes. In this light, the process of contribution that quantitative concepts and methods have made in management and functional decision making has been significant. In all aspect of daily living, quantitative concepts and methods are applied and used to assist in making decisions as well as solving real problems. In order to function effectively in the modern and ever progressing business world, irrespective of the organisation, managers and persons running the organisations must be able to apply and use quantitative concepts, methods and techniques in a reliable and confident manner as Punch (2013) reinforces. Accountants, economists, marketers, personnel managers and other persons in an organisation use information that is increasingly quantitative (McNeil, Frey & Embrechts 2010). It is therefore apparent that they need a working knowledge of the concepts and procedures appropriate for evaluating and analysing such information. Such analysis as well as business evaluation cannot be delegated to mathematicians or specialist statisticians, who, skillful though might have stylish numerical analysis will most often have little understanding of business relevance of such analysis (Bonn & Cantlon 2012). For these reasons, this report aims to provide a quantitative concept and or method that are most useful in analysing, problem solving and decision making in supply chain and logistics management or an equivalent functional management decision making perspective. In addition, the report identifies a logistics management problem or decision that the quantitative concept could be directly applied to. The last part of the report provides a discussion of the limitations of using the concept or the quantitative method. The part outlines and explains what can be done to minimise the negative consequences of the said limitations in the specific problem or decision situation discussed. 2.0 Why Multiple Regression Analysis Is the Single Most Useful Method In Respect Of Problem Solving and Decision Making In Supply Chain and Logistics Management The purpose of this concept is to analyse the relationship between two variables, independent variables and dependent variables (Hair 2009). The dependent variable is also known as the outcome variable. One of the reasons why multiple regression is the most useful is because it is very flexible than the other concepts and methods. The interaction between the variables can be incorporated as well as inclusion of polynomial terms as Hair et al., (2006) asserts. For instance, in determining the relationship between age and sex, weight and height, it is possible to include height squared as well as the product of height and sex. The relationship between height and weight would then be different for women and men. The predicted difference in weight between 5 foot person and 5’1 foot person is not equal to a 6 foot person and 6’1 foot person. This indicates that multiple regressions are very flexible more than the other methods. Multiple regression analysis allows the inclusion of many variables as possible, it uses several independent variables with each of the variables monitoring for the other (Cohen et al., 2013). Another aspect of multiple regression model that make it the most useful is that is the best predictor of relationships involving multiple predictors as Sincich (2003) reinforces. In fact, the more the predictors the more the accurate the prediction will be. Significantly, multiple regression analysis uses research to predict the likely outcomes and the future trend (Kutner, Nachtsheim & Neter 2004). This is very important as this makes it to not only to be accurate but to minimise the chances of errors. By suing research, it analysis the course and from the course it predicts the outcomes. For instance, an organisation can use it provide insights of how changes in consumer spending patterns, higher taxes or shifts in the economy will affect its business. This is important in helping the business on what to do in relation to these factor (Preacher et al., 2006). In essence, multiple regression analysis turns raw data of an organisation in to actionable information. This makes it to be the most useful method of analysis. As Allison (2002) notes, this method helps persons in management level especially managers to uncover and discover relationships and patterns that they had considered or noticed previously. For example it can help marketing manager to uncover purchasing pattern on particular days or certain times of the year. This helps the manager to ensure that there are enough supplies during this certain days or certain times of the year as Graham (2003) adds. Essentially, among the quantitative concepts or methods, multiple regression analysis is the most useful. It provides an accurate outcome after consideration of data. 3.0 Application of Multiple Regression Analysis in Supply Chain and Logistics Management Decision Supply chain has become one of the most essential aspects of building competitive advantage for organisations. It emphasizes on how to maximise and optimise the overall value of an organisation by effective utilisation of resources across all functions of an organisation (Chatterjee & Hadi 2013). Supply chain is the set of value-adding activities and functions connecting organizations suppliers and customers. The supply chain involves all the parties, involved either directly or indirectly, in meeting customer expectations and requests. Supply chain includes suppliers, manufacturers, warehouses, transporters, retailers as well as customers themselves. Within an organisation, (for example a manufacturer), the supply chain comprises all the functions involved in getting and satisfying requests from customers. These functions are such as product development, operations, marketing, accounting, finance, customer service, distribution as well as other functions involved in serving customer requests (Montgomery, Peck & Vining 2012). An effective supply chain is crucial in creating and sustaining a competitive advantage in services and products of an organisation. Notably, supply chain is highly influenced by integrating and managing key elements of information. As organizations collect more and more data through the various advances of technology, multiple regressions analysis has improved opportunities to organizations in making data (fact) driven decisions as Raudenbush & Bryk (2002) outlines. The regression analysis is the most useful method in the hands of a manager in any organisation. By describing the kind of relationship between various variable in supply chain, regression helps an organisation to understand how its business works, it would also help the organisation to make predictions about its evolution. An organisation can use multiple regression analysis in its supply chain to understand the impact of product and service prices on its profitability. They can also use it to determine the effects of economic growth on its sales to customers as well as predicting its stock prices as Mertler & Vannatta (2002) adds. This is very important as an organisation can predict the movement of exchange rates and how this movement can affect its stock prices in the securities exchange market. This is essential in managing the organizations ability to maintain a steady performance in the market. Importantly, investors depend on the stock performance of an organisation to determine if they will invest in the organisation. If the organisation stocks are performing well in the market, that is, they are increasing in value and they are trading at increasing level, investors will purchase the stocks as they look to claim a stake in the organizations growth and good performance as Jaccard & Turrisi (2003) notes. On the other hand, if the organisation stocks are performing poorly in the market, they are slumping and indicating a decreasing growth in value; investors will give a wild eye and will not look to invest in the organisation. Therefore, regression analysis plays a major role in determining the performance of organisation; it helps an organisation to improve the value of its supply chain as Dawson & Richter (2006) underscores. This does not only improve the profitability of the organisation but also the growth of the organisation as well. Another way that regression analysis is very useful to an organisation in the supply chain is that it is fundamentally applied to understand how much the independent variables affect the dependent variables. For instance it is used to explain and justifying the advertising budget (Lind, Marchal & Wathen (2006). Budget is one function in the supply chain; it is used to reach out to customers. The advertising expenditures are independent variables while sales are dependent variables. By running regression analysis of the advertising expenditures and organisational sales, it is possible to determine a good equation that best explains and describes the relationship between advertising budget and the sales {sales = 90,000 + (7 times advertising expenditure)}. It interprets that, for each dollar spent on advertising, the organisational sales increased by seven times as much. An essential caution to consider is the regression coefficient of multiple determination, or R^2, which shows the strength of the relationship on a scale from zero to one. Higher values of R^2 shows a stronger relationship between the advertising expenditure and sales while lower values suggest a weaker relationship between the advertising expenditure and sales (Collard et al., 2005). A stronger relationship indicates that the advertising is highly influencing sales; it is increasing sales by a bigger margin. A weaker relationship indicates that the advertising is not influencing sales; the advertising is not effective as it is not increasing sales. Regression analysis helps to understand this mechanism (Chatterjee & Hadi 2013). This helps managers in the marketing function of the organisation to come with strategies to improve and increase sales. This explains clearly the importance and usefulness of regression analysis in supply chain of an organisation (Preacher et al., 2006). Another way that regression analysis is very useful to an organisation is by testing hypothesis. For instance, if an organisation is going through a difficult time in increasing sales, or sales are slumping, regression analysis would help to identify the problems behind the slumping of sales (Rocca, 2014). A number of factors are identified and running a multiple regression analysis follows. In this case, running an analysis of the historical data of sales and advertising expenditure is done. Also, running an analysis of number of sales personnel and the mix of urban and out-of-town stores is carried out. By removing or removing one variable or same variables from the regression model creation at a time, it is possible to determine the explanatory effects on reasons for sales decline as Moffett, McAdam & Parkinson (2003) illustrates. This is by noting decreases and increases to R^2. This is very important as it helps an organisation to know the reasons for decrease of sales. After knowing the reasons, the organisation takes appropriate actions to increase sales. As such, multiple regression analysis is the most useful concept or method in supply and logistics management in an organisation (Hair, 2009). Another important aspect of regression analysis is that it is very useful in forecasting the future. It is a very powerful tool of forecasting the future of an organisation; it is not focused on the past. Multiple regression analysis enables an organisation to take advantage of past data to extrapolate outcomes in future (Rocca, 2014). For instance, if the marketing manager wants to know how much a $2 million increase in advertising budget would impact sales, it is very possible to give a confident prediction. This is by consulting regression equation for the advertising budget and sales. In the earlier example of sales = 90,000 + (7 times advertising expenditure), it is expected that a $2 million increase in advertising budget would generate $14 million additional sales. In human personnel, regression analysis is very useful. Human resource personnel professionals use this analysis to determine equitable compensation for the organizations employees or workforce. There are a number of factors or dimensions like amount of responsibility or number of persons to supervise that contributes to the value of a job (Inda Sukatia, 2012). The processionals then conduct a market research of the salary among comparable organizations and then recording the sales as well as the respective characteristics (that is, the values on dimension) for different positions under consideration. They then use this information in multiple regression analysis to create a regression equation. The form of the equation is such like salary = 6*responsibility + 9*number to supervise. After determining the regression line, the professionals them constructs graph of the expected or predicted salaries and the actual salaries of the incumbents of the job. Thus, the human resource professionals are able to know which positions are underpaid (the positions that are below the regression line) as well as the positions that are overpaid (positions that are above the regression line); they also know the positions that equitably paid (StatSoft, 2014). This shows the usefulness of multiple regression analysis in the personnel function of an organisation. Another significant use of multiple regressions analysis in supply chain and logistics management is optimising business processes. For example, a production manager in a manufacturing organisation can use it to create a model to help him or her understand the relationship between the temperatures and shelf life of cheese produced by the organisation (Inda Sukatia, 2012). This helps the manager to determine not only the shelf life of the products under different temperatures but also how to store and market the products. In addition, the customer service function of an organisation can also use multiple regressions to determine the wait time of customers calling as well as the number of complaints from the customers. Essentially, regression analysis is an indispensable method in supply chain for different functions of an organisation (Lind, Marchal & Wathen (2005). 4.0 Limitations of Using Multiple Regressions While the regression analysis is such powerful, it has some limitations. If the user of this model cannot be able to plot a straight line to express the relationship of variables, then its usefulness becomes limited. Moreover, it is sometimes difficult to differentiate between correlation and causation. For instance, even if the multiple regression analysis might show that increasing advertising budget influences increase in sales with a higher R^2 factor, there might also be other factors like prompt economic growth that might be really responsible for the increase in sales. The challenge in the multiple regression analysis is to choose the right variables to include in the model (Rocca, 2014). To overcome this limitation, it is very important for an analyst or user of this model to select as many variables as possible; at least several of them will come out significant. This is essential as the user is capitalizing on chance. This would help to plot a straight line as many variables will make the line (StatSoft, 2014). Furthermore, it is also important to consider many predictors of the variables of interest that may affect the specific function before making decision. Another limitation of using regression analysis is that it only ascertains relationships but does not indicate the causal agents or mechanism as Yuan et al., (2006) indicates. The analysis doe s not considers alternative causal explanations. To improve this course, it is necessary to consider other highly likely causal mechanisms that may affect a function. This would improve the understanding and hence enable make a more informed decision or solve a problem adequately. This would also improve the application of the concept in supply chain and logistics management (StatSoft, 2014). 5.0 References Allison, P, D, 2002, missing data: Quantitative applications in the social sciences: British Journal of Mathematical and Statistical Psychology, 55(1), 193-196. Bonn, C, D, & Cantlon, J, F, 2012, the origins and structure of quantitative concepts: Cognitive neuropsychology, 29(1-2), 149-173. Chatterjee, S, & Hadi, A, S, 2013, regression analysis by example: John Wiley & Sons. Cohen, J., Cohen, P., West, S. G., & Aiken, L, S, 2013, applied multiple regression/correlation analysis for the behavioral sciences: Routledge. Collard, B. C. Y., Jahufer, M. Z. Z., Brouwer, J. B., & Pang, E. C. K, 2005, an introduction to markers, quantitative trait loci (QTL) mapping and marker-assisted selection for crop improvement: the basic concepts, Euphytica, 142(1-2), 169-196. Dawson, J. F., & Richter, A, W, 2006, probing three-way interactions in moderated multiple regression: development and application of a slope difference test: Journal of Applied Psychology, 91(4), 917. Graham, M, H, 2003, confronting multicollinearity in ecological multiple regression: Ecology, 84(11), 2809-2815. Hair, J, F, 2009, Multivariate data analysis: Pearson Prentice Hall. Hair, J. F., Tatham, R. L., Anderson, R. E, & Black, W, 2006, Multivariate data analysis, Upper Saddle River, NJ: Pearson Prentice Hall. Jaccard, J., & Turrisi, R, 2003, Interaction effects in multiple regressions (Vol. 72): Sage. Johnson, R. R. B., & Christensen, L, B, 2010, educational research: Quantitative, qualitative, and mixed approaches: Sage Publications. Kutner, M. H., Nachtsheim, C., & Neter, J, 2004, applied linear regression models: McGraw-Hill/Irwin. Lind, D, A, Marchal, W, G, & Wathen, S, A, 2005, Statistical techniques in business & economics: Canberra. Lind, D. A., Marchal, W. G., & Wathen, S, A, 2006, basic statistics for business and economics, Boston: McGraw-Hill/Irwin. McNeil, A. J., Frey, R., & Embrechts, P, 2010, quantitative risk management: concepts, techniques, and tools: Princeton university press. Mertler, C, A, & Vannatta, R, A, 2002, advanced and multivariate statistical methods; Los Angeles, CA: Pyrczak. Moffett, S., McAdam, R., & Parkinson, S, 2003, an empirical analysis of knowledge management applications: Journal of knowledge Management, 7(3), 6-26. Montgomery, D. C., Peck, E. A., & Vining, G, G, 2012, introduction to linear regression analysis (Vol. 821): John Wiley & Sons. Preacher, K. J., Curran, P. J., & Bauer, D, J, 2006, computational tools for probing interactions in multiple linear regression, multilevel modeling, and latent curve analysis: Journal of Educational and Behavioral Statistics, 31(4), 437-448. Punch, K, F, 2013, introduction to social research: Quantitative and qualitative approaches, Sage. Raudenbush, S. W., & Bryk, A, S, 2002, hierarchical linear models: Applications and data analysis methods (Vol. 1), Sage. Sincich, T, 2003, a second course in statistics: regression analysis. Upper Saddle River, New Jersey: Pearson Education, Inc. Yuan, J. S., Reed, A., Chen, F., & Stewart, C, N, 2006, Statistical analysis of real-time PCR data: BMC bioinformatics, 7(1), 85. Read More