Quantitative Applications in Economics and Finance

The business manager's toolkit is filled with common tools to manage the financial performance of an organization, among them financial ratios, variance analyses, and budget projections. Capital-budgeting tools assume that decision makers have access to remarkably complete and reliable information, yet most strategic decisions must be made under conditions of great uncertainty (Courtney, Lovallo $ Clarke, 2013). A core function of financial oversight is having intimate knowledge of strategies to effectively manage changing environments, shrinking operating margins and increasing accountability. Managers are being held to specific performance goals by their organizations and stakeholders. It is readily apparent that quantitative analytic skills, or the ready access to these skills, are necessary to maintain a secure position in the industry. Historically, finance and economics experts' tools tended toward tried and true empirical models, long accepted in the business sector. One must ask, however: How successful would a manager be with access to the tools of measurement and inference that are necessary for market and business processes? Common trajectories can be uncovered from where they are hidden behind the coevolution of a large array of indicators (Du & Kamakura, 2012). This essay offers a snapshot of several economic and financial modeling techniques that organizations are using successfully to decrease variance and error. The models develop a hypothesis, apply analytic tools, and strengthen analysis and projection positions. Some industries are further ahead than others in the statistical analysis playing field and they are finding remarkable benefits to being on the cutting edge. It is not within the scope of this essay to instruct the reader in the science of statistics, rather to deliver the concept and its relevance to business and economics today.

Keywords Analysis; Econometrics; Financial Analysis; Forecasting; Hypothesis; Inference; Metrics; Models — Modeling Techniques; Predictive Behaviors; Probability; Process Control; Quantitative Analysis; Regression Analysis; Statistics; Underwriter; Variance

Economics > Quantitative Applications in Economics & Finance

Overview

Quantitative analysis, the process of applying mathematics to business, suggests that operational success can be enhanced beyond current systems, which look very much like an “applied intuition” method of management. The literature suggests that the setting of budgets, assessing changes in the marketplace, forecasting and strategy can be improved with the use of econometrics, the business of mathematically studying the underlying causes of varying outcomes. Variance is often poorly understood, and is always inefficient, wasteful and costly.

Applications

Case Study Number One: Initial Public Stock Offering

Who: The seller of common or preferred stock often enlists the services of an underwriter to determine pricing and timing of an Initial Public Stock Offering to investors. In the September, 2007 Journal of Financial Analysis, a study by Binay, Gatchev, and Pirinsky hypothesized that there is a predictable impact on IPO's sales related to the IPO underwriter's relationship with known investors (Binay et al., 2007). The group defined the event (the element to which a probability can be applied) in terms of a relational participation of the investor and the underwriter to determine the relationship's impact on future investment tendencies.

What: Sales of an initial public offering (IPO) present a risky investment option because the stock offering is usually brand new to the market and provides no historical performance with which the investor can compare. Uncertainty for the investor is the hallmark of an IPO; these offerings are commonly sold by newer companies in an early growth phase. To the casual observer, it would appear that there is much unpredictability in how well the stock will sell in its first days of going public. Given factors like market volatility, investor confidence, and risk perception, predictions are primarily speculative. Intuitively, the underwriter considers the question: Is there method, a model for predicting the relative success of the early offering; what investors should be approached early? It would appear very difficult to find any predictability in the answer to this question. History suggests that empirical experience drives the underwriters' assumptions. However, the following example just goes to show that where predictability seems unlikely, conditional probability analysis can provide a more robust prediction. Probability is simply the measure of the chance that an event will occur, given that another event has already taken place; the outcome of this type of study can support the analyst's prediction using inferential analysis.

How: Analysis is a process, and the deliverable is the recommendations for change and improvement. What Binay and others were searching for follows: "Despite the importance of underwriter-investor relationships, empirical research provides little evidence on the role of regular investors in the equity issue process in part because data on actual IPO allocations are proprietary and rarely disclosed by investment banks. [In this paper] we empirically examine the role of underwriter-investor relationships in the IPO process by examining institutional positions in IPO's as disclosed in quarterly 13F filings with the SEC. We construct a measure of relationship-based participation by institutional investors in IPO's as the difference of two probabilities-the probability that an institution investor participates in an IPO conditional on that investor's past participation in the same lead underwriters' IPO's, and the unconditional probability that an institutional investor participates in the IPO" (Binay et al., 2007). The writers hypothesized the following: "In this paper we investigate the role of regular IPO investors in the going-public process. IPO allocation practices often leave the impression that underwritings unfairly reward favorite clients and neglect other investors. Economic theory, however predicts that favoritism toward regular investors is an efficient way to extract information relevant for IPO firms” (Binay, 2007). The reader can examine in more detail, from the EBSCO Online Research Database (see bibliography) how the study was derived and administered, as well as the statistical strength of its findings.

Why: Refined and innovative analysis is costly: Time, experience and knowledge are required. The organization that makes the investment of enhancing its managers' skills or bringing in expertise is positioning itself to an advantage.

Where: The study on relational impact revealed that "We further find that regular institutional investors are more likely than casual investors to participate in IPO's with higher under pricing" (Binay, 2007). This is but a snapshot of quantitative analysis at work, validating and verifying predictive behaviors in the market clearly can prove instructive to underwriters and sellers. Absent statistically significant findings, the exercise may still be of value due to the inquiry and opportunity it brings to the investigators.

Case Study Number Two: Entrepreneurial Benefit of Storytelling

Who: Martens, Jennings, and Jennings published in the Academy of Management Journal a qualitative and quantitative study on the effects that storytelling has on potential investors' behavior response; for this research they used the statistical method of regression analysis. The reader will recall that some quantitative analyses involve developing a hypothesis, running the experiment, and gauging the effect of the independent variable on the outcome. For purposes of this essay, the quantitative work (not the qualitative component) of Martens et al. will be discussed. The group developed three distinct hypotheses surrounding resource acquisition, the premise of which is described by Jennings and Jennings in a subsequent paragraph.

Why: Applying logical (statistical) techniques to compare data can be a powerful means to support strategic initiatives in an organization. The adage that time is money supports businesses' efforts to improve their forecasting and predictions. Read further about the statistical study that identifies several dependent variables and measures their behavior in response to an independent action and the exciting outcomes of this design.

"Adopting a narrative approach to resource acquisition research, we examine the effects of storytelling on a firm's ability to secure capital. We argue that narratives help leverage resources by conveying a comprehensible identity for an entrepreneurial firm, elaborating the logic behind proposed means of exploiting opportunities and embedding entrepreneurial endeavors within broader discourses. Qualitative analyses of all 1996-2000 initial public offering prospectuses in three high-tech industries reveal how identity constructions, story elaboration, and contextual embedding are invoked within narratives. Our quantitative findings show how these aspects of an entrepreneurial narrative impact resource acquisition net of previously emphasized factors. To our knowledge, this paper offers the first systematic, large-sample test of the overarching claim that effective storytelling can facilitate external resource acquisitions. Integrating theory and research on the resources acquisition process with work by narrative scholars, we develop and test three arguments about how narratives (stories) help entrepreneurs attract capital" (Jennings, 2007).

How: The Martens, Jennings and Jennings group's three hypotheses were (Martens et al., 2007):

Hypothesis #1 the identity constructed for a firm in an entrepreneurial narrative has an influence on resource acquisition that is net of the influence of factual information about the firm's existing resource endowments.

Hypothesis #2 Elaborating the rationale behind a firm's intended actions in an entrepreneurial narrative has a positive but diminishing effect on the firm's resource acquisition ability.

Hypothesis #3 Embedding both contextually familiar and contextually unfamiliar elements in an entrepreneurial narrative has a positive but diminishing effect on resource acquisition ability.

What: The degrees of statistical strength in the results were varied, showing some correlation between the independent and dependent variables. Inference is determined by the strength of the findings; however, the experimenters' involvement in the process and any partial evidence of correlation is valuable nonetheless. The take-home message from this experiment is that further exploration is warranted. The authors close with the following words: "With respect to future research, we can envision a host of other intriguing directions for further work adopting a narrative approach to entrepreneurial phenomena. One of the most obvious is triggered by the dot.com scandals (Lowenstein, 2004): the need to investigate the nature, prevalence and effects of inauthentic entrepreneurial narratives" (Marten et al., 2007).

Case Study Number Three: Sawmill Industry

Who: Rudolf Beran from the University of California applied multivariate analysis (observation and analysis of more than one variable at a time) to the processes of board-cutting at a sawmill that was experiencing too many random errors in cut boards' thickness. As we discuss the test itself and the findings, the reader should recall that for statistically significant, meaningful tests and results, statisticians are a primary resource for advising on the actual study technique.

Why: What is the business case for embarking on a costly statistical analysis, might shareholders or owners ask? To what benefit, since aren't the majority of the boards at our sawmill good enough for processing? Beran offers the answer to this question in his introduction: "The increasing scarcity of top quality lumber in the western United States provides an economic incentive for strengthening process control in the sawmill industry" (Beran, 2007). The market is looking for quality and price all in one; and the market drives the demand.

What: What are the variables known to be contributing to error; is there more than one? Can it be identified? Among the multiple variables listed in the analysis were various stages through which the wood was processed and cut. Because the board product is subjected to so many processes throughout its preparation, the cumulative effect of all these processes would provide little insight into the variation; what is needed is a multivariate analysis.

How: How was the analysis performed? As a standard, baseline data were collected, analyzed for relations to errors and recorded. Identification of individual variable(s) impacting quality was the goal.

Where: Where were the opportunities found? Managing the saw's performance, identifying the impact of cumulative re-sawing, and setting specific size targets were key findings in the analysis. As the author states, "The main finding of this paper is that fitting a physically based statistical model to measured board thicknesses provides sound quantitative insight into the propagation of thickness error through lumber resawing. The model thereby enables effective quality control of individual saws, determination of target mean thickness by sawing code and setting priorities for sawmill improvements" (Beran, 2007). Discussion at the end of the Beran article can offer the reader a fuller understanding of not only the power but the process of statistical process improvement.

Case Study Number Four: Decision Analysis Tools for Marketing Investment

What: What is the best use of our organization's resources? If the demand for product is there, but not realized, should the company invest in more advertising? How will we know if it's worth the cost, if it's where our best return resides?

Why: No manager favors uncertainty in decision-making. Facts lay the groundwork for success, yet much decision-making continues to be made with limited data. Stanford University professor Chip Heath said, “The analogy I like is how we handle problems with memory. The solution isn't to focus harder on remembering; it's to use a system like a grocery-store list. We're now in a position to think about the decision-making equivalent of the grocery-store list (Heath, 2013).” There is truth to the theory that the efficacy of a system within a larger system inevitably impacts other sub-groups; the larger the organization, the more complexity and ambiguity. Understanding how related systems interact and their overall contribution is paramount in today's environment, especially with the reality of globalization and interconnectivity. Companies are looking for recommendations from their units that are defensible, while leaders are being held accountable for their actions. Errors in forecasting occur for many reasons, some of which include the challenge of additive error processes, conditional forecasting whose values are for an incorrect time period, inaccurate assumptions, or the silo phenomenon within organizations.

How: Some organizations are relying on simulation models, those that take a big picture view of where the organization or system is expected to be. Statistical forecasting has recognized value in capital and other resource budgeting, while more recently it has been employed as a primary tool for assessing return on marketing investments.

Who: How does econometrics bring value to decision-makers? Laura Bogomolny of the Canadian Business Journal writes, "Managers used to just throw money at their marketing department and hope for the best. Nowadays, they want proof that they are getting a return on their investment. There's an old saying familiar to anyone in the marketing biz: "Half the money I spend on advertising is wasted — I just don't know which half." However, a growing number of managers are convinced that they can unravel the mystery. Metrics, long embraced by almost all other business divisions, are finally grabbing hold in the marketing world — and many on the front lines report that innovative new methods for determining advertising's effectiveness are helping them get more bang for their buck. Tracking return on marketing investment, advocates argue, is now a real option, and as a result companies big and small are leaping at the chance to end the tradition of wasting huge sums on useless promotional campaigns before finding one that works (Bogomolny, 2004).

In the business sector, some believe that for far too long, forecasting (budgeting and projecting) has been mostly intuitive, speculative, and lacking numerical grounding. Bogomolny goes on to say, "Even CEOs of small, private companies are now at least open to the idea that there may be proof the money they pour into marketing isn't going down the drain.” And those willing to follow a marketing officer's advice on a "trust me" basis are a dwindling breed. Many execs have begun to demand that the efficiency and effectiveness of marketing expenses be carefully and scientifically tracked — and they are modifying budget decisions according to the results (Bogolmony, 2004).

Viewpoint

Contrasting View on the Utility of Quantitative Analysis

Few analysts would argue that the measurement of tangible data is easier than factoring a return on investment in something more nebulous, like marketing. Again from the Canadian Business Journal, Bogolmony says some Public Affairs and Marketing professionals are not sure that scientifically measuring their efficacy is playing fair.

"Amid the converts, however, skeptics still abound. Teasing out the impact of individual marketing projects requires advanced statistical modeling, and many question the accuracy of these techniques. Can you really capture the excitement created by a first-rate ad campaign with a numerical calculation? Some professional marketers remain unconvinced that you can accurately assess the influence of a specific initiative on an individual's purchase patterns. And many creative types — who insist that products survive on "buzz" and are loath to see advertising fall prey to corporate homogenization — are especially put off by the whole return-on-marketing-investment concept. Supporters counter that questioning marketing metrics is simply an excuse. As they see it, the naysayers are merely afraid to track how effective their work is because they fear it may highlight bad decisions made in the past.

Leonard Lodish, a professor of marketing at the Wharton school of business at the University of Pennsylvania and a leading expert in the field of return on marketing investment, certainly falls into that camp. He argues that “although there are some things that are notoriously variable in terms of measuring,” ROMI analyses are “pretty accurate.” Lodish says the real roadblock to getting companies on board is that measuring “pushes the egos of a lot of managers, because it sometimes will show you that your prior judgments didn't work very well. You have to be a very confident manager to do this kind of stuff," (Bogolmony, 2004).

The Shortage of Operational Managers with Quantitative Analytic Skills

In forward-moving sectors, employees are being “trained up" to enhance their analytic skills. In effect, bringing operational high performers to a higher level is the process of re-recruiting from within. Managers today are expected to manage quality processes such as those in lean manufacturing, continuous service and product improvement, innovation, solving problems by examining them, synthesizing measurable and defendable data, and leading development to solutions. Last but not least, growing professionals from within an organization means providing the development opportunities to produce the next generation of multi-talented financial, operational managers whose toolkit is much broader than that of their predecessors. Internal resources commonly provide the best investment a company has.

Conclusion

There should be no doubt in the reader's mind, after reviewing this essay, of the strength inherent in merging quantitative analytic skills into routine operational processes of business. Business degree programs teach quantitative financial analysis; these financial experts are excellent at what they do, and they can assess a process from afar with remarkable facility. However, the challenge for graduates is in gaining the real-world experience, the lack of which decreases the robust results possible from multi-talented managers facile in both econometrics and operations. Why are business managers not universally enjoying dramatic success in financial performance measures, given what we know? Sticking with business as usual comes at a cost; the savvy manager must understand that striking out ahead of the competition means challenging oneself and increasing his or her skills to the next level.

Terms & Concepts

Forecasting: The practice of demand-planning in everyday business; modeling the future state using statistical techniques.

Hypothesis: An uncertain proposal made to explain certain observations or facts that require further investigation in order to be verified.

Quantitative Analysis: An analysis technique used in finance, business and research to assess and understand process behavior with the use of mathematical and statistical modeling. It is a mathematical means of assessing the reality of a current system or process.

Statistical Modeling: Refers to the process of collection, analysis, interpretation and presentation of data. Statistical modeling assists decision-makers with more informed decisions based on mathematical analysis.

Return on Marketing Investment (ROMI): Represents the ratio of money gained or lost on a marketing investment relative to the amount of money invested.

Variance: The variance is one of several indices of variability that a statistician uses to characterize the distribution among the measures in a given data set.

Bibliography

Beran, R. (2006). Statistical modeling for process control in the sawmill industry. Applied Stochastic Models in Business & Industry, 22(5/6), 459-481. Retrieved November 25, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=23527519&site=ehost-live

Binay, M., Gatchev, V., & Pirinsky, C. (2007). The role of underwriter-investor relationships in the IPO process. Journal of Financial & Quantitative Analysis, 42, 785-809. Retrieved November 23, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=26847970&site=ehost-live

Bogomolny, L. (2004). Do you measure up? Canadian Business, 77, 93-103. Retrieved November 25, 2007, from EBSCO Online Database Academic Search Premier. http://search.ebscohost.com/login.aspx?direct=true&db=aph&AN=14694358&site=ehost-live

Courtney, H., Lovallo, D., & Clarke, C. (2013). Deciding how to decide. (cover story). Harvard Business Review, 91, 62-70. Retrieved November 21, 2013, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=91571432&site=ehost-live

Du, R., & Kamakura, W. (2012). Quantitative trendspotting. Journal of Marketing Research (JMR), 49, 514-536. Retrieved November 21, 2013, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=78191680&site=ehost-live

Heath, C., & Sibony, O. (2013). Making great decisions. McKinsey Quarterly,, 66-76. Retrieved November 22, 2013, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=87315652&site=ehost-live

Martens, M., Jennings, J., & Jennings, P. (2007). Do the stories they tell get them the money they need? The role of entrepreneurial narratives in resource acquisition. Academy of Management Journal, 50, 1107-1132. Retrieved November 23, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=27169488&site=ehost-live

Recommended Readings

McColloch, A.C. (1986). Institute for quantitative research in finance. Financial Analysts Journal, 42, 10-11. Retrieved November 29, 2007, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=6979848&site=bsi-live

Terui, N., & Imano, Y. (2005). Forecasting model with asymmetric market response and its application to pricing of consumer package goods. Applied Stochastic Models in Business & Industry, 21, 541-560. Retrieved November 25, 2007, from EBSCO Online Database Business Source Premier. http://search.ebscohost.com/login.aspx?direct=true&db=buh&AN=19121750&site=ehost-live

Sreenivasan, R. (2004). Applied quantitative finance: Theory and computational tools. Journal of the Royal Statistical Society, 167, 191-192. Retrieved November 29, 2007, from EBSCO Online Database Business Source Complete. http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=11713725&site=bsi-live