Review of

1-18-10_mhtm_COST FUNCTION MODEL ASSUMPTIONS NARRATIVE AND VALIDITY

COST FUNCTION MODEL ASSUMPTIONS NARRATIVE AND VALIDITY- A LINEAR AND POLYNOMIAL REGRESSION GOODNESS OF FIT FOR FORENSIC ACCOUNTING, EXPERT WITNESS TESTIMONY & COMPUTER LITIGATION SUPPORT

Abstract

This paper deals with forensic studies in accounting a= nd Business. It focuses on the integration of academic and educational issues of fraud and forensic applications of management accounting theories of the “relevant range” and validity of the cost functions goodness of fit assumptions. The paper tracks the improvement in the narrative of explaining the cost function in educational context of fraud and forensic accounting. Using pedagogical techniques, such= as case presentations, empirical studies, mock trials, as well as the practice= of expert witness testimony and computer litigation support, in both the higher education and the continuing professional education environments. This paper present technique= s and technologies of investigation involving real-world business problems of forecasting cost as a function of revenue using computers for litigation support. It describes collabo= ration between practitioners and academics in deploying new forensic methods, in theoretical, empirical and experimental approaches.

Literature Review

Forecasted Annual Reports and the Role of the Cost Fun= ction and the Relevant Range

Forecasted Annual Reports has been researched extensively in the literature, across multiple countries. "The Construction of an International Investors' Perception Model for Corporate Published Forecasted Financial Reports for the United States, Great Britain, and New Zealand,&qu= ot; (Rushinek, A., Rushinek, S., Chang, L. and Most, K., 1983)[1]= [2]= for Managerial And Decision Economics (MDE), is one such example. In the center of such studies are = the cost function and the “Relevant Range.” Once the accountants forecast the = Sales, they can forecast the costs, after they have calculated the cost function.<= span style=3D'mso-spacerun:yes'> For this purpose, they have to fig= ure out what is the cost function, and that depends on whether the activity lev= el is within the relevant range, in which traditionally accountants assume that the total cost is a Total Mixed Cost (TMC) function equal to a Total Variab= le Cost (TVC) plus a Total Fixed Cost. Most forecasters use some variation of a regression analysis for such purposes.

Another common application of forecasting financial performance is "Using Financial Ratios to Predict Insolvency." (Rushinek and Rushinek, 1987)[3]= In such a case, the idea is to forecasts the cost, the cash outflows that t= hey produce, and the point at which the net cash outflows, exceed the enterprise cash reserves, driving it into bankruptcy.= The common methodology of such an analysis is a discriminant analysi= s, which is similar to the regression analysis, in that the user has to decide whether the function is linear or not linear. Most users assume that it is linea= r and proceed without additional testing. The problem is that the linearity assumption may not be valid, and therefore the forecasts may not be as accurate, due to the low goodness of fit between the actual data and the mathematical function that describes the data.

Using Experts for Detecting & Litigating Forecast Errors and Frauds for Oil and Gas Industry

The reasons for some of such bankruptcies i= s due to the deficiencies in "Evaluating Asset Safeguarding and Data Integrity,” especially “in the EDP Audit" process. (Rushin= ek and Rushinek, 1989)[4] For that purpose companies have been "Using Experts for Detecting & Litigating Computer Crime", in Managerial Auditing application. (Rushi= nek and Rushinek, 1993)[5] Such "Forensic Audits = of Stock Market Forecasts” frequently rely on experts in a certain indus= try and even in a specific company such as “Texaco & The International Oil and Gas Industry,” looking at “Financial Models” for specific cases of “Online Broker Litigation." (Rushinek and Rushinek, 2003)[6] For such cases, an expert may have= to rely on consolidated industry financial reports, such as the reports produc= ed by the US government.

For example in "Tax Fraud Forecastin= g for the Petroleum Refining Industry,” experts compare the “Financial Ratios” of individual companies to their industry standards. They use “Reversed Accounting Theory Applied to” a specific company, such as “Shell Oil"= to figure out what their taxes should have been based on industry standards. Then, one can determine whether th= e differences between what the taxes should have been and what taxes have been reported, appears to raise the red flags of tax fraud. (Rushinek and Rushinek, 1997)[7]=

Some times “Managerial auditors mixed cost forecasting assumption departure error” lead to “litigation and professional liability,” e= ven if it is not fraud but lack of due diligence. Such errors may increase the “risk” of material misstatements of financial reports. Again the reduction of such errors= and their risk depends on using consolidated industry data, in this case it is “the electronics industry.” (Rushinek and Rushinek, 1997)[8]= For example, auditors may assume t= hat the cost function is linear when in fact it is not linear. Therefore, the cost estimate based= on forecasting errors are off, leading to enormous cost over-run, litigations,= and other additional difficulties. For this reason it is important to raise the goodness of fit on the cost functi= ons and the data that they represent.

For this reason researchers have been “Rating and Ranking Best Practices” in certain companies and industries, such as the “British Petroleum Oil Company and the Oil & Gas Industry.” That is especially important in the “Sales to Cash Forecasting” areas, “Univariate Regression Trend Analysis and Computer Modeling” to correctly estimate the future cash needs and avoid bankruptcies. Especially the volatile Oil and Energy i= ndustries. (Rushinek and Rushinek, 1997)[9]=

As experts repeat their wo= rk, running a similar computer model over and over again, they tend to automate such an analysis. A case of &= #8221;Litigation in Pennzoil and the Petroleum Refining Industry,” uses “A Finan= cial Ratio Automated Peer Review Analysis of Variance” using “Intern= et Software Development” to automate the analysis of the “OIL, GAS & ENERGY” industry. (Rushinek and Rushinek, 1998)[10]

Internet Software Development of Expert Sys= tems (ES)

A regression analysis, is basically a correlation betw= een the dependant variable and the independent variable. It is similar to ”Correlatin= g Investors’ Preferences to Stock Characteristics.” That is “An Algorithm for In= ternet Software Decision Support System Development in the Crude Oil & Refining Industry.” (Rushinek and Rushinek, 1999)[11] The computer program tends to make decisions in a much more structured, and methodical way than experts do. The idea is to emulate, or simulat= e a human expert as much as possible using a computer system in addition to a h= uman expert. Similar methodology c= an be used for ”Oil & Gas Market, Sector & Company Relative Strength Congruity Trading,” assisting an expert stock trader using “Internet Automation Approach” to speed his trades and make the stock market forecasts more accurate and less risky, when applied to a cert= ain company and a certain industry, such as “Chevron Corporation’s Stocks” and the “OIL, GAS & ENERGY” industry. (Rushinek and Rushinek, 2000)[12]

Forensic experts a= nd accounting professors can use “Internet Fraud Auditing” methodo= logy in teaching by applying “A Simulated Health Care Industry” case study to teach the jury and the students the concepts of calculating the abnormal profits that result from phony sales and/or understated costs. An expert, an ES, or a professor c= an regress an individual health care providing company on consolidated industry financial reports to identify red flags of fraud, and help “un-cook t= he books.” (Rushinek and Rushinek, 2000)[13]

<= /span>

“The Role of= the Forensic Accountant in Calculating the Damages Using the ‘But-if̵= 7; Analysis” depends on the validity of several assumptions about the co= st function. (Rushinek and Rushinek, 2001)[14] Such assumptions include the clear definition and the position within the “Relevant Range.” Furthermore, another assumption deals with the linearity of the regression model. The advantages of linearity include but are not limited to the mathematical and statistical simplicity. Since, the linear function is simp= ler than the nonlinear functions, it is easier to program it into the Expert Sy= stem (ES), and it is much easier for the user to understand it, as well as easier for an expert to explain it to a jury, or for a instructor to teach it to students. To program it into = an ES, the expert has to provide the ES with decision rules, to decide whether the cost function is linear or not. This paper provide a methodology of doing just that, in a way that is very intuitive, easy to understand, easy to teach, and very easy to demonst= rate manually in a classroom or courtroom environment (By check marking a couple= of boxes in the regression wizard in Microsoft’s Excel or similar spreadsheet programs). <= /o:p>

<= /span>

This method simply calculates both functions, compares the R-Square differences in the Goodnes= s of Fit, and selects the linear function, if the gains of using the nonlinear functions are small enough to be negligent (less than .1% for example). To further test the validity of the function, the expert and/or the expert system applies it not only to the company in question but also to the industry. The inclusion of the industry illu= strates that such a cost function is valid not only in the company in question, but= in the entire industry. This is particularly important for forensic situation where the company data is not available, unreliable, incomplete, or simply does not contain enough observations to conduct a reliable regression analysis. Therefore, in the last case that we reviewed, the expert had to rely on such an analysis to calculate the damag= es due to a day trader from his broker, since the broker stopped the trade cau= sing the trader significant losses.

<= /span>

Continuing Professional Education (CPE) Teaching Accou= nting Professors about Model Selection

Continuing Professional Education (CPE) courses play a= very prominent role in testing new theories about regression methods that should= be simpler to use and understand, concerning selection of the model with the b= est goodness of fit, which is also the simplest. Such CPE about Model Selection hav= e been focusing recently about integration of new technologies into forensic accounting expert witness testimony and computer litigation support. This courses demonstrate to accoun= ting professors, expert witnesses, and Lawyers as Social Media Us= ers, using such networking sites such as LinkedIn® and Facebook®, Myspace®, Blogs, and Wikis, while dealing with “Sarba= nes Oxley Microsoft Accelerator Forensic Accounting, Expert Witness Testimony, & Computer Litigation Support” (Rushinek and Rushinek, 2005)[15] Teaching s= uch CPE workshops for organizations such as the American Accounting Association (AAA) shows that it is an up and coming new social media networ= king technology. Since it is relat= ively new, it is hard to evaluate it. Therefore, researchers develop new assessment techniques to evaluate= it. (Rushinek and Rushinek, 2007)[16] Such "Assessment” of CPE Video Podcasts and Blogs requires some = "Hands-on" experience to learn how to “Consult, Publish & Teach Forensic Auditing, Expert Witness Testimony” using new software such as Microsoft® Office Accounting” for Computer Litigation Support. It demonstrates how to transcribe = the lecture or testimony and automatically publish the narrative, the video, the supporting documents on a peer-reviewed online journal, such as the IT (Information Technology) Journal.[17]

Due to the popularity of internet video forensic exper= ts are learning how to deploy it in their witness testimony, in the courtroom and = in the classroom (Rushinek and Rushinek, 2= 008) In the case of brokers’ misconduct; the surveillance video can show h= ow the broker approaches the day trader pulling out his computers power plug in middle of the trade. This is = an example of “How to Use New Media and Technology to Repurpose Classroom Instruction for the Accounting and Business Educator”, in an “I= nternational Fraud and Forensic Accounting Education Conference,” in Las Vegas, Ne= vada, where the most sophisticated repurposed forensic surveillance systems play a prominent role in auditing, expert witness testimony and litigation support= . As the expert explains the concept= of the regression goodness of fit, on can observe the students, and/or the jur= y in real time and assess their reaction.

<= /span>

Another way to measure the students and jury reaction to a new method is by a mock trial. (Rushinek and Rushinek, 2008, = 2009)[18] "Mock Trial & Forensic Faculty” can deploy the “WWW Marketplace, Jury, Judge & Attorneys” exposing them to the new me= thod in the "Courtroom & Classroom” using new “Instructional Technology” to evaluate their reaction. So far such reaction has been positive, so the authors proceed with this research direction.

Common Confusions in Manag= ement Accounting Textbooks and Accountants Communication

Mixing Up Total and Unit Costs and Applying= the KISS & GAAP principle

Total Mixed Cost =3D Fixed Cost Element+ Va= riable Cost Element

The typical Total Mixed Cost (TMC) equation frequently omits important terms and confuses the differences between total= and unit costs (Warner and Jones, 2009 p 167). A more consistent and accura= te statement of the TMC equation will be the following:

Total Mixed Cost (TMC) =3D Total Fixed Cost= (TFC) + Total Variable Cost (TVC)

Adding the prefix “Total” in fr= ont and dropping the suffix “Component” at the end improve the distinguish more clearly between the concepts of Total Costs (TC) and Unit = Cost (UC). Furthermore, it may be = more consistent to stick to one term such as components or element, rather than adding too many terms to discuss the same thing, as the Generally Accepted Accounting (GAAP) principle of consistency suggests. Likewise, it may be less confusing= to the reader and the students. = If the work “components” works, why not use the “KISS principle = and Keep It Simple Stupid,” or “if it works, don’t fix it.= 221;

Abbreviating the variable names as soon as possible maybe another improvement as it makes it more familiar to the students, when they have to deal with so many abbreviations later on, as the mathematical derivations become more complicated.

Adding Currencies as Activity Measures to t= he Physical Units of Activity Measurement

Typical authors of Management Accounting te= xtbook tend to describe the cost function as a “Scatter Graph for … Co= st and Miles Driven” or a similar physical unit of activity measurement. Its advantage is= that it is simpler but tends to be limited to situations with only a single acti= vity to measure cost. Most compani= es, most costs, and most activities involve more than a single physical unit. Therefore, it is important to supp= lement the examples of currencies, such as Dollars of Sales, as an activity unit of measurement, to physical units, such as Miles Driven, or Packages Delivered= . As one consolidates multiple activities, for example for a package delivery company, combining Miles Driven and Packages Delivered, on can use the Doll= ar Values of the revenues to measure the activities, in addition to the physic= al values. Therefore, the present paper focuses on such supplemental example of decomposing the components of= TMC as a function of Dollars of Sales. That is even more important for consolidated financial statements, w= hen multiple subsidiaries have totally different products and lines of business, such as IBM Corporation that sells computers and consulting services. This is even more important when a= pplying the same process to an entire industry, such as the manufacturing industrie= s, or the consolidated “U.S. Manufacturing, Mining, And Wholesale Trade Corporations.” (US Census Bu= reau, Dec. 2010)

A lot of these confusions, seem too trivial= to accountants, while they are communicating among themselves, the trouble is = that many times, accountants are communicating with the public of jurors, as forensic expert witnesses, dealing with the public, who is trying to read t= he financial reports, and make some sense out of them. At such times, it is quite difficu= lt for the public to figure out what and where is the relevant range, and why it i= s so relevant to cost forecasting and to a forensic “But If” type of analysis, as well as to teaching management accounting and similar topics.<= o:p>

Data Collection, Access, Reliability, and Sampling Errors

The US Census Bureau provided= the data in “The quarterly publication, Quarterly Financial Report (Q= FR) for Manufacturing, Mining, and Trade Corporations – Third quarter 2009 (QFR/09-Q3), December 14, 2009, US Census Bureau Report. (US Census Bureau, 2009)

<= /p>

The data describes the “= ;After-Tax Profits and Sales, Third Quarter 2009 - Seasonally Adjusted,” using A= RIMA (Auto Regressive Integrated Moving Average) method. It focuses on Manufacturing Corpor= ations. U.S. manufacturing corporations= 217; seasonally adjusted after-tax profits in the third quarter of 2009 totaled $88.5 billion, up $36.4 (±0.5) billion from the after-tax profits of $52.1 billion recorded in the second quarter of 2009, but down $29.3 (±1.6) billion from the $117.8 billion after-tax profits recorded in= the third quarter of 2008. &nb= sp;

While the data describes both= Durable Goods Manufacturers, and none-Durable.&nbs= p; Our model focuses on none durable manufacturers, such as the Oil and= Gas industries, due to the importance of forecasting correctly and the high volatility of the industry as the political scene and the financial markets fluctuate.

Mining Corporations are part of the Oil and Gas industries an= d may need some special attention. “Third quarter 2009 unadjusted after-tax profits for mining corporations with assets of $50 million and over totaled $4.7 billion, down $26.7 (±0.1) billion from the $31.4 billion after-tax profits record= ed in the third quarter of 2008. Compared with second quarter 2009 after-tax profits of $2.3 billion, third quarter 2009 after-tax profits were up $2.4 (±0.1) billion.” &n= bsp;

Access to QFR Data

“This press release can= be viewed today in portable document format (.pdf) at our QFR internet website= http://www.census.gov/econ/qf= r/index.html. Data in the release are drawn from a more comprehensive data set published in the Quarterly Financial Report for Manufacturing, Mining, and Trade Corporation= s - Third quarter 2009 (QFR/09-Q3). This quarterly publication is available for downloading at our QFR internet website http://www.census.gov/econ/qfr/inde= x.html. The quarterly publication includes summary statements of income and retained earnings, balance sheets, and related financial and operating ratios for manufacturing, mining, and trade corporations. Data presented are classified by industry and asset size. &= nbsp;

Reliability of the Estimates =



Data in this press release are based on quarterly financial reports from approximately 9,000 U.S. corporations. The data are estimated from a sample survey and are subject to sampling and nonsampling errors. &nbs= p;



Sampling error occurs because= only a subset of the entire population is measured. Estimates of sampling error can be computed based on the sample and used to construct confidence intervals aro= und the estimates. Statements of = change appearing in this report include 90 percent confidence intervals. Thus, a statement in the report su= ch as “up $2.5 (± 0.2) billion” estimates the interval (+$2.3 billion to +$2.7 billion) within which the actual value is likely to fall i= n 90 percent of samples of the same size and design, drawn from the same population. If the range of estimated change contains zero (0), then it is uncertain whether there is an increase or decrease; that is, the change is not statistically different fr= om zero (0). For any comparison cited in the text without a confidence interva= l, the change is statistically significant at the 90 percent confidence level.=          =



Nonsampling error encompasses= all other factors that contribute to the total error of a survey, including response errors, nonresponse, and coverage errors. Although no direct measu= res of nonsampling error are available, precautionary steps were taken in all phases of the collection, processing and tabulation of the data in an effor= t to minimize their influence. The quarterly publication, Quarterly Financial Report for Manufacturing, Mining, and Trade Corporations – Third quarter 2009 (QFR/09-Q3), includes more detailed explanations of nonsampling and sampling error, and additional measures of sampling variability.”&nb= sp;

Visual Data Inspection of the Lowest and Hi= ghest Bounds of the Relevant Range

The data set describes the following variab= les: Years, Sales (x), and Expenses (y). = For example at the lowest bound of the Relevant Range is year 1998, with Sales = of $444,850 (x), and Expenses (y) of $415,375, in millions of Dollars= . At the highest bound of the Releva= nt Range is Year 2008, with Sales of $923,334, and Expenses of $848,934. The TMC chart graphs the Expenses = or Costs y, the dependant predicted variable, as a function of Activity or Sal= es, x, the independent or predictor variable.&= nbsp; As one can observe from the chart and the data, a company with about $600,000,000,000 would be in the middle of the relevant range, where the validity of the linearity assumption is almost 100%. By observing the Poly and the line= ar trend lines, one can see that they are perfectly aligned. In contrast, as one moves up, furt= her into the upper bound over the Sales label of $800,000 the Poly function dep= arts from the Linear function curve, that is where the goodness of fit drops to 99.7%.

Before getting into the regression analysis= and the statistical and mathematical discussion, it is important for an Expert System (ES) as well as an Expert, especially a forensic accountant providing expert witness testimony to the jury, to explain the data in a jargon free manner and in layman terms that appeal to the common sense. Likewise, an instructor teaching elementary management accounting should describe such issues. This is when a simple 2 dimensional scatter graph and the trend lines can be very visually appealing, to preview the regression analysis and the more mathematical discussion. As one looks at t= he scatter graph, one can see that both trend lines increase from left to righ= t, as the volume of Sales grows, and the Expenses grow proportionately, at a constant slope, UVC, which means that we are inside the relevant range, whe= re the Relevant Range assumptions of functional linearity are valid.

This policy provides some kind on an intern= al control that enables the observer to quickly, by inspection of the chart, t= rap errors and correct them at an early stage before additional work and relian= ce has been erroneously placed on the data.&n= bsp; Such inspection should verify that the cost function shows that the total variable costs (TVC) increase as the activities increase ($Sales), at= a constant rate.

Regression Analysis – Focusing on the Concepts rather than the Translation

In a typical “Regression AnalysisR= 21; authors describe the TMC function again this time decomposing the Total Variable Cost (TMC) into its components, but still failing to name it accurately as TMC and TFC, instead they keep omitting the prefix “Total” from Fixed Cost and Variable Cost, as follows:

Total Cost (omitting Mixed) =3D Fixed Cost = + (Unit Variable Cost x Activity Level).

A clearer, step wise more explicit alternat= ive may be:

Total Mixed Cost (TMC) =3D Total Fixed Cost= (TFC) + Total Variable Cost (TVC) and

Total Variable Cost (TVC) =3D Unit Variable= Cost x Activity Level, therefore:

Total Mixed Cost (TMC) =3D Total Fixed Cost= (TFC) + Unit Variable Cost x Activity Level or

Y=3D a + bx,

Where:

y =3D TMC, “the dependent” pred= icted variable, “that changes in response to a change in x.”

a =3D TFC The 1^ST coefficient of= the independent predictor variable, “The y intercept, or the amount of y = when x is zero.”

b =3D UVC The 2^nd coefficient of= the “independent predictor variable.”

x =3D “The independent” predict= or “variable.”

Including the TMC, TFC, and UVC, facilitate= s the reader’s transitions from the accounting terminology to the statistic= al mathematical terminology. Thi= s way the reader may avoid making the leap of faith in the translation all by themselves, and can focus on understanding the concept rather than translat= ing the accounting terminology to the statistical mathematical terminology. The translation efforts compounded= by the new terms, such as least-squares method and linear regression, forward & backward forecast, trend line, in addition to regression analysis may= be “the straw that breaks the camel’s back.”

The Linearity Assumption of the Relevant Ra= nge and the Regression Analysis Empirically Tested

Regarding the “Relevant Range” forensic experts, providing expert witness testimony and computer litigation support, accounting authors, accountants and textbooks assume implicitly th= at “variable cost is constant per unit within the relevant range.”= (Werner, Jones, 2009, page 161) They make the linearity assumption, without providing any empirical evident to that effect or a method to test whether the TMC function is linear or not. Th= is paper provides such evidence along with an automated method that enables the replacement of a linear function by a more fitting function, such as a polynomial function. Instead = of just assuming that we should always use a linear regression, this paper off= ers a novel empirical automated proofing methodology, along with the traditional statistical hypothesis testing methodology.

Textbooks typically state that “decis= ion makers usually assume activity levels will be within a company’s rele= vant range.” (Werner, Jones, 2009, page 161) Therefore, they also assume that t= he TMC function is linear, without a simple empirical, speedy automated methodolog= y to prove and support such a decision. This paper automatically replaces the linear function, with a more appropriate function, such as a polynomial function, in the event of a significant departure from linearity. Otherwise, if the linearity assumption holds, it provides empirical = evidence to such an assumption that will promote accuracy of forecast, as well as reliability of an expert system, or an expert witness, business decision ma= ker that has to forecast or explain cost as a function of activity.<= /span>

R² Goodness of Fit Difference between = the Polynomial of the 3rd Degree and the Linear Model

Table 1

R² Goodness of Fit Differences Between the Polynomial of the 3rd Degree and the Linear Model

Column L

M

N

O

P

Q

Line 66

Model

Regression

R²

y

Line 67

1

Poly2

9= 9.9%

y =3D 1E-07x² + 0.701x + 80317<= /p>

Line 68

2

Poly3

9= 9.9%

y =3D 1E-08x³ - 0.005x² + 1372.x -1E+0= 9

Line 69

3

Linear

9= 9.8%

y =3D 0.885x + 22809

Line 70

Difference

0.1%

=3DN68-N70

Line 71

TMC =3D 0.885*Sales +$ 22,810

Table 1 describes the “R² Goodne= ss of Fit Differences Between the Polynomial of the 3rd Degree and the Linear Mod= el of a univariate regression in an excel spreadsheet in Column L, M… an= d line 66, starting with the Model, Regression, R², y function of a trend lin= e, and in Line 67, model number 1, Polynomial of the 2^nd degree (Po= ly2) with its respective pairs of R-Square, = ; 99.9%, and the trend line function of y =3D 1E-07x2 + 0.701x + 80317. The table en= ds with Line 70, the Difference  &nb= sp;      0.1%, between the Linear and the Poly3 function R-Squared, goodness of fit. This difference supports the linea= rity assumption, showing that the gain in goodness of fit due to the departure f= rom linearity is lower than .1% for a goodness of fit of 99.9%. Therefore, the expert as well as t= he expert system can proceed with linear model simplicity without significant = loss in forecasting accuracy and goodness of fit.

Rule Automation of the Selection of a Total= Fixed Cost (TMC) Trend Line Regression Model

The experts set up a predetermined rule that states the condition of selecting a certain default trend line regression model. The user can customize= the values of the decision rule according the minimal needed reliability, goodn= ess of fit, and other circumstances. Such rule may specify that if the R-Square between any pair of forecasting trend line methods difference is .1% or low= er the Expert System (ES) or the expert (author, accountant, etc) should defau= lt to the simplest linear model. Following is an example of a couple of rules related to the differen= ce between the R-Squares of the Linear and the Poly model, and their Hypothesis Testing or the value of the R-Square:

If-

Rule 1: If the Linear R-Square minus the Polynomial R-Square < .01 %, and

Rule 2: The Linear R-Square > .9 or 90%,=

Then default to the-

Action: Linear trend model for the Total Mi= xed Cost TMC =3D TFC + UVC*Sales.

Otherwise, select the model that maximizes = the R-Square but is more complicated. The table ends with the excel formula of =3DN68-N70, followed by Lin= e 71, and the cost function of TMC =3D 0.885*Sales + 22,810, of the linear trend = line, restated in the accounting rather than the statistical mathematical terminology.     &= nbsp;

Raising the level of such simplicity, as in= the linear model compared to the polynomial models is especially important when dealing with students in an elementary course, business executive who may n= ot have studied accounting, math, or statistics, or with a lay court room (Jud= ges, Lawyers, and especially jurors)  =           Maximizing The Simplicity Among Polynomial, Exponential, Logarithmic, Moving Average, = And Power Model, Without Compromising The Goodness Of Fit Of The Forecasting Tr= end Line

The typical management accounting textbook describes how to “forecast backward and Forecast Forward,” it instructs the reader “to click on the Options tab,” but fails to mention to check mark the box for “R-Square,” the goodness of f= it of the trend line and the forecasting model. This paper illustrates the importa= nce of such a measure, and how easily it can be added to a manual regression calcu= lation. Moreover, for further simplicity, = the user can automate it and program it using VBA (Visual BASIC (Beginners All Purpose Symbolic Interactive Code) for Application) programming language bu= ilt into excel, and other spreadsheets, as well as more advanced programming languages for ES. =

As the ES program automates the calculation= of the goodness of differences among a dozen of models, such as Exponential, Logarithmic, Moving Average, and Power model, the ES will proceed to pick t= he simplest and the best fit, the linear model in the present case, without ex= posing the author, business executive, reader, courtroom to the mathematical and statistical complexities, yet intuitively enable them to understand and correctly assess the goodness of fit and its importance. Using the earlier narrative, the E= S will proceed to generate the appropriate consistent and clear text, to facilitate the understanding of the concept for a novice student and a business execut= ive, as well as making it easier for an instructor to teach.

Summary Conclusions and Implications

Summary

In summary, this paper reviewed the narrati= ve that describes cost behavior and related “does” and “don’ts.” F= or example, the paper reviews the tendencies of accountants to mix inconsistent terminology, for a single concept, unnecessarily confusing the reader, such= as cost components and elements. Inconsistently adding or dropping the prefix “Total” from Total Mixed, Fixed or Variable Costs, without any apparent reason. For example = using Fixed Cost to describe the Total Fixed, while using Variable Cost for Unit Variable Cost. So, it may not= be perfectly clear whether they are talking about Total Fixed, Mixed, or Varia= ble cost, or maybe about Unit Fixed, Mixed, or Variable Cost. Another issue is making assumption= about the relevant range, linearity, and trend line model selection, with complete disregard to the goodness of fit, or any empirical evidence to support such= an assertion, such as the R-Square. Another issue, maybe the selection of activity measurement that is limited to the number of physical units, excluding the possibility of using Sales in Dollars or other currencies, etc.. Likewise, the paper suggested bett= er narratives, and automated evidence producing methods to support such assert= ions and assumptions or depart from such assumption as the circumstances dictate. It provided empirical evidence for the case of the superior simplicity of a linear model over a polynomial model, and its related rules, as well as their decision paramete= rs.

The conclusion is that for non-durable manufacturing the linear function maybe the best within the relevant range,= of annual report of Total Mixed Cost as a function of Total Variable and Fixed Costs. Likewise, this paper s= uggest that in addition to considering the data of an individual product, company,= or consolidated financial statements, an entire industry trend line may further enhance the model and control for early detection of errors, outliners, or unusual devi= ants from the national industry standards. It concludes that automation may improve the accuracy without over burdening the user with having to understand fully complicated computational methods, statistical analysis, and mathematical constructs. Yet making speedy assumptions that= are supported by the evidence, promoting simplicity without compromising accura= cy and validity.

The implications of this study are that a l= ot more work may be done in testing such conclusions over longer period of tim= es, various levels of aggregation all the way to an individual product. Likewise, applying the same concep= t to varying combinations of dependent and independent variables. For example, testing the same methodology for Sales Forecasting as a function of the Julian date, prior to calculating the costs as a function of sales, due to the fact that Sales Fo= recast normally precedes cost forecasts.

<= /p>

//////////////////////////////////////////////////////////= /////////////////////////////////////

References:

Rushinek, A., Rushinek, S., Chang, L. and M= ost, K. "The Construction of an International Investors' Perception M= odel for Corporate Published Forecasted Financial Reports for the United States, Great Britain, and New Zealand," MANAGERIAL AND DECISION ECONOMICS (MD= E), THE INTERNATIONAL JOURNAL OF RESEARCH AND PROGRESS IN MANAGEMENT ECONOMICS,= Vol. 4, No. 4, 258‑265, December 1983.

Rushinek, A. and Rushinek, S. "Evaluating Asset Safeguarding and Data Integrity in the EDP Audit&quo= t;, MANAGERIAL AUDITING JOURNAL, Vol. 4, No. 2, 3‑6, 1989.

http://www.emeraldinsight.com/Insight/viewC= ontentItem.do;jsessionid=3D98F6451F5BD9FAD182CCEE6888D98CA7?contentType=3DA= rticle&contentId=3D868055

Rushinek, A. and Rushinek, S. "Using E= xperts for Detecting & Litigating Computer Crime", MANAGERIAL AUDITING JOURNAL, Vol. 8, No. 7, November, 1993, 19-21. http://www.emeraldinsight.com/Insight/viewC= ontentItem.do?contentType=3DArticle&hdAction=3Dlnkpdf&contentId=3D8= 68184

Article URL: http://www.emeraldinsight.com/10.1108/02686909310046862

Rushinek, A. and Rushinek, S. "Forensic Audits of Stock Market Forecasts For Tex= aco & The International Oil and Gas Industry & Financial Models for Onl= ine Broker Litigation", OIL, GAS, & ENERGY QUARTERLY, Vol. 51, No. 4, June, 2003, 659-678.

Rushinek, A. and Rushinek, S. "Tax Fra= ud Forecasting For the Petroleum Refining Industry: Financial Ratios & Reversed Accounting Theory Applied to Shell Oil", OIL AND GAS TAX QUARTERLY, March, 1997.

Rushinek, A. and Rushinek, S. “Managerial auditors mixed cost forecasting assumption departure error estimates for litigation and professional liability risk reduction: the electronics industry”, MANAGERIAL AUDITING JOURNAL, Vol.12 No. 6, 199= 7.

http://www.emeraldinsight.com/Insight/viewC= ontentItem.do?contentType=3DArticle&hdAction=3Dlnkpdf&contentId=3D8= 68360

http://www.emeraldinsight.com/10.1108/02686= 909710180643

Rushinek, A. and Rushinek, S.”Rating = and Ranking Best Practices of the British Petroleum Oil Company and the Oil &am= p; Gas Industry: Sales to Cash Forecasting Univariate Regression Trend Analysis and Computer Modeling”, OIL AND ENERGY QUARTERLY, December, 1997.

Rushinek, A. and Rushinek, S.”Litigation in Pennzoil and the Petroleum Refining Industry: A Financial Ratio Automated Peer Review Analysis of Variance Inte= rnet Software Development” OIL, GAS & ENERGY QUARTERLY, Vol.47, No.1, September, 1998, 17-32.

Rushinek, A. and Rushinek, S.”Correlating Investors’ Preferences to Stock Characteristics: An Algorithm for Internet Software Decision Support System Development in the Crude Oil & Refining Industry” OIL, GAS & ENERGY QUARTERLY, Vol. 48, No.1, September, 1999, 87-101.

Rushinek, A. and Rushinek, S. ”Oil & Gas Market, Sector & Company Relat= ive Strength Congruity Trading Theory (MSCRSSCTT): An Internet Automation Appro= ach Applied to Chevron Corporation’s Stocks” OIL, GAS & ENERGY QUARTERLY, Vol. 48, No.4, June, 2000, 671-682.

Rushinek, A. and Rushinek, S. Internet Fraud Auditing: A Simulated Health Care Industry Case Study, JOURNAL OF FORENSIC = ACCOUNTING, Vol. 1, No. 1, 125-234, 2000. http://www.rtedwards.com/journals/JFA/content= s.html http://www.rtedw= ards.com/journals/JFA/I-1/125.pdf

Rushinek, A. and Rushinek, S. The Role of the Forensic Accountant in Calculating the Damages Using the ‘But-if Anal= ysis in a Case of Internet Day Trader & Online Broker Misconduct Litigation, JOURNAL OF FORENSIC ACCOUNTING, Vol. 1, No. 2, 241-250, 2001. http://www.rtedwards.com/journals/JFA/content= s.html

Avi Rushinek, Sara F. Rushinek, Lucia S. Ch= ang, Kenneth S. Most "The construction of an international investors' perce= ption model for corporate published forecasted financial reports for the USA, the= UK and New Zealand," Managerial and Decision Economics, Volume 4 Issue 4, Pages 258 - 265, Published Online: 9 Nov 2006, Copyright © 2009 = John Wiley & Sons, Ltd. http://www3.interscience.wiley.com/journal/= 113456935/abstract?CRETRY=3D1&SRETRY=3D0

Rushinek, A. and Rushinek, S. "Sarbanes = Oxley Microsoft Accelerator Forensic Accounting, Expert Witness Testimony, & Computer Litigation Support Workshop", American Accounting Association Annual Meeting 2005, Saturday, August 4,2005, San Francisco, California, sponsored by Teaching and Curriculum Section<= /span>. http://aaahq.org/AM2005/cpe/cpe02.htm<= /a>

=

Rushinek, A. and Rushinek, S. "Assessment Automation Software for Accounting, Teaching, and Recruiting Programs", American Accounting Association Meeting 2006, August 6, 2006, Washington, D.C., sponsored by Teaching and Curriculum Sect= ion. http://aaahq.org/AM2006/cpe/cpe01.htm<= /b> http://aaahq.org/pubs/AE= N/2006/AM2006.pdf

Rushinek, A. and Rushinek, S. "Assessmen= t CPE Video Podcasts and Blogs "Hands-on:" Consult, Publish & Teach Forensic Auditing, Expert Witness Testimony & Microsoft® Office Accounting Computer Litigation Support," American Accounting Associati= on Annual Meeting 2007, Saturday, August 4, 2007, Chicago, Illinois, sponsored= by Teaching and Curriculum Section. http://aaahq.org/am2007/cpe/cpe08.htm

Rushinek, A. and Rushinek, S. How to Use New = Media and Technology to Repurpose Classroom Instruction for the Accounting and Business Educator 2008 International Fraud and Forensic Accounting Education Conference, May 9, 2008, Las Vegas, Nevada http://www.docstoc.com/docs/5525868/2008-Inte= rnational-Fraud-and-Forensic-Accounting-Education-Conference

Rushinek, A. and Rushinek, S. "Mock Trial & Forensic Faculty WWW Marketplace, Jury, Judge & Attorneys "Courtroom & Classroom Instructional Technology Research & Service," American Accounting Association Annual Meeting 2008, Anaheim, California, CPE Session 2: Saturday, August 2, 8:00 AM-11:00 AM, sponsored = by Teaching and Curriculum Section. http://aaahq.org/AM2008/cpe/cpe02.htm

Rushinek, A. and Rushinek, S. "Rol= e Play Forensic Accounting, Auditing & Tax Expert Witness: Providing Testimony= and Computer Litigation Support Teaching, Services, Research and Development,&q= uot; American Accounting Association Annual Meeting 2008, Anaheim, California, a= nd CPE Session 11: Saturday, August 2, 1:00 PM-4:00 PM, sponsored by Teaching = and Curriculum Section.

http://aaahq.org/AM2008/cpe/cpe11.htm

Rushinek, A. and Rushinek, S. How to Use Case= s, Role Playing and Mock Trials in the Classroom: The Role of the Accounting a= nd Business Educator with the Aid of New Computer Media, American Accounting Association Annual Meeting 2008, August 6, 2008, Anaheim, California http://aaahq.org/TeachCurr/newsletters/Summer= 2008.pdf

Rushinek, A. and Rushinek S= . ""Tele-Medicin= e's SMART Surveillance AV HD IP Security Network Camera Audit Systems,= '" Jay Weiss Center for Social Medicine and Health Equit= y, University of Miami Miller School of Medicine, 2009 JWC SMART Forum, Loi= s Pope LIFE Center, April 24,2009, Miami, Florida. http://aaahq.org/AM2009/Emerg= ing.cfm

Rushinek, A. & Rushinek, S., (2009, August). "IFRS Compliant Real-Time 2-Way AV (Audio/Video) HD (High Definition) Recording Audit of Effective Learning Strategies for Forensic IT Networks of Expert Witnesses Research that Focuses on Investigative Accounting Testimon= y, & Computer Litigation Support, Effective Learning Strategies" 2009= American Accounting Association (AAA) Annual Meeting = NY, NY., Monday August 3, 2009, retrieved 10-27-2009 http://aaahq.org/AM2009/ELS01.cf= m Rushinek, A. and Rushinek S. "Cell Phone Real Time Forensic Audit Of Tele-Medicine e-Bill Records over the WWW AV IP Network Security Surveillan= ce Reconnaissance PTZ Cam Systems, CCTV & DVD-Recording: Repurposed Stand-= Up Instructions for Distance or Blended Education and CPE," 2009 AAA American Accounting Association, Annual Meeting Proceedings, August 1-5, 2009, New Y= ork, New York. = http://aaahq.org/AM2009/Emerging.cfm Rushinek, A. & Rushinek, S., (2009, August). "Cell Phone Real Time Forensic Audit of Tele-Medicine e-Bill Records o= ver the WWW AV IP Network Security Surveillance Reconnaissance PTZ Cam Systems, CCTV & DVD-Recording: Repurposed Stand-Up Instructions for Distance or Blended Education and CPE, Emerging and Innovative Research Session" 2= 009 American Accounting Association (AAA) Annual Meeting = NY, NY., Tuesday August 4, 2009, retrieved 10-27-2009 http://aaahq.org/AM2009/Emerg= ing.cfm Rushinek, A. and Rushinek S. "Health Care Forensic Information Technology Networks of Expe= rt (FITN) Witnesses Teams Researching Investigative Tax, Accounting & Audi= ting (FIA) Forecasts Archives of Testimony, Assessment, Publishing, Teaching, Service, & WWW Multi-Media Consulting," = 2009 AAA American Accounting Association, Annual Meeting Proceedings, August 1-5, 2009, New Y= ork, New York. Retrieved 10-27-2009 http://aaahq.org/AM2009/Emerg= ing.cfm Werner, M. and Jones, K. (2009), Introduction to Management Accounting – A User Perspective, Third Edition, Kendall Hunt Publishing, 200= 9. <= /p> US Census Bureau (= = 2009), the quarterly publication, <= span style=3D'letter-spacing:-.15pt'>Quarterly Financial Report for Manufacturin= g, Mining, and Trade Corporations – Third quarter 2009 (QFR/09-Q3= ), December 14, 2009, US Census Bureau Report http://www2.census.gov/econ/qfr/current/qfr_mg.pdf Retrieved 1-15-2010= <= /p> \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ Appendix: <= /p> Chart 1: <= /o:p> Screen Shot = 1 <= /o:p> <= /p> <= /p> Sample Data Set 1<= /span> Foot Notes to Data Set 1= 1-The seasonally adjusted est= imates provided in this table were derived using a combination of SIC-based and NAICS-based estimates.   &nb= sp;            =             &nb= sp;            =                    &= nbsp;           &nbs= p;            &= nbsp; 2- Beginning with the 2007Q4 = data release, updated Regression ARIMA (Auto Regressive Integrated Moving Averag= e) models for seasonal adjustment were implemented to improve the identification of s= easonal patterns in unadjusted net sales and net income after taxes for all manufacturing, all nondurable manufacturing, and all durable manufacturing.=         &= nbsp;           &nbs= p;            &= nbsp;           3- Revised             =             &nb= sp;            =                        =             &nb= sp;            =             &nb= sp;            =             &nb= sp;         ///////////////////////////////////////////////////// Appendix 1: Sample Forensic Accounting Continuing Professional Education (CPE) Workshops for t= he American Accounting Association (AAA) Rushinek, A. and Rushinek, S. “Sarbanes Oxley Microsoft Accelerator Forensic Accounting, Expert Witness Testimony, & Computer Litigation Support Workshop”, American Accounting Association Annual Meeting 2005= , Saturday, August 4,= 2005, San Francisco, California, sponsored by Tea= ching and Curriculum Section. http://aaahq.org/AM2005/cpe/cpe02.htm An International Meeting of The American Accounting Association 2005 Annual Meeting August 7–10, 2005 San Francisco, California Come to the City by the Bay! ---------------------------------------------= ----------------------------------- CPE Session 2: Saturday, August 6, 8:00 AM &#= 8211; 4:00 PM Sarbanes-Oxley Microsoft® Accelerator For= ensic Accounting, Expert Witness Testimony and Computer Litigat= ion Support Description/Objectives: Sarbanes-Oxley Microsoft Accelerator Forensic accounting workshop demonstrates how Accounting academics and practitioners= can use Microsoft Share Point Team Services and Share Point Server to assist an internal audit team in implementing, investigating, and improving compliance audit services. It describes how to become an Accounting Expert Witness and= how to provide Expert Witness Testimony, with an emphasis on HIPAA (Health Insurance Portability and Accountability), Sarbanes-Oxley, and Basel II Compliance Forensic Accounting. We will discuss R&D for experienced expert witnesses, including the issues of dealing with the forensic team of professionals. This workshop focuses on using forensic accounting technique= s in general, and providing Computer Litigation Support in particular. The works= hop objectives include helping experienced expert witnesses to review their ski= lls and upgrade them, while demonstrating to prospective expert witnesses the nature of the work, its intellectual challenges and financial opportunities= and rewards. We plan to meet such objectives, by lectures, role playing, demonstrations, and most importantly active participation of the audience as well as discussions and arguments of pros and cons of various issues.<= /o:p> Format/Structure: DVDs for litigation support, including the pr= os and cons of different standards such as DVD+/-RW/VR (Digital Versatile Disk, Re= ad Write, and Video Recording), DVD-RAM (Random Access Memory), HD (High Definition)-DVD, Blu-Ray DVD, DVD Copying, dubbing, burning, and recording, focusing on the technical and legal facets. Intended Audience: This study targets practicing and academic litigation and litigation support team members, including lawyers, economis= ts, accountants, auditors, doctors, engineers, corporate executives, expert witnesses, plaintiffs and defendants, as well as students and others intere= sted in the topic. This workshop will review the financial opportunities for exp= ert witnesses, fee structures, relationship with the legal profession, and opportunities for students that want to specialize in the area of forensic accounting and litigation support, and/or combine the study of accounting a= nd law, and computer information systems. Presenters: Sara Rushinek, University of Miami= Avi Rushinek, University of Miami<= /span> Sponsor: Teaching and Curriculum Section ---------------------------------------------= ----------------------------------- * Denotes special requirements or prerequisit= e Back to CPE Listings Back to Annual Meeting H= ome Page <= o:p> <= o:p> <= o:p> An International Meeting o= f the American Accounting Association 2005 Annual Meeting August 7–10, 2005 San Francisco, California Come to the C= ity by the Bay!<= /h2> CPE Session 2: Saturday, August 6, 8:00= AM – 4:00 PM Sarbanes-Oxley Microsoft® Accelerat= or Forensic Accounting, Expert Witness Testimony, and Computer Litigation Support Description/Objectives: Sarbanes-Oxley Microsoft Accelerator Forensic accounting workshop demonstrates how Accounting academics and practitioners can use Microso= ft Share Point Team Services and Share Point Server to assist an internal audit team in implementing, investigating, and improving compliance aud= it services. It describes how to become an Accounting Expert Witness and h= ow to provide Expert Witness Testimony, with an emphasis on HIPAA (Health Insurance Portability and Accountability), Sarbanes-Oxley, and Basel II Compliance Forensic Accounting. We will discuss R&D for experienced = expert witnesses, including the issues of dealing with the forensic team of pr= ofessionals. This workshop focuses on using forensic accounting techniques in genera= l, and providing Computer Litigation Support in particular. The workshop objectives include helping experienced expert witnesses to review their skills and upgrade them, while demonstrating to prospective expert witnesses the nature of the work, its intellectual challenges and finan= cial opportunities and rewards. We plan to meet such objectives, by lectures, role playing, demonstrations, and most importantly active participation= of the audience as well as discussions and arguments of pros and cons of various issues. Format/Structure: DVDs for litigation support, including the pros and cons of different standards such as DVD+/-RW/VR (Digital Versatile Disk, Read Write, and = Video Recording), DVD-RAM (Random Access Memory), HD (High Definition)-DVD, Blu-Ray DVD, DVD Copying, dubbing, burning, and recording, focusing on = the technical and legal facets. Intended Audience: This study targets practicing and academic litigation and litigation support team members, including lawyers, economists, accountants, audit= ors, doctors, engineers, corporate executives, expert witnesses, plaintiffs = and defendants, as well as students and others interested in the topic. This workshop will review the financial opportunities for expert witnesses, = fee structures, relationship with the legal profession, and opportunities f= or students that want to specialize in the area of forensic accounting and litigation support, and/or combine the study of accounting and law, and computer information systems. Presenters: Sara Rushinek, University of Miami Avi Rushinek, University of Miami Sponsor: Teaching and Curriculum Section * Denotes special requirements or prerequisite Back to CPE Listings Back to Annual Meeting Home Page <= /p> <= o:p> <= strong>Note:= CPE fields of study are in parentheses Rushinek, A. and Rushinek, S. “Assessment CPE Video Podcasts and Blogs "Hands-on:" Consult, Publish & T= each Forensic Auditing, Expert Witness Testimony & Microsoft® Office Accounting Computer Litigation Support,” American Accounting Association Annual Meeting 2007, Saturday, August 4, 2007, Chicago, Illinoi= s, sponsored by Teaching and Curriculum Section. http://aaahq.org/am2007/cpe/cpe08.htm CPE Session 8: Saturday, August 4, 1:00 PM – 4:00 PM= Assessment CPE Video Podcasts and Blogs "Hands-on:&qu= ot; Consult, Publish & Teach Forensic Auditing, Expert Witness Testimony &a= mp; Microsoft® Office Accounting Computer Litigation Support (Specialized Knowledge & Applications) Description/Objectives: Repurpose DVD-recorded lectures using new Blu-ray® and HD-DVD® in the classroom and CPE, for the CPA, CMA, CIA, CISA/M WWW certificate Preps. Demonstrate download of the Microsoft Office Accounting,= and upload the lecture on iTunesU®, Google Video®, and YouTube®. Re= view test automation compliance with NASBA, AACSB, COSO, Sarbanes-Oxley, HIPAA, = and Basel II. Publish for students, alumnus, professionals, and the general pub= lic. Web cast using multi-media streaming tools to be more effective and efficie= nt, and provide a more enjoyable experience. Use the new Microsoft® Office = 2007 Accounting, and Vista® involving Effective Learning Strategies, Emerging and Innovative Research, and Internet Research Forums. Instructors encourage existing and prospective faculty and students to be involved and contribute= in Assessing Submissions, Publications, Faculty Development, Teaching, Researc= h, Practice, Service, Awards, and Placement streamline automation and SEO (Sea= rch Engine Optimization). Presenters: Avi Rushinek, University of Miami Sara Rushinek, University of Miami Sponsor: Teaching and Curriculum Section. Rushinek, A. and Rushinek, S. “How to Use New Media and Technology to Repurpose Classroom Instruction for the Accounting and Busine= ss Educator” 2008 International Fraud and Forensic Accounting Education Conference, May 9, 2008, Las Vegas, Nevada= http://www.docstoc.com/docs/5525868/2008-Inte= rnational-Fraud-and-Forensic-Accounting-Education-Conference Rushinek, A. and Rushinek, S. "Mock Trial & Forensic Fa= culty WWW Marketplace, Jury, Judge & Attorneys "Courtroom & Classroom Instructional Technology Research & Service," American Accounting Association Annual Meeting 2008, Anaheim, California, CPE Session 2: Saturday, August 2, 8:00 AM-11:00 AM, sponsored by Teaching = and Curriculum Section. http://aaahq.org/AM2008/cpe/cpe02.htm<= /a> = CPE Session 2: Saturday, August 2, 8:00 AM – 11:00 AM = Mock Trial & Forensic Faculty WWW Marketplace, Jury, Judge & Attorneys — Courtroom & Classr= oom Instructional Technology Research & Service = (Accounting - Basic) <= strong>Description/Objective= s: The workshop deals with: E-Discovery Testimony, Accounting Corporate Information Unauthorized Disclosure, Identity Theft, Money Laundering, Fraud investigations, Office Accounting Cops & Robber Simulations, Phony Sales & Fraud detections, Spreadsheet Database Research Tracking, Digital Med= ia Evidence CDs & DVDs, CYBER CRIME, Security ethical hacking anti-Virus controls, Cell Phones & PDAs Evidence, Trade Secrets Alleged Theft, Electronic Evidence Management. Threats Software & Hardware tool Integration Scenarios Deposition Preps, Jury Deliberation Litigation Arbitration & Mediation Industry Specific Academia, Hedge Fund, Governm= ent & Medical Shenanigans, Plagiarism, Theft & Misrepresentation Contro= ls Red-flags White-color Schemes, Scams Cons Crooks Drug & Human Trafficki= ng Certification Qualification Monitor Confessions Boiler-Plate Rooms, Schemes, Counsels, Judges, & Lawyers Gone Wild (900)Sue-U2.com Foreign Offshore Banking, Tax Avoidance Evasion, & Wealth Management PI Family Divorce Real-estate Stock-Market Litigations Hired Guns Cross Examination Structured Settlements. <= strong>Advance Preparation: = Email and or call presenters (arush@miam= i.edu, arushine@aol.com, (305) 666-7890) a= bout cases, articles, issues, or any relevant suggestions and/or materials, and/= or ideas for discussion and analysis during the workshop to presenters. Dr. Avi Rushinek, University of Miami, Accounting Department, Jenkins 421, C. Gable= s, Fl, 33124 USA (Run)Air-Webs/Phone (Cell) U. Miami, 5250 University Drive, C.I.S. Dept. Jenkins 421, C. Gables, Fl, 33124 (786)247-9327/7466 (Cell) Ac= c. Dept. 314 Jenkins, UM, Coral Gables, Fl, 33124, USA or 1205 Mariposa Ave. #= 208, Coral Gables, Fl, 33146. http://www.onafree.com/defa= ult.aspx http://drjd.net/default.aspx http://dvdu.org/default.aspx http://moax.org/default.aspx http://www.webjobnet.com <= strong>Presenters: Avi Rushinek, University of Miami Sara Rushinek, University of Miami <= strong>Note:= CPE fields of study are in parentheses Rushinek, A. and Rushinek, S. "Role Play Forensic Accountin= g, Auditing & Tax Expert Witness: Providing Testimony and Computer Litigat= ion Support Teaching, Services, Research and Development," American Accoun= ting Association Annual Meeting 2008, Anaheim, California, and CPE Session= 11: Saturday, August 2, 1:00 PM-4:00 PM, sponsored by Teaching and Curriculum Section. http://aaahq.org/AM2008/cpe/cpe11.htm CPE Session 11: Saturday, August 2, 1:00 PM &= #8211; 4:00 PM Role Play Forensic Accounting, Auditing &= Tax Expert Witness: Providing Testimony and Computer Litigation S= upport Teaching, Services, Research and Development<= o:p> (Accounting - Basic) Description/Objectives: The workshop deals with: Rogue Wireless Access Points, Internal Threat Management, Vulnerability Exploits Web, Intrusion Detection, Eaves Dropping AV, Surveillance Recording Audit, On Demand Live Consulting, Internet Access Data points, Consumer Electronics Gone Wild, IT (Information Technology) Governance, Log Forensics Stealth Regulatory Compliance Audit Trails, Mobile Device Digital Consumer Electronics Transcription Stenography, voice to text automation, Cell Phone Video Cam Justice & Court TV Shows Camera Imaging Distribution, Lawsuit Incident Response, WWW Network URL TCP/IP HTTP, Stealth Data Intelligence Office Acc= ounting Information Systems Demos "But-for/If" Analytics Metrics Trap Door Trojan Horses, Spying & Due Diligence Enterprise Risk & Damage Assessment Market Rates & Charges Malpractice & Business Valuation "Lost Profits" Board of Directors Litigation Service Confessions Acting Telling a Story Sound-byte Tips & Tricks Interrogate Terror Home-land Security Bribery Black-mail Corruption & Where is the Money Laundering. Advance Preparation: Email and or call presenters (arush@miami.edu, arushine@aol.com, (305) 666-7890) about cases, articles, issues, or any relevant suggestions and/or materials, and/or ideas for discussion and anal= ysis during the workshop to presenters. Dr. Avi Rushinek, University of Miami, Accounting Department, Jenkins 421, C. Gables, Fl, 33124 USA (Run)Air-Webs/Phone (Cell) U. Miami, 5250 University Drive, C.I.S. Dept. Jenkins 421, C. Gables, Fl, 33124 (786)247-9327/7466 (Cell) Acc. Dept. 314 Jenkins, UM, C. Gables, Fl, 33124, USA. http://www.onafree.com/default.aspx http://drjd.net/default.aspx http://dvdu.org/default.aspx http://moax.org/default.aspx http://www.webjobnet.com Presenters: Avi Rushinek, University of Miami<= /span> Sara Rushinek, University of Miami= ________________________________________= Note: CPE fields of study are in parentheses<= o:p>

Table 1	R² Goodness of Fit Differences Between the Polynomial of the 3rd Degree and the Linear Model
	Column L	M	N	O	P	Q
Line 66	Model	Regression	R²	y
Line 67	1	Poly2	9= 9.9%	y =3D 1E-07x² + 0.701x + 80317<= /p>
Line 68	2	Poly3	9= 9.9%	y =3D 1E-08x³ - 0.005x² + 1372.x -1E+0= 9
Line 69	3	Linear	9= 9.8%	y =3D 0.885x + 22809
Line 70		Difference	*0.1%*	=3DN68-N70
Line 71		TMC =3D 0.885*Sales +$ 22,810