The right of Sue Nugus to be identified as the author of this work is asserted in accordance with the Copyright, Designs and Patents Act 1988. You can also submit your request on-line via the Elsevier homepage (http: //www.elsevier .com), by selecting 'Customer Support' and then 'Getting Permissions'.
Introduction
Approaches to forecasting
Most of the forecasting examples in the book would be described as objective or quantitative forecasts. These forecasts must be developed in such a way that actual data can be fed into the model so that a comparison can be made between the forecast and the actual data.
Data collection
The last set of data to examine is shown in Figure 2.4 and consists of quarterly data for the number of conferences. Figure 2.3 Historical data for 24 monthly periods showing evidence of seasonality. The chart shows that there appears to be a five-year cyclical pattern in the data.
Using statistical measures
To calculate the arithmetic mean of the sample of output weights, the AVERAGE function is used as follows in cell B4. Therefore, it estimates the standard deviation of the sample mean based on the population mean (Press et al.1992, p. 465)1.
Summary
Using an average or an arithmetic mean produces a value that is typical or representative of a given set of data. The arithmetic mean or simple mean of a data set produces a straight line through the data.
Moving averages
The three-month moving average shown by the green line in Figure 3.2 reflects changes in data and is easily affected by irregularities and fluctuations. The five-month moving average shown by the blue line, on the other hand, is smoother and shows less impact of irregularities and fluctuations.
Weighted moving average
In the proportional method, each value in the moving average is multiplied by a specified weight, and the total weight is usually equal to 1. The effect of applying this technique to the second set of data from Chapter 2, which showed some evidence of a trend, can be seen in Figure 3.6, and Figure 3.7 displays the results graphically.
Adaptive filtering
The adjusted weights will no longer add up to 1, because if the actual amount is less than the forecast, the combined weights will end up being less than 1. The reverse is true in that if the actual amount is more than the forecast, the combined weights total more than 1.
Exponential smoothing
At a low coefficient value of 0.20, a high degree of smoothing is expected, which is shown in the graph in Figure 3.10. Looking at the chart in Figure 4.3, it is clear that there is a strong linear relationship between the two variables.
Calculating the least square line
The regression line and regression equation only apply within the independent range or X interval of the data. 2 The Analysis Toolkit is a standard part of the Excel suite, but if a custom installation was selected during setup, this option may not have been selected.
Function vs command
The regression formula can be created by referencing the intercept or constant in cell B17 and the X coefficient in cell B18.
The graphic approach
Although the individual calculations behind the line cannot be seen, choosing the FORMATTRENDLINE OPTION allows the formula to be displayed as shown in Figure 4.11.
Standard error
The output of the Regression command in Figure 4.9 is placed on a sheet called LR Command, and this can be seen in the formula when referring to the Student t distribution, standard error, and number of observations. The results of these formulas are shown in Figure 4.12 and Figure 4.13 shows the lines plotted on the graph.
Multiple linear regression
The following example in Figure 4.14 represents expenditure on advertising and sales promotion together with sales achieved, where the advertising and sales promotion are dependent variables and the sales achieved is the independent variable. In Excel there are no built-in functions for multiple regression and therefore the command method is required.
Multiplicative time series analysis
A time series analysis model
Figures 5.3 and 5.4 show the results section of the Calculation and Processing sheet, and Figures 5.5 and 5.6 show the. The trend describes the long-term behavior of the series after removing the seasonal effects and the irregularities due to short-term random fluctuations.
A model for analysing expected values
More information about using this function can be found in the Help function in Excel. This places a small arrow to the right of each column heading in the expected value table.
Selecting the right technique
For any forecast to be useful, it must be accurate to an acceptable level, and the level of accuracy can be significantly affected by the forecasting approach chosen. Exponential Recurring forecast Some statistics Just recent smoothing with or without knowledge to set the data and seasonality of the system, but currently.
Accuracy and reliability
Charts
Statistical methods
Once the average error for a particular forecasting method is calculated, it can be compared to the average error for another method. Although the mean error has no intrinsic statistical property that makes it useful for estimating confidence limits, it is a useful indicator of forecast accuracy.
Subjective methods
Business forecasting involves predicting what may happen in the future and business planning involves calculating what is required to fulfill the forecast ie. the level of analysis in the plan reflects the end use for which it is intended.
Approaches to business planning
Getting started
The spreadsheet in Figure 9-1 is a simple profit projection that may be useful to the author, but is unlikely to be useful to anyone else. If the spreadsheet author needed a quick profit estimate based on known data and growth rates for sales units, prices, and costs, the spreadsheet provided that information quickly and in a more concise form than would have been achievable using a calculator and the recording the data. results on paper.
Ownership and version
Formatting
When formatting a spreadsheet, it is important to consider the entire plan and not just the cells that are currently being worked on. It is important to understand that formatting cells only changes the display and does not affect the results of calculations which are still performed to the full degree of accuracy, usually 16 significant decimal places.
Documentation
A third form of documentation that can be particularly useful for large systems is the "sentence at the end of the line" technique. Requiring less file space than comment boxes and always in view, it can be useful to have a brief description of the activity taking place on each line of a plan.
Minimising absolute values
Because no growth in the price is required, the opening value of 12.55 was copied for the four quarters. With this spreadsheet, it would be necessary to overwrite the price in the first quarter and then copy the new value for the remaining three quarters.
Separating growth and cost factors
However, as already mentioned, this plan incorporates absolute values into the formulas for sales and cost growth. In this plan an option is included in the growth factors for the price, despite the fact that in this plan the price does not change.
Optimizing layout
It is important to always think ahead when developing a plan and although the price is not changing currently, it may be necessary to implement a percentage increase in the future. Building the ability to change into the plan can save time later – and for now the growth factor is simply set to zero.
Arithmetic cross-checks
Charts
Mutiple sheets
This color coding can then be carried over to summary reports and other reports dealing with different parts of the plan. To assign a name to a worksheet, simply double-click the sheet reference at the bottom of the screen and type the required name.
Templates
When the template is complete, the spreadsheet should be protected, specifying only the cells into which data can be entered as unprotected cells. The location of the template file defaults to the directory where other Microsoft Office template files are located.
Data input forms
There are many benefits that can be derived from using data entry forms including the fact that data can be checked more easily. Even if this is not possible, the order of the items in the data entry form does not have to be the same as the order in which they are referenced in the logic, which means that the data entry form can be created to be as compatible with the input data source as possible.
The model
Getting started
It forms the basic structure of the plan and enables cross-referencing when developing formulas. Figures 10.1 to 10.6 show the six worksheets that make up the complete plan with sample data.
Developing the Profit and Loss Account
In the second and subsequent periods, the volume of sales should increase by the value entered in cell C5 of the input sheet. The remaining lines in the income statement follow the same principles and Figure 10.7 shows the formulas for this part of the financial plan.
Developing the Profit and Loss Appropriation Account
Developing the balance sheet
The opening balance for trade payables is taken from the entry sheet, but in the first quarter the payables are calculated by taking total direct costs minus labor (which is paid in cash) plus total other costs minus wages and depreciation. Therefore the required formula in cell B43, which can be copied for the remaining periods, is:
Funds flow statement
The same procedure is required for the reserves by adding the opening reserves from the balance sheet section of the Input Statement and then adding the reserves from the profit and loss account in subsequent quarters. The total capital and liabilities can then be calculated by adding together the total net assets and the owners' equity.
Ratio analysis
Cash flow statement
The cash expenses include labor and salaries, which means the following formula is required in cell C11 and can be copied for the remaining periods. There will be no tax or dividend distribution, although an accommodation for this has been included in the cash flow plan.
Capital investment appraisal
There are several discounted cash flow (DCF) measures available in the spreadsheet, with Net Present Value (NPV), Profitability Index (PI) and Internal Rate of Return (IRR) being some of the most commonly used. PI is defined as the sum of the present values of the cash flows divided by the present value of the investment.
Developing a capital investment appraisal plan
For this function to work properly in this context, it is necessary that the values in the first row of the table are sorted in ascending order. The number of rows to count down for the result, where the first row of the table is 1.
Learning curve costing
The effect of this is to emphasize the impact of the learning curve effect itself on profits. The formula for variable costs in column E begins in year 1 by adding material costs and labor costs and multiplying by production.
Break-even analysis
The formulas needed to calculate the break-even point for the plan in Figure 11.7 are not complex and are shown in Figure 11.8. The result will be a break-even statement for the firm as a whole, which will include a range of volumes, one for each product.
Economic order quantities (EOQ)
An additional input element is required for calculating EOQ with a supplier policy and that is the minimum order requirement. As can be seen from Figure 11.10, although the inventory cost is higher for the supplier policy scenario, the overhead costs are still lower due to the lower selling price.
Sales campaign appraisal
In this example, it is estimated to be between 30 and 40% of the previous week's inquiries. This is achieved by calculating the result of a series of relatively small changes in the specified input factors on the objective function, or output of the business model.
Three approaches to what-if analysis
Sensitivity analysis is a technique that determines the relative importance of certain input variables in a business scenario or plan. A user of sensitivity analysis wants to determine whether a 5% increase in raw material costs or a 2% increase in labor costs will have a negative impact on the firm's profits.
Manual what-if analysis on opening assumptions
Furthermore, if a data entry form has been created as shown in Chapter 9, it is easier to see the input data that can be changed. In this case, it can be difficult to know which change has the greatest impact on the result.
Data tables
Therefore, it is necessary to refer to the cell in the main model in which the input values are to be entered. If the opening price in the model is changed to 50, the net result for December in the model will be NOK 56,120, as it is in the table.
Backward iteration or goal seeking
Goal Seek is accessed by selecting TOOLS GOAL SEEK and the dialog box shown in Figure 12.8 appears, which must be filled in with the appropriate information. After setting up the spreadsheet, TOOLS SOLVER is selected and the Solver Parameters dialog box appears, as shown in Figure 12.10.
Preparing a plan for risk analysis
This chart will generally be a bell shaped curve and the exact shape of the curve will reflect the degree of risk present in the investment based on the input data series. Cells and ranges in the output selection part of the input form should be labeled for future reference in the results sheet.
Incorporating the RAND function
The range D17 to F19 is named 'looktab' and will be referred to as a lookup table in the result sheet. The system generates a different random number each time the table is recalculated and so pressing F9 will return a different set of values to the main model and in turn the investment ratios will be recalculated using different data each time.
The results worksheet
The more iterations of the model, the better the results, so for this example the plan was set to perform 2000 recalculations and thus collect 2000 different results. The results of the risk analysis are most clearly visible on the graph, but 2000 data points are too many to display on the graph.
Frequency distribution
The results can be displayed graphically by selecting the range G4 to H15 and using the Chart button to automatically create a graph like the one shown in Figure 13.4. Due to the use of the RAND function, the results of this exercise will never be exactly the same as those displayed here.
Using the risk analysis model
If it is appropriate to specify ranges for only some of the input variables, the same value can be entered for the minimum and maximum in the input sheet. The method used for this example can be used with most well-designed deterministic plans.
Scope of budgeting
The essence of a budget is that it is a target set for management to keep within, achieve or exceed. Thus, a budget is always associated with a specific point or center of departmental responsibility within the organization.
Benefits of budgeting
Different approaches to budgeting
One result of a ZBB system that is not the case with the traditional approach is that management is forced to demonstrate the necessity of each expenditure item in the budget. ZBB is not considered directly relevant to the production process as detailed expenses in this area are typically automatically accounted for on a variable or direct basis.
Budget preparation
It is not good enough to say that there was a certain amount in the promotional account last year and that this figure should be increased by 10% because of inflation. Clearly, it is more expensive to operate a ZBB system than a traditional system, and therefore a cost-benefit study is appropriate before embarking on a ZBB.
Spreadsheets for budgets
Detailed cost calculations of each responsibility center are made – This requires the involvement of a wide range of staff and can lead to a significant amount of negotiation.