Advanced Techniques for Forecasting Financial Statements Accounts
The techniques in Chapter 9 assumed that many items listed on the financial state-ments were proportional to sales, but that is not true in some cases. In this exten-sion, we show how to forecast an item when it is not proportional to sales.
If we assume that the relationship between a certain type of asset and sales is linear, then we can use simple linear regression techniques to estimate the require-ments for that type of asset for any given sales increase. For example, MicroDrive’s sales, inventories, and receivables during the last 5 years are shown in the lower section of Figure 9A-1, and both of these current asset items are plotted in the upper section as scatter diagrams versus sales. Estimated regression equations, determined using a financial calculator or a spreadsheet, are also shown with each graph. For example, the estimated relationship between inventories and sales (in millions of dollars) is
Inventories 5 2$35.7 1 0.186(Sales)
The plotted points are not very close to the regression line, which indicates that changes in inventory are affected by factors other than changes in sales. In fact, the correlation coefficient between inventories and sales is only 0.71, indicating that there is only a moderate linear relationship between these two variables. Still, the regression relationship is strong enough to provide a reasonable basis for forecast-ing the target inventory level, as described next.
We can use the regression relationship between inventories and sales to fore-cast 2013 inventory levels. Since 2013 sales are projected at $3,300 million, it fol-lows that 2013 inventories should be $578 million:
Inventories 5 2$35.7 1 0.186($3,300) 5 $578 million
This is $99 million less than the preliminary forecast based on the projected fi-nancial statement method. The difference occurs because the projected financial
© N
ikad
a/iS
tock
pho
to.c
om
Web Extension 9A
CHE-BRIGHAM-11-0504-WEB EXTN-009A .indd 1 21/03/12 8:33 PM
© C
enga
ge L
earn
ing.
All
right
s re
serv
ed. N
o di
strib
utio
n al
low
ed w
ithou
t exp
ress
aut
horiz
atio
n.
9wA-2 web Extension 9A Advanced Techniques for Forecasting Financial Statements Accounts
statement method assumed that the ratio of inventories to sales would remain con-stant when, in fact, it will probably decline. Note also that, although our graphs show linear relationships, we could easily have used a nonlinear regression model had such a relationship been indicated by the data.
MicroDrive Inc.: Linear Regression Models (Millions of Dollars)Figure 9A-1
2,000
300
400
500
600
700 400
350
300
250
200
2,250 2,500
Inventories = —35.7 + 0.186 (Sales)Receivables = 62 + 0.097 (Sales)
2,750 3,000 Sales ($)
Inventories($)
Receivables($)
0 2,000 2,250 2,500 2,750 3,000 Sales ($)0
Year Sales InventoriesAccounts
Receivable
2008 $2,058 $387 $268
2009 2,534 398 298
2010 2,472 409 304
2011 2,850 415 315
2012 3,000 615 375
CHE-BRIGHAM-11-0504-WEB EXTN-009A .indd 2 21/03/12 8:33 PM
© C
enga
ge L
earn
ing.
All
right
s re
serv
ed. N
o di
strib
utio
n al
low
ed w
ithou
t exp
ress
aut
horiz
atio
n.