Audit evidences and modelling audit risk using goal programming
29
Figure 3 Relationship between audit risk and audit sample (see online version for colours)
5 Simulation model
Three models can be used for the analysis of audit risk and to determine audit evidence.
The first model is the non-pre-emptive model which it does not give any preference or
priority to any of the goals. The second model is the sequential and the third one is called
streamline pre-emptive method. However, we only show how audit DR can determine the
audit evidence using the first model which is non-pre-emptive model.
5.1 Non-pre-emptive
model
In this model, we assume that the auditor assigns no priority to the goals. The auditor
could feasibly use this model by modifying the constant values of the variables for the
audit resources. We set out below the stages that need to be followed to determine the
optimal number of pieces of audit evidence that should be collected and supported
through audit samples for each major account in order to create the final audit opinion.
5.1.1 Stage one
An audit partner should optimise three objectives: minimising the audit risk components
to increase audit quality, maximising audit profit, and using the audit resources to achieve
the highest level of audit efficiency. High audit quality will be achieved by minimising
the audit risk. The main variable is the optimal number of audit samples, as the higher the
audit risk, the greater the number of pieces of evidence that must be collected to support
the audit opinion. If the auditor uses more audit samples than are needed, the audit costs
will be higher, more time will be needed to collect and analyse those extra samples, and
staff will spend more time compiling the samples; ultimately, audit quality with be higher
as audit risk will be reduced. The auditor may define sample sizes for each class of
30
S. Askary et al.
accounts, of which there are five main ones: assets, liabilities, revenue, expenses and
other accounts such as contra accounts (accumulated depreciation, provision for doubtful
debts, etc.), stockholder accounts (retained earnings, different shares, etc). There will also
be other documents required to support the audit report. The other accounts are usually
not many in number, but they may be important in nature. We defined x
1
to x
5
as
variables to measure the optimal number of pieces of evidence for the audit:
x
1
= Samples size to be collected for asset accounts
x
2
= Samples size to be collected for liability accounts
x
3
= Samples size to be collected for revenue accounts
x
4
= Samples size to be collected for expense accounts
x
5
= Sample size to be collected for other accounts.
Information about the sample size for the previous year’s audit could be used as a guide
for the sample size in the current year, if the auditor is continuing to audit the same client.
If the auditor is doing an audit for a new client, then the auditor will determine the sample
size during the audit planning stage by looking at previous experience with clients of the
same size and nature. In any case, the auditor will ensure the sample size is correct by
using statistical sampling techniques to double check the accuracy of the calculation.
∑ Samples
t – 1
is the total of sample collected last year for the client.
1
1
.
.
.
.
.
t
t
Samples
Ass S
Liab S
Rev S
Exp S
Others S
−
−
=
+
+
+
+
∑
∑
∑
∑
∑
∑
where
Ass.S = Total samples collected all assets accounts last year
∑
Liab.S = Total samples collected all liabilities accounts last year
∑
Rev.S = Total samples collected Revenue and Gains accounts last year
∑
Exp.S = Total samples collected all Expense and loss accounts last year
∑
Other.S = Total samples collected all other accounts last year
∑
The overall goal of the auditor is to minimise the DR, to increase audit quality. To use
GP, the auditor must determine the DR for each class of accounts. The overall DR is the
sum of the DR for each class of accounts as follows:
.
.
.
1
Asset
Liab
Rev
Exp
others
t
DR DR
DR
DR
DR
DR
−
=
+
+
+
+
∑
5.1.1.1 Goal
1
The auditor should determine the DR value, which is CR × IR divided by AR. Suppose
the auditor’s CR and IR values are 0.05 and 0.04, and the AR value is assumed to be as
low as 0.01 or 1%. Then DR would be 0.20, or 20%. The following determines the value
of DR for the asset accounts. If ∑Ass. is defined as the total samples taken from all the
Audit evidences and modelling audit risk using goal programming
31
asset accounts divided by total number of samples taken to complete the audit
(∑ Samples), then DR
Assest
is:
.
Assest
Ass
DR
DR
Samples
∑
=
×
∑
The auditor’s goal is to minimise the DR so that it is less than or equal to 0.2, or 20%, in
total. Therefore, suppose the DR for assets, from the above formula, is 0.04 or 4%, and
the DR for liabilities is 0.03, the DR for revenue is 0.06, the DR for expenses is 0.04, and
the DR for other accounts is 0.02. Accordingly, the first goal constraint will be
formulated as follows:
1
2
3
4
5
Audit Risk goal: 0.04
0.03
0.06
0.04
0.02
0.20
x
x
x
x
x
+
+
+
+
≤
5.1.1.2 Goal
2
The second goal is that the operating audit cost budget is less than or equal to the total
planned audit costs. The audit partner estimates that the cost of each sample of
x
1
to x
5
is $30, $25, $20, $25, and $25, respectively. This could be determined by the
following formula:
(
)
1
1
1
1 Δ%
t
t
t
t
t
All Audit Cost
Indirect Audit Cost
Estimated Cost of all Sample
Samples
−
−
−
∑
− ∑
=
∑
× +
This formula estimates the cost of each sample starting with the total audit cost from the
previous year and deducting all the indirect audit costs from the total cost to give the
direct audit costs. The auditor obtains the cost for each sample using the number of
samples for each class of accounts and dividing the direct audit cost by the number of
samples, then multiplying this figure by the percentage increase in the costs for this year.
For the asset account, for instance, the estimated cost per sample can then be determined
from this formula:
1
t
t
Ass.
Cost of Samples for Assets Accounts
Samples
Estimated Cost of all Sample
−
∑
=
∑
×
Assume the audit partner estimates that the total cost should be around X = $150,000 this
year, then the goal can be expressed in terms of the decision variables as:
1
2
3
4
5
Total cost goal: 30
25
20
25
25
150,000
x
x
x
x
x
+
+
+
+
≤
5.1.1.3 Goal
3
Maintaining the minimum number of staff for the audit engagement while keeping the
audit risk low is another optimisation goal for an audit firm.
(
)
1
1 Δ%
t
t
t
Total Direct Personel Cost
Estimated No Staff
Staff
−
∑
=
× +
∑
32
S. Askary et al.
1
.
t
t
Ass
No Staff Asset Accounts
Estimated No Staff
Samples
−
∑
=
×
∑
For example, suppose the audit manager plans to use a fraction of 0.08 staff members to
work on the asset account, 0.3 on the liability account, 0.2 on the revenue account, 0.07
on the expense account and 0.3 on the other accounts, but the total number of the staff
being used should not be greater than 35. Then, the goal can be expressed in terms of the
decision variables as:
1
2
3
4
5
Staff limitation goal: 0.08
0.3
0.2
0.07
0.3
35.
x
x
x
x
x
+
+
+
+
≤
5.1.1.4 Goal
4
Finishing the task within m days (m ≤ 100) is another constraint and objective for every
audit engagement.
1
Δ%
t
t
t
Total Engagement Period
Estimated Days per Sample
Samples
−
∑
=
×
∑
Then, for instance, for the asset accounts the number of days to be spent should be:
1
.
t
t
Ass
Estimated day for Assets Accounts
Samples
Estimated Days per Samples
−
∑
=
∑
×
Suppose, for example, that the task should be finished within 100 days and that the
auditor will need for each sample from the asset account, liabilities account, revenues
account, expenses account and other account 0.2, 0.5, 2.5, 1.5 and 0.5 days, respectively.
Then, the goal can be expressed in terms of the decision variables as:
1
2
3
4
5
Time limitation goal: 0.2
0.5
2.5
1.5
0.5
100
x
x
x
x
x
+
+
+
+
≤
5.1.2 Stage two
At this stage, the auditor should determine the penalty weight for going over each goal.
From this, we develop auxiliary variables y
1
to y
4
, where y
i
represents the amount by
which the i
th
goal is exceeded. Subsequently, the objective function Z is the minimisation
of the sum product of the weights and auxiliary variables.
Table 1
Summarises the model, goals, and penalty weights
Factors
Audit samples for accounts
Goal
Penalty
weight
Assets Liabilities Revenues Expenses Others
x
1
x
2
x
3
x
4
x
5
Detection risk
0.04
0.03
0.06
0.04
0.02
0.2
40
Cost
30
25
20
25
25
150
7
Audit team
0.08
0.3
0.2
0.07
0.3
35
5
Audit period
0.2
0.5
2.5
1.5
0.5
100
10
Audit evidences and modelling audit risk using goal programming
33
From Table 1, we obtain the following integer non-linear programming formulation of
this GP problem:
Minimise
1
2
3
4
40
7
5
10
Z
y
y
y
y
=
+
+
+
Subject to:
1
2
3
4
5
1
1
2
3
4
5
2
1
2
3
4
5
3
1
2
3
4
5
4
0.04
0.03
0.06
0.04
0.02
0.2
30
25
20
25
25
150,000
0.08
0.3
0.2
0.07
0.3
35
0.2
0.5
2.5
1.5
0.5
100
x
x
x
x
x
y
x
x
x
x
x
y
x
x
x
x
x
y
x
x
x
x
x
y
+
+
+
+
−
=
+
+
+
+
−
=
+
+
+
+
−
=
+
+
+
+
−
=
Solution:
1
2
3
4
5
1
2
3
4
498,
1,
1,
1,
1,
0,
17,
5,
4.
x
x
x
x
x
y
y
y
y
=
=
=
=
=
=
=
=
=
The y
i
values indicate the following:
• The audit risk goal is met.
• The audit cost is exceeded by $17, which is negligible.
• The number of staff is exceeded by 5; i.e., 40 staff members would be needed.
• The number of days is exceeded by 4; i.e., the total days would be 104 days.
The reader should note that the first goal is met as it has the highest weight/penalty.
However, if the audit manager wanted to give greater priority to another goal, then this
could be simply done by changing the weights allocated.
Do'stlaringiz bilan baham: |