Forecasting Cash Flows

1
Forecasting Cash Flows

• Chapter 11

2
Why Forecast?

• Liquidity Management
• Financial Control
• Variation of Actual from Forecast
• Strategic Objectives
• Capital Budgeting
• Cost Management
• Optimize Cash Usage
• Exchange Rate Exposure Management

3
Forecasting Concerns

• Forecasting Horizons
• Short-Term
• Daily
• Weekly
• Monthly
• Medium-Term
• Up to 1 year
• Long-Term
• Over 1 year

• Degrees of Certainty
• Certain
• Interest, Royalties
• Tax Payments
• Predictable
• Collections
• Payroll
• Vendor Payments
• Less Predictable
• New Product Sales, etc.

4
Selecting a Forecast Method

• Establishing Data Relationships
• Relationships between available data and cash
  flows.
• Past Experience
• Trends
• Operating Relationships
• Selecting a Method
• Accurate Cost Effective

5
Testing and Validation

• In-Sample Validation
• Data used in model development
• Out-of-Sample
• Historical data not used in model development
• On-Going Validation

6
Forecasting Methods

• Receipts Disbursements
• Kinda Like a Cash Budget
• Short Medium Term Forecasting
• Cash Receipts Schedule
• Collections, Maturing Investments, etc.
• Cash Disbursements Schedule
• Vendor Payments, Payroll, Interest

7
Receipts and Disbursements Method

• Steps in the Process
• Develop a Sales Forecast for the Period
• Range of Possible Values
• Estimate Receipts
• Cash Sales
• Collections from Credit Sales
• Other Cash Receipts
• Estimate Cash Disbursements for the Period
• Payroll
• Vendor Payments

8
Receipts and Disbursements Method

• Net Cash Flow Receipts - Disbursements
• Subtract the Target (or Minimum) Cash Balance to
  Obtain the Cash Surplus or Cash Shortfall.

9
Receipts and Disbursements Method
10
Forecasting Receipts from Credit Sales
11
Forecasting Receipts from Credit Sales
12
Forecasting Receipts from Credit Sales

• July Sales 20,250,000
• August Sales 19,500,000
• Forecasted September Sales 15,750,000
• Forecasted September Collections
• 0.25 x 20,250,000 5,062,500
• 0.50 x 19,500,000 9,750,000
• 0.25 x 15,750,000 3,937,500
• 18,750,000

13
Forecasting Methods

• Distribution Forecast
• Based on Historical Distributions of Cash Flows
• Averages
• Regression
• Estimate Percentage of Cash Inflows or
  Disbursements on Particular Days or Weeks
• Short-Term Forecasting

14
The Distribution Method of Forecasting

• Forecasted September Receipts
  18,750,000
• You have determined that the monthly receipts are
  typically distributed in the following manner
• Week 1 - 26.7
• Week 2 - 13.3
• Week 3 - 32
• Week 4 - 28

15
The Distribution Method of Forecasting

• The Weekly Receipts Forecast Is
• Week 1 0.2667 x 18,750,000 5,000,000
• Week 2 0.1333 x 18,750,000 2,500,000
• Week 3 0.32 x 18,750,000 6,000,000
• Week 4 0.28 x 18,750,000 5,250,000

16
(No Transcript)
17
Forecasting Methods

• Percentage of Sales
• Determine the Relationships between Sales Levels
  and Asset Accounts
• Express Asset Accounts that Vary with Sales as a
  Percentage of Sales
• Multiple Percentages and Forecasted Sales
• Long-Term Forecasting

18
Percentage of Sales Forecasting
19
Percentage of Sales Forecasting
20
Time Series Forecasting

• Simple Moving Average
• Rolling Average of Past Historical Values
• Will Lag Trends in Actual Cash Flow
• Exponential Smoothing
• Ft1 axt (1 - a)Ft

21
Moving Averages

• Moving Averages Smooth the Past History Data
• The Moving Average Eliminates Some of the
  Randomness in the Data
• Each Average is Computed by Dropping the Oldest
  Observation and Including the Next Observation

22
Moving Averages
23
Moving Averages

• Using Moving Average Smoothers to Estimate the
  Trend-Cycle
• A Larger Number of Observations Included in the
  Moving Average will Result in a Smoother
  Trend-Cycle

24
(No Transcript)
25
The Data with a 5 MA Smoother
26
Exponential Smoothing

• Allows a Greater Weight for More Recent
  Observations
• F t1 (? x Yt) (1 - ?) x (Ft)
• ? Smoothing Constant
• 0 lt ? lt 1
• A smoothing constant closer to 1.0 places more
  weight on the most recent actual observation.
• A smoothing constant closer to 0 places more
  weight on the most recent forecast.

27
Exponential Smoothing
28
Regression

• Linear Regression Fits a Trend Line Through the
  Data
• The intercept and slope of the line are chosen to
  minimize the sum of the squared error terms.
• Yt a b(Y t-1) e
• Linear Regression is Used to Find the Optimal
  Values of a and b.

29
Regression Models
30
Regression Models
31
Regression Models