A. Planning, Budgeting, and Forecasting Flashcards
What is the purpose of a “budget”? Is a budget more quantitative or qualitative focused?
An operational plan and a control tool for an entity that identifies the resources and commitments needed to satisfy the entity’s goals over a period.
Budgets are primarily quantitative as opposed to qualitative.
They set specific goals for income, cash flows, and financial position.
What is a “pro forma statement”?
A budgeted financial statement based on historical documents that is adjusted for events as if they had occurred.
Describe the four (4) phases of the “budget cycle”.
Budget Cycle:
(1) “Budget Creation & Agreement” - budget is created that addresses entity as a whole as well as its subunits. All managers agree to fulfill their part of the budget.
(2) “Performance Evaluation” - budget is used to test current performance against expected performance.
(3) “Corrective Action” - variations from the plan are examined, and corrective actions are taken when possible.
(4) “Feedback & Revision” - feedback is collected, and the plan is revisited and revised if needed.
What are the four (4) main reasons for budgeting?
Four (4) main reasons for budgeting:
(1) Planning - provides framework to achieve the goals of the organization
(2) Communication & Coordination - forces different parts of the organization to communicate their plans and needs to each other.
(3) Monitoring - sets standards / performance indicators by which managers can monitor the organization’s progress in meeting its goals
(4) Evaluation - serve as guides / instruments for employee evaluations
What are some characteristics of successful budgeting?
Budgets should be:
- Aligned with the corporate strategy
- Kept separate, but should flow from the strategic planning (higher-level, longer-term, and structured in companywide terms) and forecasting processes (lower accountability than a budget, not approved by management, and often are not formally analyzed against variances)
- Used to alleviate potential bottlenecks and to allocate funds to those areas that can use resources most efficiently
- Management must endorse the budget
- Characterized as a motivating tool to help employees work towards organizational goals
- Seen as an internal control device
- A higher authority than the team that developed the budget must review and approve the budget
- The final budget should not be easily changed, but be flexible enough to be useful
What are the characteristics of a successful budget process?
Characteristics include:
(1) Budget period - most commonly used period is the fiscal year, but shorter or longer durations of time may be used
(2) Participants in the budget process:
- “Authoritative Budgets” (top-down budgets) - top management sets everything related to the budget and expects lower management and employees to adhere to the budget and meet goals
- “Participative Budgets” (bottom-up budgets) - managers at all levels cooperate to set budgets for their areas and top management retains final approval
- The ideal process contains a combination of authoritative and participative budget approaches
(3) Basic steps in budgeting
a. Budget proposal
b. Budget negotiation
c. Budget review and approval
d. Budget revision
(4) Use of cost standards
- A “standard” is any carefully determined price, quantity, service-level, or cost.
- Include “authoritative” (determined solely by management) or “participative” standards (determined through dialogue between management and all other involved parties)
- Sources for Standards Settings: activity analysis, historical data, market expectations, strategic decisions, benchmarking
What is used to determine proper resource allocation among competing opportunities in the budgeting process?
The overall “strategy” of the corporation determines the allocation of scarce resources among competing opportunities.
What is a “master budget” and what are its components?
- The overall plan for operations for a company or SBU over a year, an operating period, or a shorter duration.
- Sets quantitative goals for all operations, including detailed plans for raising capital
- A map showing where the company is heading. Should be designed and “heading” towards company strategy
- Broken down further in 3 parts:
1) Operating budget
2) Financial budget
3) Capital budget
What are three (3) quantitative methods that are used to plan for the future (i.e. forecast)?
1) Data analysis
- Involves analyzing a given set of data to establish the relationship and/or pattern in data
- Can be used to predict or forecast outcomes based on given set of circumstances
2) Model building:
- Involves creating mathematical models to establish relationships between different factors
3) Decision Theory:
- Deals with uncertainty by looking at various potential outcomes that can happen in the future, along with the likelihood of these outcomes
What is a “regression analysis”? What are its assumptions? What is its formula? What are some objective benchmarks that allow users to evaluate the reliability of the regression equation?
A statistical method used to determine the impact that one variable (or group of variables) has on another variable. Provides the best linear, unbiased estimate of the relationship between the dependent variable (Y) and one or more independent variables (X or X’s).
Assumptions used in linear regression:
- Linearity - relationship between X (or X’s) and Y is linear
- Stationary - process underlying relationship is stationary
- Independent variables (X’s) in a multiple regression are independent of one another
- Differences between actual and predicted values of dependent variable are normally distributed with a mean of zero and a constant deviation (i.e. the dependent variable is not correlated with itself)
Terms to know:
- Simple regression analysis: 1 independent variable
- Multiple regression analysis: 2+ independent variables
Formula(s): Y = a + b1X1 + b2X2 + b3X3 + …. bnXn
where:
Y = dependent variable
a = the Y-intercept (value of Y when X=0)
b = regression coefficient / slope of the line
X = value for the independent value
Regression analysis equations systematically reduce estimation errors and are therefore called “least squared regression”. Regression analysis fits a line (least squared line) through all data points, which minimizes the difference between the line (prediction) and the data point (actual). The statistical formula that is used produces the least amount of error between these two items (i.e. predicted and actual values).
Objective benchmarks:
a) R-squared:
- Value between 0 and 1
- Indicates the degree to which changes in the dependent variable can be predicted by changes in the independent variable / the % of the variation in the dependent variable that can be accounted for by the variability in the independent variable (0 = no predictability / little correlation; 1 = high predictability / complete correlation)
b) T-value:
- Measures whether an independent variable (X) has a valid, long-term relationship to a dependent variable
- Generally should be more than 2
c) Standard error of the estimate:
- Measures the dispersion around the regression line
- Allows users to assess the accuracy of the prediction
- Higher dispersion = Higher unpredictability
Disadvantages:
- Assumes relationship will hold into the future
- Could be outliers that skew the data set
- Is the relationship between the dependent and independent variable really all that reasonable
What is a “time series analysis”? What are the four (4) components that combine to provide the overall pattern in a time series analysis?
A series of measurements of a variable taken at any time interval (hour, day, month, etc) with the objective of finding patterns in the data that can aid in making forecasts or predictions of future values.
Essentially a regression analysis, but uses “time” as the independent variable (Y).
Four (4) components that combine to provide the overall pattern in a time series analysis:
(1) Trend: gradual shits to higher or lower values
(2) Cycles: cyclical fluctuations in the data set that are explained by cycles in the economy
(3) Seasonality:
(4) Irregular Variations - random / unexplained variations
Benefits:
- Helps analysts see patterns that are clearly upward / downward so managers can make decisions to expand or exit the business or product market
Disadvantages - relies on past data that assumes will continue into the future
What is “smoothing” in terms of forecasting techniques? What are the three (3) smoothing methods?
An analytical method that levels out the random fluctuations from the irregular component of the time series. Effective for time series that do not exhibit significant patterns due to trend, cyclical, or seasonal effects. Generally very accurate for short-range forecasts.
Three “smoothing” methods:
(1) Moving Averages:
- Uses the average of the most recent data value set of a given time period
- Formula: Moving Avg = (Sum of Data Values for a Time Period / Time Periods)
(2) Weighted Moving Averages:
- Same as moving averages, but gives more weight to most recent data
- Formula: Weight. Moving Avg = (% x A) + (% x B) + … (% x Z)
(3) Exponential Smoothing:
- Uses a weighted average of past time series, selected only one weight–that of the most recent set of data. The weights of the other data values are automatically computed, getting smaller as the time period moves farther into the past.
- Formula: F(t+1)= aY(t) + (1 - a)*F(t), where:
- F(t+1) = forecast for the time period (t+1), where “t” is the current time period; this is the objective of the current forecast
- Y(t) = actual value of the time series in period “t”. Ex. actual sales for Q4
- F(t) = forecast of the time period for period “t”. This was the objective of the previous forecast. Ex. sales that were forecasted for Q4
- a = smoothing constant (0 < a < 1). Derived through trial and error on an initial data set to determine what value of “a”, which must be between 0 and 1, results in the forecast closest to the actual sales for the period
- Has minimal data requirements and therefore is a good smoothing method to use when forecasts are required for large numbers of dependent variables
- Requires only two pieces of information: (1) the forecast and (2) actual values for the last (current) period
- Will not work as well when there are significant trends or variations
What is a “learning curve analysis”? What are the two (2) methods used to measure the learning curve?
A systematic method for estimating costs based on increased learning by the business, group, or individual, which allows them to become more efficient at completing tasks. The idea is that costs will decrease as learning increases due to increased efficiencies.
Calculation of the learning curve is based on the “learning rate”, which is the % by which average time decreases from the previous level as output doubles. (ex. It takes a worker 10 hours to assemble the first unit, but only 8 hours to assemble the second unit due to his/her learning experience. The “learning rate” = 80%)
Two ways to measure the learning curve:
(1) Incremental unit-time learning model (non-accepted method)
(2) Cumulative average-time learning model (accepted method)
Incremental Unit-Time Learning Model (non-preferred):
- Measures increased efficiency by adding the incremental time for each unit to the previous total time. “Average time per unit” is then calculated by diving total time by the number of units
- Formula: (Total Time Spent to Make Units) / (Total Units Made)
- Steps:
a) Determine the number of widgets produced
b) Calculate the hours it takes to produce each widgemultiplying the time it takes for the incremental unit by the learning rate
c) Calculate the “cumulative total time” - Ex. Using the “learning rate” example above, the learning curve value = 9 hours per unit ([10 hours + 8 hours] / 2 units)
Cumulative Average-Time Learning Model (preferred):
- Like the method above, determines the “cumulative average time per unit” by multiplying the total time for the “x”th unit by the “learning rate”
- Calculates “cumulative total time” by multiplying the incremental unit by the “cumulative average time per unit”
- Formula: Learning Curve = (Incremental Unit) x (Cumulative Avg Time per Unit)
- Steps:
a) Determine the number of units produced
b) Calculate the “cumulative average time per widget” by multiplying the time it takes for the incremental unit by the learning rate
c) Calculate the “cumulative total time” - Ex. Going off the example from above,
What is an “expected value” analysis?
A type of forecast that takes into consideration expected economic conditions and applies those expectations to the (sales) forecast.
Formula: EV = S x P(x), where:
- EV = Expected value
- S = amount associated with a specific outcome
- P(x) = probability associated with a specific outcome
Calculation: Multiply the outcome of each possible outcome by the probability associated with that possible outcome, and then add all of them together.
Benefits:
- Helps the organization decide whether to undertake certain actions
Limitations:
- Only as good as the estimated potential outcomes for each scenario and the probability assigned to each scenario
- Assumes that the decision maker is risk neutral
What is a “sensitivity analysis” forecast?
Used by decision makers to help decide what changes in a given situation (“state of nature”) are most likely to produce a particular outcome (“payoff”).
Can be used to study impacts on both quantifiable or qualitative states of nature and payoffs.
A form of “what-if” analysis because it is conducted by changing a specific variable (i.e. an input) and determining how sensitive the outcome is to the changes in the input.
Benefits:
- Shows managers how susceptible the outcomes of decisions are to changes in any parameter or estimate.
