Forecasting

Question 1

What is forecasting a budget?

Answer 1

Budgets are based on forecasts
Forecasts are used to provide realistic assumptions for budgets
Forecasting can be done using very simple assumptions
More complex forecasting techniques and models will provide more reliable forecasts

Question 2

What is linear regression analysis?

Answer 2

A statistical technique for identifying “a line of best fit” from a set of data

It is based on the following assumptions:
there is a linear relationship between two variables represented by x and y
this relationship in the future can be predicted from the relationship in the past

Question 3

How do you calculate the line of best fit in linear regression analysis?

Answer 3

Line of best fit:
y = a + bx

Where:
y = value of dependent variable
x = value of independent variable
a is the intercept on y axis
b is the gradient
a and b are values obtained from a statistical analysis of historical data for values of x and y

Question 4

How do you calculate least squares regression analysis?

Answer 4

B= (n∑xy – ∑x∑y) / n∑x2 – (∑x)2

a= y – bx

Question 5

How can linear regression be used in budgeting?

Answer 5

Linear Regression can be used to make forecasts or estimates whenever:
a linear relationship is assumed between 2 variables
historical data is available

Can be used for example:
Sales budget
Total costs

Question 6

What are the limitations of linear regression?

Answer 6

Assumes a linear relationship between variables
Only measures relationship between two variables
Only interpolated forecasts tend to be reliable
Assumes history will repeat itself
Predictions only reliable if there is a significant correlation between the data

Question 7

What is correlation?

Answer 7

It is possible to calculate the strength of the connection (or correlation) between two variables. The stronger the connection, the more reliable the line of best fit should be for forecasting.

The strength of the connection can be measured by a correlation coefficient, r

Question 8

What is correlation coefficient?

Answer 8

Value of r must lie between –1 and +1
When r = +1 there is a perfect positive correlation between the values of x and y
When r = -1 there is a perfect negative correlation between the values of x and y
When r = 0 there is no correlation between the values of x and the values of y

Question 9

How do you calculate correlation coefficient?

Answer 9

r= (n∑xy – ∑x∑y) / square root of ([n∑x2 – (∑x)2 ] [n∑y2 – (∑y)2 ])

Question 10

What is the correlation of determination?

Answer 10

r2
Indicates the proportion of the variations in the value of the dependent variable y that can be explained by variations in the value of the independent variable x

Question 11

What is inflation?

Answer 11

Need to adjust historical data and future forecasts for inflation
Historical data should be adjusted to the same index level for prices or costs
When a forecast is made an adjustment should be made for anticipated inflation in the forecast period

Question 12

What is time series analysis?

Answer 12

A time series is a series of figures relating to the changing value of a variable over time
The data often conforms to a certain pattern over time
This pattern can be extrapolated into the future and hence forecasts are possible
The actual data can be broken down into four components that are said to affect the results

Question 13

What are the components of a time series?

Answer 13

The Trend (T)
describes the long term general movement of the data
Cyclical Variations (C)
economic cycle of booms and slumps
Seasonal Variations (S)
a regular variation around the trend over a fixed time period, usually one year
Residual Variations (R ) 
irregular, random fluctuations in the data

Question 14

How do you calculate the time series additive forecasting model?

Answer 14

Actual = T + C + S + R

Question 15

How do you calculate the time series Multiplicative forecasting model?

Answer 15

Actual = T x S x C x R

Question 16

What is a moving average?

Answer 16

Used to calculate a trend line
A series of averages calculated from time series historical data
A moving average is associated with the mid-point of the time periods used to calculate the average
If the time period is even in number need to “centre” the moving average

Question 17

What are seasonal variations?

Answer 17

Once the trend has been established the seasonal variation can be determined

Question 18

How do you calculate the seasonal variations using the additive model?

Answer 18

Prediction = T + S

Question 19

How do you calculate the seasonal variations using the multiplicative model?

Answer 19

Prediction = T x S

Question 20

What is the additive method in seasonal variations?

Answer 20

Compare actual sales with the moving average value
For each season the seasonal variation is the difference between these two
If actual value is higher than the trend line value the seasonal variation should be added
If actual value is lower than the trend line value the seasonal variation should be deducted
An average variation for each season is calculated
Sum of seasonal variations has to be 0

Question 21

What is the multiplicative method in seasonal variations?

Answer 21

Compare actual sales with the moving average value
For each season the seasonal variation is calculated by expressing the actual sales for the period as a % value of the moving average figure
An average variation for each season is calculated
Multiplication of seasonal variation has to = 1

Question 22

What are the advantages of measuring seasonal variations?

Answer 22

Forecasts are based on clearly-understood assumptions
Trend lines can be reviewed after each successive time period and reliability of forecasts can be assessed
Forecasting accuracy can improve with experience

Question 23

What are the disadvantages of measuring seasonal variations?

Answer 23

Assumption that past is a reliable guide to the future
Assumption that straight-line trend exists
Assumption that seasonal variations are constant