Forecasting Technique Part1

ACCA F2 - Management Accounting

Forecasting is the estimate of future position based on present and historical position taking account of foreseeable factors that influence cost, revenue and performance. All that mathematical tools (Go to article "Business Mathematics") used to determine past and current performance are equally important in determining future position. In this section, we go through some more mathematical/statistical tools used in forecasting.

Forecasting starts by assigning fixed values to some determinants like sales volume, cost per sales and sales revenue which are established and justified by trend, fluctuations and prudence estimation of product and market position. Use of probability, regression, correlation, trend analysis and variance analysis is common in forecasting.

Probability and matrix: We see in article titled 'managerial process and styles' that managerial task comprises with making choices out of available alternatives; these alternatives are surrounded by uncertainty. Therefore, managers use probability and payoff to match their risk exposure to expected return.

Regression analysis is widely used for forecasting. It establishes relation between dependent variables and independent variables. E.g. Prediction of sales volume with the growth in per capita income. Decrease in price of product with increase in supply. A simple liner regression equation is presented here.

Y = a + b x            (where value of Y depends on value of X and a is a constant)

For forecasting past data values for X and Y are used to calculate the value of "a" and "b".

For "n" sample size, b=(n∑xy-∑x∑y)/(n∑x²-(∑x)²) and  a=(∑y/n)-(b∑x/n)               

Once values of "a", "b" and "x" are established, then past trend is projected to forecast for upcoming intervals/periods.

A multiple linear regression equation looks like Y = a + b₁ X₁ + b₂  X₂ +…… + b_n X_n

This technique is mostly suitable for forecasting to businesses in stable and slowly changing sector and market. The forecast can be unrealistic if used to rapidly changing, dynamic, turbulent, and innovative business sectors.

Installation of analysis tool pack is essential for using Excel 2007 to calculate regression.

Ø  Office button > Excel option > Add-Ins > Excel Add-Ins > Go > Check - "Analysis Toolpack"

Then copy input values to excel sheet, go to Data Ribbon  > Analysis > Data Analysis > select Regression and fill in the inputs required.