9. Sales forecasting Flashcards
<p>How is Centering defined?</p>
<p>A method used in the calculation of a moving average where the average is plotted or calculated in relation to the central figure.</p>
<p>How is Correlation defined?</p>
<p>The relationship between two sets of variables.</p>
<p>How is Correlation Coefficient defined?</p>
<p>A measure of the extent of the relationship between two sets of variables.</p>
<p>How is Moving Average defined?</p>
<p>A succession of averages derived from successive segments (typically of constant size and overlapping) of a series of values.</p>
<p>How is Scatter Graph defined?</p>
<p>A graph showing the performance of one variable against another independent variable on a variety of occasions. It is used to show whether a correlation exists.</p>
<p>How is Time Series Analysis defined?</p>
<p>A method that allows a business to predict future levels from past figures.</p>
<p>What are the four main components of Time Series Data?</p>
<p>- Trend
<br></br>- Seasonal Fluctuations
<br></br>- Cyclical Fluctuations
<br></br>- Random Fluctuations</p>
<p>How can you identify the trends?</p>
<p>- Moving Average --> to calculate this you add the number or sales up per year/quarter etc. then divide by the same amount you have added up
<br></br>- four-period centred moving average --> can be used by finding the midpoint of a four-year moving total and a eight-year moving total.
<br></br>- to work this out divide the eight year moving total by eight</p>
<p>How can you predict future trends from a graph?</p>
<p>- Using a line of best fit
<br></br>- by plotting the values from the four-period moving average onto a graph accurately and then adding the line of best fit 'by eye', so that points fit equally either side of the line.
<br></br>- This line can be extended should help give a reasonable prediction</p>
<p>What equations can you use to help draw the line of best fit?</p>
<p>- when drawing a line of best fit it should pass through the coordinates (X,Y), where X is the average of the years and Y is the average sales:
<br></br>
<br></br>- X = (the sum of the years)/ (number of years)
<br></br>- Y = (the total sales in the trend)/ (the number of years)</p>
<p>How can we calculate the variation from the trend?</p>
<p>- predictions may not be accurate because it is taken from the trend, and the trend 'smoothed out; variations in sales.
<br></br>- To make an accurate prediction the business will have to fine the average variation over the period and take this into account.
<br></br>- To find out how much variation there is we must --> V= Actual Sales - Trends</p>
<p>How can we use trends from Seasonal Variation to make more accurate predictions?</p>
<p>- trends are the smoothed out figures that smoothed out the variation.
<br></br>- If a set of data has been sorted with four-period centred average using the financial Quarters in the year we can make a more accurate prediction by calculating the average seasonal variation in the each quarter ( e.g. average of the 4th Quarters from the data etc.)</p>
<p>When is Quantitative Sales Forecasting likely to be more reliable?</p>
<p>- The forecast is for a short period of time in the future, such as six months, rather than a long time. such as five years
<br></br>- The are revised frequently to take account of new data and other informaiton
<br></br>- The market is slow changing
<br></br>- Market research data, including test marketing data, is avaible
<br></br>- Those preparing the forecast have a good understanding of how to use data to produce a forecast
<br></br>- Those preparing the forecast have a good 'feel' for the market and can adjust the forecast to take account of their hunches and guesses about the future</p>
<p>How can some forecasting account for changes even if the forecast should be reliable?</p>
<p>- by creating a forecast range
<br></br>- by preparing 3 set of figures --> a optimistic, a pessimistic and a central forecast
<br></br>- these give the best and worst case scenarios aswell as the central forecast, the one that is most likely to occur
<br></br>- this give the departments in a business an indication of the possible variations they might have to face.</p>
<p>How can we use Causal Modelling with Time Series Analysis?</p>
<p>- Time series analysis only describes what is happening to information
<br></br>-Casual modelling tries to explain data, usually by finding a link between on set of data and another
<br></br>- this can be done by plotting the two variables on a scatter graph and looking at the correlations which can be show better with a line of best fit</p>
