EXAM TIME TABLE 2018 -- +2 +1 10TH

Tamilnadu School Books Change for Academic Years


Academic year

2, 7, 10 & 12


3, 4, 5 & 8


STD - 1
STD - 2
STD - 3
STD - 4
STD - 5
STD - 6
STD - 7
STD - 8
STD - 9
STD - 10
STD - 11
STD - 12


FORECAST Function Dictionary

Function Dictionary Function Examples Function Categories

Month Sales
1 £1,000
2 £2,000
3 £2,500
4 £3,500
5 £3,800
6 £4,000
    Type the month number to predict :  12
    The Forecast sales figure is :  £7,997  =FORECAST(E11,F4:F9,E4:E9)
What Does It Do ?            
This function uses two sets of values to predict a single value.
The predicted value is based on the relationship between the two original sets of values.
If the values are sales figures for months 1 to 6, (Jan to Jun), you can use the function
to predict what the sales figure will be in any other month.
The way in which the prediction is calculated is based upon the assumption of a Linear Trend.
ItemToForecast is the point in the future, (or past), for which you need the forecast.
RangeY is the list of values which contain the historical data to be used as the basis
of the forecast, such as Sales figures.
RangeX is the intervals used when recording the historical data, such as Month number.
No special formatting is needed.
The following table was used by a company considering expansion of their sales team.
The Size and Performance of the previous teams over a period of three years were entered.
The size of the New Sales team is entered.
The =FORECAST() function is used to calculate the predicted performance for the new sales
team based upon a linear trend.
Year Size Of
Sales Team
1996 10 £5,000
1997 20 £8,000
1998 30 £8,500
    Size Of The New Sales Team : 40
    Estimated Forecast Of Performance : £10,667  =FORECAST(E43,E39:E41,D39:D41)


Popular posts from this blog

Tamilnadu SIXTH all books free download, Tamilnadu 6th all books free download

Tamilnadu plus1 all books free download, Tamilnadu 11th all books free download