63
Copyright © Lexibook 007
ENGLISH
As a memorandum
We have n data on a sample of measures, results, people, objects…
Each data is made up of a value (one variable x) or two (two variables
x and y). We look to calculate the average of these values and the
breakdown of these values around the average, the standard deviation.
This data is calculated in the following way:
∑x = x
1
+x
+x
+....x
n-1
+x
n
∑x
= x
1
+x
+x
+....x
n-1
+x
n
∑xy = x
1
y
1
+x
y
+x
y
+....x
n-1
y
n-1
+x
n
y
n
Average
Standard deviation of the sample for x:
Standard deviation of the population for x:
Variance V = s2 or 2
When there are two variables we try to find a link in the data provided
between x and y. We study the simplest solution: a relation such as
y=a+bx.
cov(x,y) is the covariance :
The validity of this hypothesis is confirmed in the following calculation:
Referred to as the co-efficient of the linear correlation. The result is
always between –1 et +1 and a result greater than or equal to √3/2 in
absolute value is considered true.
5. STATISTICS
Preliminary Comments
SC460IM0237.indb 63
20/6/07 15:28:10
