Suppose x1, x2, … , xn are n observations with mean as x.
By definition of mean, [i.e. The mean or average of observations, is the sum of the values of all the observations divided by the total number of observations]
We have,
\(X=\frac{X_1X_2\,+.....+X_n}{n}\)
nx = x1 + x2 + … + xn …[1]
So,
in this case we have assumed mean(a) is equal to mean of the observations(x)
And we know that
di = xi - a
where,
di is deviation of a (i.e. assumed mean) from each of xi i.e. observations.
So,
In the above case we have
d1 = x1 - x
d2 = x2 - x
.
.
.
dn = xn - x
and sum of deviations
d1 + d2 + … + dn = x1 - x + x2 - x + … + xn - x
= x1 + x2 + … + xn - (x + x + … {upto n times})
= nx - nx [Using 1]
= 0
Hence,
sum of deviations is zero.