If each of the observation \(x_1,x_2,.....,x_n\) is increased by ‘a’, where \(a\) is a negative or positive number, show that the variance remains unchanged.
CBSE Class 11 Maths · Statistics
CBSE Class 11 Maths · Statistics
Imagine every student in a class gets 5 extra marks added to their score. The average score of the class also goes up by 5. But here is the clever part — every student is still the same distance away from the new average as before. The gaps between scores never change. Variance measures only those gaps, not the actual scores. So adding the same number to everyone leaves the variance completely untouched.
Answer: Since the deviations \( y_i - \bar{y} = x_i - \bar{x} \) remain the same after adding \( a \), we get \( \sigma_y^2 = \sigma^2 \). Therefore the variance stays unchanged when each observation is increased by any positive or negative number \( a \).
Variance measures how spread out the observations are around their mean. When we add the same number \( a \) to every observation, the whole data set shifts by \( a \), and the mean also shifts by \( a \). Since each observation and the mean move together by the same amount, the distance of each observation from the mean stays exactly the same. Because variance depends only on these distances (deviations from the mean), it does not change. This idea is called the shift invariance of variance.
The variance of observations \( x_1, x_2, \ldots, x_n \) is \[\sigma^2 = \frac{1}{n} \sum_{i=1}^{n} (x_i - \bar{x})^2\] where the mean is \[\bar{x} = \frac{1}{n} \sum_{i=1}^{n} x_i\]
Since the deviations \( y_i - \bar{y} = x_i - \bar{x} \) remain the same after adding \( a \), we get \( \sigma_y^2 = \sigma^2 \). Therefore the variance stays unchanged when each observation is increased by any positive or negative number \( a \).
A common mistake is to forget that the mean also increases by \( a \). Students sometimes compute \( (x_i + a) - \bar{x} \) instead of \( (x_i + a) - (\bar{x} + a) \), which wrongly suggests the variance changes. Always update the mean to \( \bar{x} + a \) before finding the new deviations.
Since the deviations \( y_i - \bar{y} = x_i - \bar{x} \) remain the same after adding \( a \), we get \( \sigma_y^2 = \sigma^2 \). Therefore the variance stays unchanged when each observation is increased by any positive or negative number \( a \).
One common mistake: Students sometimes think that adding a positive number \( a \) to every observation must increase the variance, confusing the shift in the mean \( \bar{y} = \bar{x} + a \) with a change in spread. In reality, since both the observations and the mean shift by exactly the same amount \( a \), every deviation \( y_i - \bar{y} \) remains equal to \( x_i - \bar{x} \), so the variance is completely unchanged.
Answer: Since the deviations \( y_i - \bar{y} = x_i - \bar{x} \) remain the same after adding \( a \), we get \( \sigma_y^2 = \sigma^2 \). Therefore the variance stays unchanged when each observation is increased by any positive or negative number \( a \).
Question: A data set consists of observations \( y_1, y_2, \ldots, y_n \) with mean \( \bar{y} \) and variance \( \sigma^2 \). Each observation is decreased by a fixed positive number \( k \). Show that the variance of the new data set is still \( \sigma^2 \).
Solution: Let the original observations be \( y_1, y_2, \ldots, y_n \) with mean \( \bar{y} \) and variance \( \sigma^2 \). The original mean is \[\bar{y} = \frac{1}{n} \sum_{i=1}^{n} y_i\] and the original variance is \[\sigma^2 = \frac{1}{n} \sum_{i=1}^{n} (y_i - \bar{y})^2.\] Now define the new observations \( z_i = y_i - k \) for every \( i = 1, 2, \ldots, n \), where \( k > 0 \) is a fixed constant. Step 1: Find the new mean \( \bar{z} \). \[\bar{z} = \frac{1}{n} \sum_{i=1}^{n} z_i = \frac{1}{n} \sum_{i=1}^{n} (y_i - k) = \frac{1}{n} \sum_{i=1}^{n} y_i - k = \bar{y} - k.\] Step 2: Find the deviation of each new observation from the new mean. \[z_i - \bar{z} = (y_i - k) - (\bar{y} - k) = y_i - k - \bar{y} + k = y_i - \bar{y}.\] The deviation is exactly the same as before subtracting \( k \). Step 3: Compute the variance of the new data set, call it \( \sigma_z^2 \). \[\sigma_z^2 = \frac{1}{n} \sum_{i=1}^{n} (z_i - \bar{z})^2 = \frac{1}{n} \sum_{i=1}^{n} (y_i - \bar{y})^2 = \sigma^2.\] Therefore the variance of the new data set equals \( \sigma^2 \), which is unchanged.
Answer: The variance remains \( \sigma^2 \) (unchanged) after decreasing every observation by \( k \).
Question: Five students scored marks \( 4, 8, 12, 16, 20 \) in a quiz. Each mark is increased by \( 5 \). Verify that the variance of the new marks is the same as the variance of the original marks.
Solution: Original observations: \( 4, 8, 12, 16, 20 \). Here \( n = 5 \). Step 1: Compute the original mean \( \bar{x} \). \[\bar{x} = \frac{4 + 8 + 12 + 16 + 20}{5} = \frac{60}{5} = 12.\] Step 2: Compute the original variance \( \sigma^2 \). Deviations from the mean: \( (4 - 12) = -8 \), \( (8 - 12) = -4 \), \( (12 - 12) = 0 \), \( (16 - 12) = 4 \), \( (20 - 12) = 8 \). Squared deviations: \( 64, 16, 0, 16, 64 \). \[\sigma^2 = \frac{64 + 16 + 0 + 16 + 64}{5} = \frac{160}{5} = 32.\] Step 3: Form the new observations by adding \( 5 \) to each mark. New observations: \( 9, 13, 17, 21, 25 \). Step 4: Compute the new mean \( \bar{x}' \). \[\bar{x}' = \frac{9 + 13 + 17 + 21 + 25}{5} = \frac{85}{5} = 17.\] Note that \( \bar{x}' = \bar{x} + 5 = 12 + 5 = 17 \), as expected. Step 5: Compute deviations of the new observations from the new mean. Deviations: \( (9 - 17) = -8 \), \( (13 - 17) = -4 \), \( (17 - 17) = 0 \), \( (21 - 17) = 4 \), \( (25 - 17) = 8 \). These deviations are identical to the original deviations. Step 6: Compute the new variance \( \sigma'^2 \). \[\sigma'^2 = \frac{(-8)^2 + (-4)^2 + 0^2 + 4^2 + 8^2}{5} = \frac{64 + 16 + 0 + 16 + 64}{5} = \frac{160}{5} = 32.\] Since \( \sigma'^2 = \sigma^2 = 32 \), the variance is unchanged after adding \( 5 \) to each observation.
Answer: Original variance \( = 32 \); New variance \( = 32 \). The variance is unchanged.
Question: If each of the observations \( x_1, x_2, \ldots, x_n \) is multiplied by a non-zero constant \( k \), show that the variance of the new observations is \( k^2 \) times the original variance.
Answer: When each observation \( x_i \) is multiplied by a non-zero constant \( k \), the new mean becomes \( \bar{y} = k\bar{x} \) and the deviation of each new observation satisfies \( y_i - \bar{y} = k(x_i - \bar{x}) \). Therefore the new variance is \( \sigma_y^2 = k^2 \sigma^2 \), i.e. the variance gets multiplied by \( k^2 \). This is different from the original result: multiplying by a constant changes the variance, unlike adding a constant which leaves it unchanged.