0:00 / 0:00

Transforming Data

The measure of central tendency and measure of spread could change if we shift or scale the values in a given data set.

Shifting: adding or subtracting the same constant cc to every data value

3  3 +c3\ \rightarrow\ 3\ +c


Scaling: multiplying or dividing every data value by the same constant dd

3  3d3\ \rightarrow\ 3d


Shifting & Scaling: multiplying/dividing every data value by the same constant dd and adding/subtracting the same constant cc

3  3d+c3\ \rightarrow\ 3d+c


Example

25, 60, 65

PAGE BREAK
What happens to the data set?




Watch Out!
When we talk about transforming the spread, we are talking about range, IQR, and standard deviation, but NOT variance!
If dd is a negative value, multiply the spread by the absolute value d\mid d\mid.

PAGE BREAK

Example


Mean(y+5)=10+5=15Mean(3y)=3y=3(10)=30Mean(3y+5)=3(10)+5=35Mean(y+5) = 10+5 = 15 \\Mean(3y) = 3y = 3(10)=30 \\Mean(3y+5)=3(10)+5=35





Median(y+5)=8+5=13Median(3y)=3y=3(8)=24Median(3y+5)=3(8)+5=29Median(y+5) = 8+5 = 13 \\Median(3y) = 3y = 3(8)=24 \\Median(3y+5)=3(8)+5=29





SD(y+5)=6.325SD(3y)=3y=3(6.325)=18.975SD(3y+5)=3(6.325)=18.975SD(y+5) = 6.325 \\SD(3y) = 3y = 3(6.325)=18.975 \\SD(3y+5)=3(6.325)=18.975





PAGE BREAK



You multiply all values in a data set by 55 and then add four. The resulting mean is 34 and the resulting standard deviation is 25.

Determine the mean and standard deviation of the original data.



0:00 / 0:00

Linear Transformation

Given a linear equation Y=c+dXY=c+dX where cc is the shifting constant and dd is the scaling constant, then:

μY=c+dμXσY2=d2σX2σY=dσX\mu_Y=c+d\mu_X \\\sigma^2_Y=d^2\sigma^2_X \\\sigma_Y=|d| \sigma_X

Wize Concept
The standard deviation σ\sigma is the square-root of the variance σ2\sigma^2.


Example
Suppose that a data set has a mean of 8 and a standard deviation of 3. You wish to multiply all of the numbers by 4 and then subtract 3.
Determine the new mean and standard deviation.

We are transforming XYX\rightarrow Y
Let XX represent the first data set and let YY represent the new data set.
Then Y=4X3Y=4X-3
  • New mean: Y=4(8)3=29\overline{Y}=4(8)-3=29
  • New standard deviation: sY=4×3=12s_Y=4\times3=12


0:00 / 0:00

Example: Linear Transformation

Jenny is a hairdresser at Swish Salon. She pays the salon a fixed fee of $5,000 to rent a chair (this is subtracted from her profit). In addition, the salon gets 15% of her revenue (which means she keeps 85% of her revenue). Suppose Jenny makes an average of $60,000 in revenue with a standard deviation of $8,000. What is the mean and standard deviation of her profit?

Hint: Profit = Revenue - Costs

Y=profitY=profit
X=revenueX=revenue

c=5000c=-5000 (fixed cost, subtracted from profit)
d=0.85d=0.85 (she keeps 85% of the revenue because the salon gets 15% of the revenue)

Profit=0.85(Revenue)5000Profit = 0.85(Revenue) - 5000
Y=0.85X5000Y=0.85X-5000

Mean(profit):
μY=c+dμXμY=5000+0.85(60,000)=46,000\mu_Y=c+d\mu_X \\ \mu_Y=-5000+0.85(60,000)=46,000

SD(profit):
σY=dσXσY=0.85(8000)=6800 \\\sigma_Y=|d| \sigma_X \\ \sigma_Y=0.85(8000)=6800



You are working overseas in Tokyo in as a consultant. Your average annual earning is ¥5,000,000 with a standard deviation of ¥1,500,000. The exchange rate is ¥1 = $0.009 USD. Each year, you must pay your agency a fixed fee of $10,000 USD.

Find the mean and standard deviation of your annual net earnings in US dollars.