Comprehensive Biostatistics Notes: Summation Rules, Frequency Distributions, and Descriptive Measures

Overview of Statistical Data and Summation Notation

A dataset is defined as a collection of numerical measurements representing observed values of a variable. Symbolically, a dataset containing nn observations is denoted as x1,x2,x3,…,xnx_1, x_2, x_3, \dots, x_n. The total sum of these nn numerical observations is written as:

∑i=1nxi=x1+x2+x3+⋯+xn\sum_{i=1}^n x_i = x_1 + x_2 + x_3 + \dots + x_n

For example, consider a specific dataset consisting of 55 distinct measurements: 33, 3.53.5, 44, 66, and 3.73.7. Representing these observations symbolically yields x1=3x_1 = 3, x2=3.5x_2 = 3.5, x3=4x_3 = 4, x4=6x_4 = 6, and x5=3.7x_5 = 3.7. To compute the sum of these 55 numbers, the addition operation is expressed as:

∑i=15xi=x1+x2+x3+x4+x5=3+3.5+4+6+3.7=20.2\sum_{i=1}^5 x_i = x_1 + x_2 + x_3 + x_4 + x_5 = 3 + 3.5 + 4 + 6 + 3.7 = 20.2

In statistical analysis, summation operations are performed continuously across datasets. To avoid repeatedly writing the addition symbol (++) between terms, the capital Greek letter sigma (Σ\Sigma) is universally adopted as a shorthand mathematical operator to signify the operation of summation. The index variable ii indicates the term identifier, while the lower and upper limits placed at the bottom and top of the sigma symbol define the starting term and ending term of the summation, respectively.

Double Subscript Notation and Matrix Summation

When a variable or characteristic is classified across two distinct dimensions or factors, a double subscript notation YijY_{ij} is employed. In this notation, the primary subscript ii identifies the group, row, or category, while the secondary subscript jj identifies the individual observation or item within that specific group.

Consider an agricultural biostatistics study tracking milk yield across multiple cattle breeds. Let ii denote the breed index (i=1,2,…,ni = 1, 2, \dots, n) and let jj denote the individual cow index within a breed (j=1,2,…,kj = 1, 2, \dots, k). If milk yield is evaluated across 44 breeds (n=4n = 4) with 55 cows observed per breed (k=5k = 5), the variable YijY_{ij} represents the milk yield produced by cow number jj belonging to breed ii. The collected observations are systematically organized in a rectangular data matrix containing rows for breeds and columns for cows:

Breed 1: Y11,Y12,Y13,Y14,Y15Y_{11}, Y_{12}, Y_{13}, Y_{14}, Y_{15}

Breed 2: Y21,Y22,Y23,Y24,Y25Y_{21}, Y_{22}, Y_{23}, Y_{24}, Y_{25}

Breed 3: Y31,Y32,Y33,Y34,Y35Y_{31}, Y_{32}, Y_{33}, Y_{34}, Y_{35}

Breed 4: Y41,Y42,Y43,Y44,Y45Y_{41}, Y_{42}, Y_{43}, Y_{44}, Y_{45}

Summation across a specific row (summing over index jj for a fixed breed ii) is denoted using dot notation, where a dot (⋅\cdot) replaces the index over which summation has taken place. For instance, the total milk yield for Breed 3 across all 55 cows is expressed as:

Y3⋅=∑j=15Y3j=Y31+Y32+Y33+Y34+Y35Y_{3\cdot} = \sum_{j=1}^5 Y_{3j} = Y_{31} + Y_{32} + Y_{33} + Y_{34} + Y_{35}

If the row totals for the four breeds are given as Y1⋅=14Y_{1\cdot} = 14, Y2⋅=20Y_{2\cdot} = 20, Y3⋅=20Y_{3\cdot} = 20, and Y4⋅=24Y_{4\cdot} = 24, the total aggregate milk yield across all breeds and all cows (Y⋅⋅Y_{\cdot\cdot}) requires a double summation over both indices ii and jj:

∑i=14∑j=15Yij=Y⋅⋅=Y1⋅+Y2⋅+Y3⋅+Y4⋅=14+20+20+24=78\sum_{i=1}^4 \sum_{j=1}^5 Y_{ij} = Y_{\cdot\cdot} = Y_{1\cdot} + Y_{2\cdot} + Y_{3\cdot} + Y_{4\cdot} = 14 + 20 + 20 + 24 = 78

Mathematical Properties and Algebraic Identities of Sigma

The summation operator Σ\Sigma satisfies core algebraic properties that facilitate mathematical simplification of statistical equations:

  1. Summation of a constant scalar CC across nn terms:

∑i=1nC=C+C+⋯+C=nC\sum_{i=1}^n C = C + C + \dots + C = n C

For example, if C=2C = 2 and n=5n = 5:

∑i=152=5×2=10\sum_{i=1}^5 2 = 5 \times 2 = 10

  1. Factorization of a constant multiplier CC:

∑i=1nCXi=C∑i=1nXi\sum_{i=1}^n C X_i = C \sum_{i=1}^n X_i

  1. Summation of a linear offset (Xi+C)(X_i + C):

∑i=1n(Xi+C)=∑i=1nXi+nC\sum_{i=1}^n (X_i + C) = \sum_{i=1}^n X_i + n C

For example, if C=2C = 2 and n=5n = 5:

∑i=15(Xi+2)=∑i=15Xi+5×2=∑i=15Xi+10\sum_{i=1}^5 (X_i + 2) = \sum_{i=1}^5 X_i + 5 \times 2 = \sum_{i=1}^5 X_i + 10

  1. Summation of squared binomial sums:

∑i=1n(Xi+Yi)2=∑i=1n(Xi2+Yi2+2XiYi)=∑i=1nXi2+∑i=1nYi2+2∑i=1nXiYi\sum_{i=1}^n (X_i + Y_i)^2 = \sum_{i=1}^n (X_i^2 + Y_i^2 + 2 X_i Y_i) = \sum_{i=1}^n X_i^2 + \sum_{i=1}^n Y_i^2 + 2 \sum_{i=1}^n X_i Y_i

  1. Summation of squared deviations from a fixed constant aa:

∑i=1n(Xi−a)2=∑i=1nXi2−2a∑i=1nXi+na2\sum_{i=1}^n (X_i - a)^2 = \sum_{i=1}^n X_i^2 - 2 a \sum_{i=1}^n X_i + n a^2

∑i=1n(Xi+a)2=∑i=1nXi2+2a∑i=1nXi+na2\sum_{i=1}^n (X_i + a)^2 = \sum_{i=1}^n X_i^2 + 2 a \sum_{i=1}^n X_i + n a^2

  1. Square of a sum versus sum of squares with a added offset:

(∑i=1nXi+a)2=(∑i=1nXi)2+2a∑i=1nXi+a2(\sum_{i=1}^n X_i + a)^2 = (\sum_{i=1}^n X_i)^2 + 2 a \sum_{i=1}^n X_i + a^2

  1. Expansion of quadratic sums over indices:

∑i=13(Xi2+Xi)=∑i=13Xi2+∑i=13Xi\sum_{i=1}^3 (X_i^2 + X_i) = \sum_{i=1}^3 X_i^2 + \sum_{i=1}^3 X_i

∑i=13(Xi+3)2=∑i=13Xi2+6∑i=13Xi+27\sum_{i=1}^3 (X_i + 3)^2 = \sum_{i=1}^3 X_i^2 + 6 \sum_{i=1}^3 X_i + 27

∑i=15SiXi2=S1X12+S2X22+S3X32+S4X42+S5X52\sum_{i=1}^5 S_i X_i^2 = S_1 X_1^2 + S_2 X_2^2 + S_3 X_3^2 + S_4 X_4^2 + S_5 X_5^2

∑i=14Ui(Ui+2)=∑i=14Ui2+2∑i=14Ui\sum_{i=1}^4 U_i (U_i + 2) = \sum_{i=1}^4 U_i^2 + 2 \sum_{i=1}^4 U_i

∑i=14(aXi2+r)=a∑i=14Xi2+4r\sum_{i=1}^4 (a X_i^2 + r) = a \sum_{i=1}^4 X_i^2 + 4 r

Worked Examples and Computational Exercises on Summation

Given a sample size of n=5n = 5 with summary statistics ∑i=15Xi=20\sum_{i=1}^5 X_i = 20, ∑i=15Yi=17\sum_{i=1}^5 Y_i = 17, ∑i=15Xi2=90\sum_{i=1}^5 X_i^2 = 90, ∑i=15Yi2=75\sum_{i=1}^5 Y_i^2 = 75, and ∑i=15XiYi=66\sum_{i=1}^5 X_i Y_i = 66, evaluate the following statistical expressions:

  1. Evaluation of ∑i=15(2Xi+3)\sum_{i=1}^5 (2 X_i + 3):

∑i=15(2Xi+3)=2∑i=15Xi+3n=2(20)+3(5)=40+15=55\sum_{i=1}^5 (2 X_i + 3) = 2 \sum_{i=1}^5 X_i + 3 n = 2(20) + 3(5) = 40 + 15 = 55

  1. Evaluation of ∑i=15Xi(Xi−1)\sum_{i=1}^5 X_i (X_i - 1):

∑i=15Xi(Xi−1)=∑i=15Xi2−∑i=15Xi=90−20=70\sum_{i=1}^5 X_i (X_i - 1) = \sum_{i=1}^5 X_i^2 - \sum_{i=1}^5 X_i = 90 - 20 = 70

  1. Evaluation of ∑i=15(2Xi+1)(Xi−3)\sum_{i=1}^5 (2 X_i + 1)(X_i - 3):

∑i=15(2Xi+1)(Xi−3)=∑i=15(2Xi2−5Xi−3)=2∑i=15Xi2−5∑i=15Xi−3n=2(90)−5(20)−3(5)=180−100−15=65\sum_{i=1}^5 (2 X_i + 1)(X_i - 3) = \sum_{i=1}^5 (2 X_i^2 - 5 X_i - 3) = 2 \sum_{i=1}^5 X_i^2 - 5 \sum_{i=1}^5 X_i - 3 n = 2(90) - 5(20) - 3(5) = 180 - 100 - 15 = 65

  1. Evaluation of ∑i=15(Xi−5)2\sum_{i=1}^5 (X_i - 5)^2:

∑i=15(Xi−5)2=∑i=15Xi2−10∑i=15Xi+25n=90−10(20)+25(5)=90−200+125=15\sum_{i=1}^5 (X_i - 5)^2 = \sum_{i=1}^5 X_i^2 - 10 \sum_{i=1}^5 X_i + 25 n = 90 - 10(20) + 25(5) = 90 - 200 + 125 = 15

  1. Evaluation of ∑i=15(Xi+5Yi)\sum_{i=1}^5 (X_i + 5 Y_i):

∑i=15(Xi+5Yi)=∑i=15Xi+5∑i=15Yi=20+5(17)=20+85=105\sum_{i=1}^5 (X_i + 5 Y_i) = \sum_{i=1}^5 X_i + 5 \sum_{i=1}^5 Y_i = 20 + 5(17) = 20 + 85 = 105

  1. Evaluation of ∑i=15(Xi−5)Yi\sum_{i=1}^5 (X_i - 5) Y_i:

∑i=15(Xi−5)Yi=∑i=15XiYi−5∑i=15Yi=66−5(17)=66−85=−19\sum_{i=1}^5 (X_i - 5) Y_i = \sum_{i=1}^5 X_i Y_i - 5 \sum_{i=1}^5 Y_i = 66 - 5(17) = 66 - 85 = -19

Consider next a 6×36 \times 3 matrix of double-indexed observations XijX_{ij}, where i=1,2,…,6i = 1, 2, \dots, 6 represents rows and j=1,2,3j = 1, 2, 3 represents columns:

Row 1 (i=1i=1): X11=4X_{11} = 4, X12=3X_{12} = 3, X13=2X_{13} = 2

Row 2 (i=2i=2): X21=8X_{21} = 8, X22=6X_{22} = 6, X23=7X_{23} = 7

Row 3 (i=3i=3): X31=5X_{31} = 5, X32=5X_{32} = 5, X33=7X_{33} = 7

Row 4 (i=4i=4): X41=6X_{41} = 6, X42=2X_{42} = 2, X43=3X_{43} = 3

Row 5 (i=5i=5): X51=1X_{51} = 1, X52=6X_{52} = 6, X53=4X_{53} = 4

Row 6 (i=6i=6): X61=5X_{61} = 5, X62=7X_{62} = 7, X63=8X_{63} = 8

Evaluating row totals (Xi⋅X_{i\cdot}) yields:

X1⋅=∑j=13X1j=4+3+2=9X_{1\cdot} = \sum_{j=1}^3 X_{1j} = 4 + 3 + 2 = 9

X2⋅=∑j=13X2j=8+6+7=21X_{2\cdot} = \sum_{j=1}^3 X_{2j} = 8 + 6 + 7 = 21

X3⋅=∑j=13X3j=5+5+7=17X_{3\cdot} = \sum_{j=1}^3 X_{3j} = 5 + 5 + 7 = 17

X4⋅=∑j=13X4j=6+2+3=11X_{4\cdot} = \sum_{j=1}^3 X_{4j} = 6 + 2 + 3 = 11

X5⋅=∑j=13X5j=1+6+4=11X_{5\cdot} = \sum_{j=1}^3 X_{5j} = 1 + 6 + 4 = 11

X6⋅=∑j=13X6j=5+7+8=20X_{6\cdot} = \sum_{j=1}^3 X_{6j} = 5 + 7 + 8 = 20

Evaluating column totals (X⋅jX_{\cdot j}) yields:

X⋅1=∑i=16Xi1=4+8+5+6+1+5=29X_{\cdot 1} = \sum_{i=1}^6 X_{i1} = 4 + 8 + 5 + 6 + 1 + 5 = 29

X⋅2=∑i=16Xi2=3+6+5+2+6+7=29X_{\cdot 2} = \sum_{i=1}^6 X_{i2} = 3 + 6 + 5 + 2 + 6 + 7 = 29

X⋅3=∑i=16Xi3=2+7+7+3+4+8=31X_{\cdot 3} = \sum_{i=1}^6 X_{i3} = 2 + 7 + 7 + 3 + 4 + 8 = 31

Evaluating overall grand total and related composite operations:

∑i=16∑j=13Xij=X1⋅+X2⋅+X3⋅+X4⋅+X5⋅+X6⋅=9+21+17+11+11+20=89\sum_{i=1}^6 \sum_{j=1}^3 X_{ij} = X_{1\cdot} + X_{2\cdot} + X_{3\cdot} + X_{4\cdot} + X_{5\cdot} + X_{6\cdot} = 9 + 21 + 17 + 11 + 11 + 20 = 89

(∑i=16∑j=13Xij)2=(89)2=7921(\sum_{i=1}^6 \sum_{j=1}^3 X_{ij})^2 = (89)^2 = 7921

∑j=13(X1j−X2j)=X1⋅−X2⋅=9−21=−12\sum_{j=1}^3 (X_{1j} - X_{2j}) = X_{1\cdot} - X_{2\cdot} = 9 - 21 = -12

∑j=13(X3j+X4j)=X3⋅+X4⋅=17+11=28\sum_{j=1}^3 (X_{3j} + X_{4j}) = X_{3\cdot} + X_{4\cdot} = 17 + 11 = 28

(X1⋅−X2⋅)=9−21=−12(X_{1\cdot} - X_{2\cdot}) = 9 - 21 = -12

Frequency Distributions for Categorical Discrete Data

When handling qualitative or categorical discrete data, observation categories are assigned arbitrary integer labels XiX_i to organize tally counts. These numeric labels serve purely as identifiers and do not allow standard arithmetic operations.

Consider an industrial manufacturing craftsman producing 44 distinct component types labeled AA, BB, CC, and DD. Over the course of a day, 2020 components (N=20N = 20) are produced in the following sequence:

B,C,C,A,D,C,C,B,D,C,A,C,D,C,B,C,C,B,D,DB, C, C, A, D, C, C, B, D, C, A, C, D, C, B, C, C, B, D, D

Assigning numerical codes A→1A \rightarrow 1, B→2B \rightarrow 2, C→3C \rightarrow 3, and D→4D \rightarrow 4, the complete frequency distribution table is constructed by counting absolute frequencies (fif_i), relative frequencies (ri=fiNr_i = \frac{f_i}{N}), cumulative frequencies (gi=∑k=1ifkg_i = \sum_{k=1}^i f_k), and relative cumulative frequencies (si=giNs_i = \frac{g_i}{N}):

Component Type AA (X1=1X_1 = 1): f1=2f_1 = 2, r1=0.10r_1 = 0.10, g1=2g_1 = 2, s1=0.10s_1 = 0.10

Component Type BB (X2=2X_2 = 2): f2=4f_2 = 4, r2=0.20r_2 = 0.20, g2=6g_2 = 6, s2=0.30s_2 = 0.30

Component Type CC (X3=3X_3 = 3): f3=9f_3 = 9, r3=0.45r_3 = 0.45, g3=15g_3 = 15, s3=0.75s_3 = 0.75

Component Type DD (X4=4X_4 = 4): f4=5f_4 = 5, r4=0.25r_4 = 0.25, g4=20g_4 = 20, s4=1.00s_4 = 1.00

Total Count: ∑fi=20\sum f_i = 20, ∑ri=1.00\sum r_i = 1.00

Interpretative insights from the table include:

  1. The value r2=0.20r_2 = 0.20 in the relative frequency column indicates that exactly 20%20\% of all components produced during the day are of Type BB.

  2. The value s2=0.30s_2 = 0.30 in the relative cumulative frequency column indicates that 30%30\% of all manufactured components belong to either Type AA or Type BB.

Categorical frequency distributions are visually represented using Bar Charts (where bar height reflects category frequency or relative frequency) and Pie Charts (where angular slice area represents proportional percentage contribution, e.g., Type CC occupies 45%45\% of the circle, Type DD occupies 25%25\%).

Frequency Distributions for Quantitative Discrete Data

Quantitative discrete data represent numerical values obtained via direct counting where arithmetic operations hold explicit quantitative meaning.

Consider a public health survey recording the monthly consumption of cold tablets across 5050 families (N=50N = 50) during a winter month. The raw collected dataset consists of:

7,5,3,3,4,5,3,2,8,3,3,2,4,4,3,6,8,6,7,4,5,4,6,4,5,7, 5, 3, 3, 4, 5, 3, 2, 8, 3, 3, 2, 4, 4, 3, 6, 8, 6, 7, 4, 5, 4, 6, 4, 5,2,3,4,2,7,3,5,4,6,2,2,3,4,5,4,8,4,3,2,2,6,4,5,7,82, 3, 4, 2, 7, 3, 5, 4, 6, 2, 2, 3, 4, 5, 4, 8, 4, 3, 2, 2, 6, 4, 5, 7, 8

The unique outcome values range from 22 to 88 tablets. The constructed frequency distribution table summarizes absolute frequency (fif_i), relative frequency (rir_i), cumulative frequency (gig_i), relative cumulative frequency (sis_i), and weighted value product (fiXif_i X_i):

Tablet Count X1=2X_1 = 2: f1=8f_1 = 8, r1=0.16r_1 = 0.16, g1=8g_1 = 8, s1=0.16s_1 = 0.16, f1X1=16f_1 X_1 = 16

Tablet Count X2=3X_2 = 3: f2=10f_2 = 10, r2=0.20r_2 = 0.20, g2=18g_2 = 18, s2=0.36s_2 = 0.36, f2X2=30f_2 X_2 = 30

Tablet Count X3=4X_3 = 4: f3=12f_3 = 12, r3=0.24r_3 = 0.24, g3=30g_3 = 30, s3=0.60s_3 = 0.60, f3X3=48f_3 X_3 = 48

Tablet Count X4=5X_4 = 5: f4=7f_4 = 7, r4=0.14r_4 = 0.14, g4=37g_4 = 37, s4=0.74s_4 = 0.74, f4X4=35f_4 X_4 = 35

Tablet Count X5=6X_5 = 6: f5=6f_5 = 6, r5=0.12r_5 = 0.12, g5=43g_5 = 43, s5=0.86s_5 = 0.86, f5X5=36f_5 X_5 = 36

Tablet Count X6=7X_6 = 7: f6=4f_6 = 4, r6=0.08r_6 = 0.08, g6=47g_6 = 47, s6=0.94s_6 = 0.94, f6X6=28f_6 X_6 = 28

Tablet Count X7=8X_7 = 8: f7=4f_7 = 4, r7=0.08r_7 = 0.08, g7=50g_7 = 50, s7=1.00s_7 = 1.00, f7X7=32f_7 X_7 = 32

Totals: ∑fi=50\sum f_i = 50, ∑ri=1.00\sum r_i = 1.00, ∑fiXi=225\sum f_i X_i = 225

To determine the percentage of families consuming strictly more than 44 tablets per month, evaluate the relative cumulative frequency at X3=4X_3 = 4. The value s3=0.60s_3 = 0.60 confirms that 60%60\% of families consume at most 44 tablets. Subtracting from unity gives 1.00−0.60=0.401.00 - 0.60 = 0.40, establishing that exactly 40%40\% of families consume more than 44 cold tablets per month.

Frequency Distributions for Continuous Data and Sturges' Rule

Continuous data arise from precise measurement processes where observed values can take any real number within a continuous interval. Grouping continuous data requires defining contiguous, non-overlapping class intervals of equal width.

Consider weight measurements of 4040 butter blocks (N=40N = 40) rounded to the nearest integer gram:

52,35,24,47,36,34,38,46,47,36,38,50,47,34,41,40,42,40,26,29,52, 35, 24, 47, 36, 34, 38, 46, 47, 36, 38, 50, 47, 34, 41, 40, 42, 40, 26, 29,30,32,35,37,37,41,21,31,26,35,45,23,43,34,43,…30, 32, 35, 37, 37, 41, 21, 31, 26, 35, 45, 23, 43, 34, 43, \dots

The standardized four-step algorithm for constructing a continuous frequency distribution is defined as follows:

  1. Determination of Class Count (kk): While class selection generally ranges between 55 and 2525, Sturges' Rule provides an exact formula based on sample size nn:

k=1+3.322log⁡10(n)k = 1 + 3.322 \log_{10}(n)

For N=40N = 40:

k=1+3.322log⁡10(40)=1+3.322(1.60206)=1+5.3227=6.3227k = 1 + 3.322 \log_{10}(40) = 1 + 3.322(1.60206) = 1 + 5.3227 = 6.3227

Rounding to the next higher integer yields k=7k = 7 classes.

  1. Precision Unit and Bounds Offset (SS): Because raw measurements are rounded to the nearest integer (rounding unit =1= 1), an observation of 3535 represents continuous values in the true range (34.5,35.5)(34.5, 35.5). The measurement variability precision offset SS is:

S=Rounding Unit2=12=0.5S = \frac{\text{Rounding Unit}}{2} = \frac{1}{2} = 0.5

  1. Calculation of True Limits and Adjusted Range (RR):

minadj=minraw−S=21−0.5=20.5\text{min}_{adj} = \text{min}_{raw} - S = 21 - 0.5 = 20.5

maxadj=maxraw+S=52+0.5=52.5\text{max}_{adj} = \text{max}_{raw} + S = 52 + 0.5 = 52.5

R=maxadj−minadj=52.5−20.5=32R = \text{max}_{adj} - \text{min}_{adj} = 52.5 - 20.5 = 32

  1. Calculation of Class Width (ww): Class width is obtained by dividing true range RR by class count kk and rounding up to match precision:

w=Rk=327=4.5714≈5w = \frac{R}{k} = \frac{32}{7} = 4.5714 \approx 5

Comprehensive Descriptive Statistics and Quantile Calculations

When calculating positional quantiles (quartiles QpQ_p, deciles DpD_p, percentiles PpP_p) for grouped continuous data, the index position npn p is computed first. The quantile value is then interpolated using the linear interpolation formula:

Pp=L+np−Fcfi×wP_p = L + \frac{n p - F_c}{f_i} \times w

where LL is the lower true boundary of the quantile class, npn p is the target positional rank, FcF_c is the cumulative frequency of all classes preceding the quantile class, fif_i is the absolute frequency of the quantile class, and ww is the class width.

Applying these parameter estimations to the butter weight dataset (N=40N = 40, Xˉ=36.875\bar{X} = 36.875, s=8.056s = 8.056, s2=64.9s^2 = 64.9) yields the following structural parameters:

  1. Positional Index Values (npn p):

Q1 rank:40×0.25=10Q_1 \text{ rank}: 40 \times 0.25 = 10

Q2 (Median) rank:40×0.50=20Q_2 \text{ (Median) rank}: 40 \times 0.50 = 20

Q3 rank:40×0.75=30Q_3 \text{ rank}: 40 \times 0.75 = 30

D6 rank:40×0.60=24D_6 \text{ rank}: 40 \times 0.60 = 24

P10 rank:40×0.10=4P_{10} \text{ rank}: 40 \times 0.10 = 4

P90 rank:40×0.90=36P_{90} \text{ rank}: 40 \times 0.90 = 36

  1. Calculated Positional Quantiles:

Q1=30.5+10−910×5=31.0Q_1 = 30.5 + \frac{10 - 9}{10} \times 5 = 31.0

Q2=35.5+20−1910×5=36.0Q_2 = 35.5 + \frac{20 - 19}{10} \times 5 = 36.0

Q3=40.5+30−298×5=41.125Q_3 = 40.5 + \frac{30 - 29}{8} \times 5 = 41.125

D6=35.5+24−1910×5=38.0D_6 = 35.5 + \frac{24 - 19}{10} \times 5 = 38.0

P10=25.5+4−24×5=28.0P_{10} = 25.5 + \frac{4 - 2}{4} \times 5 = 28.0

P90=45.5+36−37…=48.5P_{90} = 45.5 + \frac{36 - 37}{\dots} = 48.5

  1. Grouped Mode (Mod\text{Mod}):

Mod=L+d1d1+d2×w=35.5+(10−9)(10−9)+(10−8)×5=35.5+13×5=37.167\text{Mod} = L + \frac{d_1}{d_1 + d_2} \times w = 35.5 + \frac{(10 - 9)}{(10 - 9) + (10 - 8)} \times 5 = 35.5 + \frac{1}{3} \times 5 = 37.167

  1. Relative Variability and Shape Coefficients:

Coefficient of Variation (CV)=sXˉ×100=8.0636.87×100=21.85%\text{Coefficient of Variation (CV)} = \frac{s}{\bar{X}} \times 100 = \frac{8.06}{36.87} \times 100 = 21.85\%

Pearson Skewness (Sk1)=Xˉ−Mods=36.87−37.838.06=−0.119\text{Pearson Skewness } (Sk_1) = \frac{\bar{X} - \text{Mod}}{s} = \frac{36.87 - 37.83}{8.06} = -0.119

Quartile Skewness (Skp)=Q3−2Q2+Q1Q3−Q1=42.375−2(36.825)+31.042.375−31.0\text{Quartile Skewness } (Sk_p) = \frac{Q_3 - 2 Q_2 + Q_1}{Q_3 - Q_1} = \frac{42.375 - 2(36.825) + 31.0}{42.375 - 31.0}

Kurtosis Coefficient (Ce)=Q3−Q12(P90−P10)\text{Kurtosis Coefficient } (Ce) = \frac{Q_3 - Q_1}{2(P_{90} - P_{10})}

For an ungrouped discrete numerical sequence consisting of N=13N = 13 ordered observations 20,23,25,28,30,31,33,35,36,38,40,43,4520, 23, 25, 28, 30, 31, 33, 35, 36, 38, 40, 43, 45, quantile locations are computed via positional interpolation:

Q1 rank:N+14=144=3.5→Q1=25+282=26.5Q_1 \text{ rank}: \frac{N+1}{4} = \frac{14}{4} = 3.5 \rightarrow Q_1 = \frac{25 + 28}{2} = 26.5

Q2 rank:N+12=142=7→Q2=33Q_2 \text{ rank}: \frac{N+1}{2} = \frac{14}{2} = 7 \rightarrow Q_2 = 33

Q3 rank:3(N+1)4=10.5→Q3=38+402=39Q_3 \text{ rank}: \frac{3(N+1)}{4} = 10.5 \rightarrow Q_3 = \frac{38 + 40}{2} = 39

D6 rank:6(N+1)10=8.4→D6=35+0.4(36−35)=35.4D_6 \text{ rank}: \frac{6(N+1)}{10} = 8.4 \rightarrow D_6 = 35 + 0.4(36 - 35) = 35.4

P40 rank:40(N+1)100=5.6→P40=30+0.6(31−30)=30.6P_{40} \text{ rank}: \frac{40(N+1)}{100} = 5.6 \rightarrow P_{40} = 30 + 0.6(31 - 30) = 30.6

Linear Transformations of Random Variables

When a primary variable XX undergoes a linear transformation of the form Y=aX+bY = a X + b, its expected mean, variance, and standard deviation transform according to explicit mathematical principles:

Yˉ=aXˉ+b\bar{Y} = a \bar{X} + b

sY2=a2sX2s_Y^2 = a^2 s_X^2

sY=∣a∣sXs_Y = |a| s_X

Given baseline statistics Xˉ=36.875\bar{X} = 36.875 and variance sX2=64.9s_X^2 = 64.9 (sX=8.056s_X = 8.056), applying the linear transformation Y=5X+3Y = 5 X + 3 yields:

Yˉ=5Xˉ+3=5(36.875)+3=184.375+3=187.375\bar{Y} = 5 \bar{X} + 3 = 5(36.875) + 3 = 184.375 + 3 = 187.375

sY2=52sX2=25×64.9=1622.5s_Y^2 = 5^2 s_X^2 = 25 \times 64.9 = 1622.5

sY=5sX=5×8.056=40.28s_Y = 5 s_X = 5 \times 8.056 = 40.28

Applying a secondary nested transformation Z=3Y+8Z = 3 Y + 8 yields:

Zˉ=3Yˉ+8=3(187.375)+8=562.125+8=570.125\bar{Z} = 3 \bar{Y} + 8 = 3(187.375) + 8 = 562.125 + 8 = 570.125

sZ2=32sY2=9×1622.5=14602.5s_Z^2 = 3^2 s_Y^2 = 9 \times 1622.5 = 14602.5

sZ=3sY=3×40.28=120.84s_Z = 3 s_Y = 3 \times 40.28 = 120.84

Cotton Thread Tensile Strength Case Study

An experiment evaluates the breaking force capacity (tensile strength) of 5050 cotton thread strands (N=50N = 50). The raw force observations are recorded as follows:

21.2,21.3,27.1,25.0,33.7,27.3,29.4,21.9,20.6,29.5,21.2, 21.3, 27.1, 25.0, 33.7, 27.3, 29.4, 21.9, 20.6, 29.5,21.8,37.5,32.7,29.5,30.2,23.9,23.0,26.4,29.1,31.3,21.8, 37.5, 32.7, 29.5, 30.2, 23.9, 23.0, 26.4, 29.1, 31.3,29.0,36.8,29.2,23.5,33.5,29.6,26.8,28.7,34.8,18.6,29.0, 36.8, 29.2, 23.5, 33.5, 29.6, 26.8, 28.7, 34.8, 18.6,29.6,25.4,34.1,27.5,29.6,22.2,22.7,31.3,33.2,37.0,29.6, 25.4, 34.1, 27.5, 29.6, 22.2, 22.7, 31.3, 33.2, 37.0,21.3,36.9,24.6,28.2,24.8,28.1,22.5,34.5,23.6,38.4,24.021.3, 36.9, 24.6, 28.2, 24.8, 28.1, 22.5, 34.5, 23.6, 38.4, 24.0

To construct the continuous frequency distribution table:

  1. Class Count via Sturges' Rule:

k=1+3.322log⁡10(50)=1+3.322(1.69897)=1+5.644=6.644≈7k = 1 + 3.322 \log_{10}(50) = 1 + 3.322(1.69897) = 1 + 5.644 = 6.644 \approx 7

  1. Precision Unit Offset SS: Data are measured to one decimal place (0.10.1 unit), giving S=0.12=0.05S = \frac{0.1}{2} = 0.05.

  2. True Range RR:

minadj=17.3−0.05=17.25\text{min}_{adj} = 17.3 - 0.05 = 17.25

maxadj=38.4+0.05=38.45\text{max}_{adj} = 38.4 + 0.05 = 38.45

R=38.45−17.25=21.2R = 38.45 - 17.25 = 21.2

  1. Class Width ww:

w=Rk=21.27=3.0286≈3.1w = \frac{R}{k} = \frac{21.2}{7} = 3.0286 \approx 3.1

The full continuous frequency distribution (Table 4.1) comprises 77 class intervals of width 3.13.1:

Class 1 (17.25−20.3517.25 - 20.35): Midpoint X1=18.8X_1 = 18.8, f1=2f_1 = 2, r1=0.04r_1 = 0.04, g1=2g_1 = 2, s1=0.04s_1 = 0.04

Class 2 (20.35−23.4520.35 - 23.45): Midpoint X2=21.9X_2 = 21.9, f2=7f_2 = 7, r2=0.14r_2 = 0.14, g2=9g_2 = 9, s2=0.18s_2 = 0.18

Class 3 (23.45−26.5523.45 - 26.55): Midpoint X3=25.0X_3 = 25.0, f3=10f_3 = 10, r3=0.20r_3 = 0.20, g3=19g_3 = 19, s3=0.38s_3 = 0.38

Class 4 (26.55−29.6526.55 - 29.65): Midpoint X4=28.1X_4 = 28.1, f4=17f_4 = 17, r4=0.34r_4 = 0.34, g4=36g_4 = 36, s4=0.72s_4 = 0.72

Class 5 (29.65−32.7529.65 - 32.75): Midpoint X5=31.2X_5 = 31.2, f5=3f_5 = 3, r5=0.06r_5 = 0.06, g5=39g_5 = 39, s5=0.78s_5 = 0.78

Class 6 (32.75−35.8532.75 - 35.85): Midpoint X6=34.3X_6 = 34.3, f6=6f_6 = 6, r6=0.12r_6 = 0.12, g6=45g_6 = 45, s6=0.90s_6 = 0.90

Class 7 (35.85−38.9535.85 - 38.95): Midpoint X7=37.4X_7 = 37.4, f7=5f_7 = 5, r7=0.10r_7 = 0.10, g7=50g_7 = 50, s7=1.00s_7 = 1.00

Totals: ∑fi=50\sum f_i = 50, ∑ri=1.00\sum r_i = 1.00

Quality Control and Comparative Analysis of Light Bulbs

A manufacturing plant produces two types of light bulbs. Performance metrics are evaluated using the sample mean lifespan (measure of central tendency) and coefficient of variation (measure of relative dispersion):

Type 1 Light Bulb: Mean lifespan Xˉ1=200 hours\bar{X}_1 = 200\text{ hours}, Standard deviation S1=11 hoursS_1 = 11\text{ hours}.

Coefficient of Variation (CV1)=S1Xˉ1=11200=0.055→5.5%\text{Coefficient of Variation } (CV_1) = \frac{S_1}{\bar{X}_1} = \frac{11}{200} = 0.055 \rightarrow 5.5\%

Type 2 Light Bulb: Mean lifespan Xˉ2=240 hours\bar{X}_2 = 240\text{ hours}, Standard deviation S2=12 hoursS_2 = 12\text{ hours}.

Coefficient of Variation (CV2)=S2Xˉ2=12240=0.050→5.0%\text{Coefficient of Variation } (CV_2) = \frac{S_2}{\bar{X}_2} = \frac{12}{240} = 0.050 \rightarrow 5.0\%

Conclusion: Type 2 light bulbs are superior in quality compared to Type 1 light bulbs. Type 2 exhibits both a higher average operational lifespan (240 hours>200 hours240\text{ hours} > 200\text{ hours}) and lower relative variability (CV2=5.0%<CV1=5.5%CV_2 = 5.0\% < CV_1 = 5.5\%), demonstrating higher product consistency.

Reconstruction of Incomplete Frequency Distributions

Consider an incomplete frequency distribution representing weight measurements of N=50N = 50 students with uniform class width w=5w = 5:

  1. Class 1 (30.5−35.530.5 - 35.5): Given relative frequency r1=0.06r_1 = 0.06.

f1=r1×N=0.06×50=3f_1 = r_1 \times N = 0.06 \times 50 = 3

Midpoint X1=30.5+35.52=33\text{Midpoint } X_1 = \frac{30.5 + 35.5}{2} = 33

g1=3,s1=0.06g_1 = 3, s_1 = 0.06

  1. Class 2 (35.5−40.535.5 - 40.5): Given cumulative frequency g2=19g_2 = 19

f2=g2−g1=19−3=16f_2 = g_2 - g_1 = 19 - 3 = 16

r2=1650=0.32r_2 = \frac{16}{50} = 0.32

Midpoint X2=38,s2=1950=0.38\text{Midpoint } X_2 = 38, s_2 = \frac{19}{50} = 0.38

  1. Class 3 (40.5−45.540.5 - 45.5): Given relative cumulative frequency s3=0.64s_3 = 0.64

g3=s3×N=0.64×50=32g_3 = s_3 \times N = 0.64 \times 50 = 32

f3=g3−g2=32−19=13f_3 = g_3 - g_2 = 32 - 19 = 13

r3=1350=0.26, Midpoint X3=43r_3 = \frac{13}{50} = 0.26, \text{ Midpoint } X_3 = 43

  1. Class 4 (45.5−50.545.5 - 50.5): Given cumulative frequency g4=46g_4 = 46

f4=g4−g3=46−32=14f_4 = g_4 - g_3 = 46 - 32 = 14

r4=1450=0.28,s4=4650=0.92, Midpoint X4=48r_4 = \frac{14}{50} = 0.28, s_4 = \frac{46}{50} = 0.92, \text{ Midpoint } X_4 = 48

  1. Class 5 (50.5−55.550.5 - 55.5): Final cumulative total g5=50g_5 = 50

f5=50−46=4f_5 = 50 - 46 = 4

r5=450=0.08,s5=1.00, Midpoint X5=53r_5 = \frac{4}{50} = 0.08, s_5 = 1.00, \text{ Midpoint } X_5 = 53