Quantitative Techniques & Statistics

Formulas

Measures of Central Tendency

• Mean (x̄) P
fx
x̄ = (Direct Method)
N
• Median (M ) !
N
2 − cf
M =L+ ×h
f

Where L=Lower limit, cf =Cumulative freq of previous class, f =frequency, h=class width.

• Mode (Z)  
f1 − f0
Z =L+ ×h
2f1 − f0 − f2

• Quartiles (Qi ) !
iN
4 − cf
Qi = L + ×h for i = 1, 2, 3
f

• Deciles (Dk ) & Percentiles (Pk )

 th  th
kN kN
Dk = Value of term, Pk = Value of term
10 100

1
Measures of Dispersion
• Range (R)
R=L−S (Largest - Smallest value)

• Interquartile Range (IQR) & Quartile Deviation

IQR = Q3 − Q1
Q3 − Q1
Q.D. =
2
Q3 − Q1
Coeff. of Q.D. =
Q3 + Q1

• Mean Deviation (M.D.) P

f |x − x̄|
M.D. =
N
M.D.
Coeff. of M.D. =
x̄
• Standard Deviation (σ or S.D.)
rP
f (x − x̄)2
σ=
N

• Coefficient of Variation (C.V.)

σ
C.V. = × 100
x̄

Skewness & Kurtosis

• Karl Pearson’s Coefficient of Skewness (Sk )

Mean − Mode
Sk =
σ
If Mode is ill-defined:
3(Mean − Median)
Sk =
σ
• Kurtosis (β2 )
µ4
β2 =
µ22
Where µ2 is variance and µ4 is the fourth moment.

– β2 = 3: Mesokurtic (Normal)
– β2 > 3: Leptokurtic (Peaked)
– β2 < 3: Platykurtic (Flat)

2
Probability Theory
• Permutations & Combinations

n n!
Pr = (Arrangement)
(n − r)!

n n!
Cr = (Selection)
r!(n − r)!

• Basic Probability
Favorable Outcomes
P (A) =
Total Outcomes
• Addition Rule
P (A ∪ B) = P (A) + P (B) − P (A ∩ B)

• Conditional Probability

P (A ∩ B) P (A ∩ B)
P (A|B) = and P (B|A) =
P (B) P (A)

• Multiplication Rule
P (A ∩ B) = P (A) · P (B|A)
If independent: P (A ∩ B) = P (A) · P (B)

• Bayes’ Theorem
P (Ei )P (A|Ei )
P (Ei |A) = P
P (Ej )P (A|Ej )

Probability Distributions
• Binomial Distribution
P (x) =n Cx · px · q n−x
Where n=trials, x=successes, p=prob of success, q = 1 − p.

• Poisson Distribution
e−m mx
P (x) =
x!
Where m = np (mean).

• Standard Normal Distribution

X −µ
Z=
σ
Where µ=Mean, σ=Standard Deviation.

3
Correlation & Regression
• Karl Pearson’s Correlation (r)
P
(X − X̄)(Y − Ȳ )
r= P
p pP
(X − X̄)2 (Y − Ȳ )2

• Spearman’s Rank Correlation (R)

6 d2
P
R=1−
n(n2 − 1)

Where d = Rank1 − Rank2 .

• Regression Equations X on Y:
σx
X − X̄ = bxy (Y − Ȳ ) where bxy = r
σy

Y on X:
σy
Y − Ȳ = byx (X − X̄) where byx = r
σx

Linear Programming (Graphical Method)

Steps:

1. Formulate Objective Function: Maximize or Minimize Z = ax + by.

2. Write Constraints as inequalities (e.g., cx + dy ≤ k).

3. Convert inequalities to equations to plot lines on graph.

4. Identify Feasible Region (Intersection of all constraints).

5. Find Corner Points of the feasible region.

6. Substitute coordinates of corner points into Z to find optimal solution.