15 Questions — Exam Ready Statistics & Probability — Quick Notes

Statistics & Probability

Complete Exam Notes — 15 Questions Syllabus
Covers all 3 units: Statistics · Correlation & Regression · Probability

UNIT 1: STATISTICS
Definitions, Scope & Limitations
ˆ Statistics: Science of collecting, organising, analysing, interpreting and presenting data.
ˆ Scope: Used in economics, business, medicine, engineering, social sciences.
ˆ Limitations: Studies aggregates (not individuals); results may be misused; does not
reveal the full picture; qualitative data cannot be measured directly.

Sampling Methods
ˆ Random Sampling: Every element has equal probability of selection. (Lottery, random
number table)
ˆ Stratified Sampling: Population divided into strata; samples drawn from each stratum
proportionally.
ˆ Systematic Sampling: Every k th element selected. k = N/n (population/sample size).
ˆ Cluster Sampling: Population divided into clusters; entire clusters selected randomly.
ˆ Convenience Sampling: Selection based on ease of access (non-probabilistic).

Collection, Classification & Tabulation of Data

ˆ Primary Data: Collected first-hand (surveys, experiments). More accurate but costly.
ˆ Secondary Data: Already collected and published (census reports, journals).
ˆ Classification: Organising data into classes — Geographical, Chronological, Qualitative,
Quantitative.
ˆ Tabulation: Presenting data in rows and columns for easy analysis.

Frequency Distribution
Range
ˆ Class Width =
No. of classes
ˆ Cumulative Frequency: Running total of frequencies up to a class boundary.
f
ˆ Relative Frequency = , where N =
P
f
N

Diagrammatic & Graphical Representation

ˆ Bar diagram (simple, multiple, sub- ˆ Frequency polygon

divided) ˆ Ogive (cumulative frequency curve):
ˆ Pie chart / Pie diagram Less-than and more-than ogive
ˆ Histogram ˆ Frequency curve (smoothed polygon)

Ogive Tip: Median = x-value at N/2 on the less-than ogive.

1
15 Questions — Exam Ready Statistics & Probability — Quick Notes

Measures of Central Tendency

Key Formulae
Mean (Arithmetic):
P P
fx x
x̄ = P (Grouped), x̄ = (Ungrouped)
f n

Step-Deviation (Coding) Method:

P
fd x−A
x̄ = A + P · h, d=
f h

Median: Middle value when data is sorted.

N
2
− cf
M =L+ ×h
f
where L = lower class boundary, cf = cumulative frequency before class, f = class frequency,
h = class width.
Mode: Value with highest frequency.
f1 − f0
Z =L+ ×h
2f1 − f0 − f2
where f1 = modal class freq, f0 = preceding, f2 = succeeding.

Empirical Relationship

Mode = 3 × Median − 2 × Mean

Partition Values

Key Formulae
rN
4
− cf
Quartiles: Qr = L + × h, r = 1, 2, 3
f
rN
− cf
Deciles: Dr = L + 10 × h, r = 1, . . . , 9
f
rN
100
− cf
Percentiles: Pr = L + × h, r = 1, . . . , 99
f
Note: Q2 = D5 = P50 = Median

2
15 Questions — Exam Ready Statistics & Probability — Quick Notes

Measures of Dispersion

Key Formulae

Measure Formula
Range R = Xmax − Xmin
P
f |x − x̄|
Mean Deviation (about mean) MD = P
f
x̄)2
P
f (x −
Variance σ2 = P
sP f
f x2
Standard Deviation σ= P − x̄2
f
Q3 − Q1
Quartile Deviation (Semi-IQR) QD =
2

Key Formulae
Coefficient of Variation (CV):
σ
CV = × 100%
x̄
Lower CV ⇒ more consistent/uniform data.

Skewness & Kurtosis

Key Formulae
Karl Pearson’s Coefficient of Skewness:
x̄ − Z 3(x̄ − M )
Sk = or Sk =
σ σ
Bowley’s Coefficient (Quartile Skewness):

Q3 + Q1 − 2Q2
Sk =
Q3 − Q1

ˆ Sk > 0: Positively skewed (tail on right)

ˆ Sk < 0: Negatively skewed (tail on left)
ˆ Sk = 0: Symmetric
Kurtosis measures peakedness of distribution:
µ4
β2 = , γ2 = β2 − 3
µ22

ˆ γ2 = 0: Mesokurtic (Normal)
ˆ γ2 > 0: Leptokurtic (more peaked)
ˆ γ2 < 0: Platykurtic (flatter)

3
15 Questions — Exam Ready Statistics & Probability — Quick Notes

UNIT 2: CORRELATION & REGRESSION

Scatter Diagram
A graphical tool to visualise the relationship between two variables X and Y .
Patterns: Points going up-right ⇒ positive correlation; down-right ⇒ negative; random ⇒
no correlation.

Coefficient of Correlation (Karl Pearson’s r)

Key Formulae
P P P P
(x − x̄)(y − ȳ) n xy − x y
r = pP P =p P P P P
(x − x̄)2 · (y − ȳ)2 [n x2 − ( x)2 ][n y 2 − ( y)2 ]
ˆ r ∈ [−1, +1]
ˆ r = +1: Perfect positive; r = −1: Perfect negative; r = 0: No linear correlation
ˆ r2 = Coefficient of determination (proportion of variation explained)

Spearman’s Rank Correlation

Key Formulae
6 d2i
P
ρ=1−
n(n2 − 1)
where di = Rxi − Ryi (difference in ranks). Used when data is ordinal or non-normal.
m3 − m P 2
With tied ranks: Add correction factor for each tie group of size m to d.
12

Lines of Linear Regression

Two Regression Lines

σy
Regression of Y on X: Y − ȳ = byx (X − x̄), byx = r
σx
σx
Regression of X on Y : X − x̄ = bxy (Y − ȳ), bxy =r
σy
p
Relationship: r = ± byx · bxy (signs of both b must match)
Both lines pass through (x̄, ȳ).

Key Formulae
Least Squares Normal Equations (for Y = a + bX):
X X X X X
Y = na + b X, XY = a X +b X2
P P P
n XY − X Y
b= P P , a = Ȳ − bX̄
n X 2 − ( X)2

4
15 Questions — Exam Ready Statistics & Probability — Quick Notes

Partial & Multiple Correlation

ˆ Partial Correlation (r12.3 ): Correlation between X1 and X2 keeping X3 constant.
r12 − r13 · r23
r12.3 = p 2 2
(1 − r13 )(1 − r23 )

ˆ Multiple Correlation (R1.23 ): Correlation of X1 with the combined effect of X2 and

X3 .
2 2
2 r12 + r13 − 2r12 · r13 · r23
R1.23 = 2
1 − r23

Multiple Linear Regression

Y = b0 + b1 X 1 + b2 X 2 + · · · + bk X k
Coefficients estimated via least squares. Measures the combined linear effect of multiple pre-
dictors on Y .

UNIT 3: PROBABILITY
Basic Concepts & Events
ˆ Sample Space S: Set of all possible outcomes.
ˆ Event: A subset of S.
ˆ Mutually Exclusive Events: A ∩ B = ∅ (cannot occur simultaneously).
ˆ Exhaustive Events: A ∪ B = S.
Favourable outcomes
ˆ Classical Probability: P (A) =
Total outcomes
ˆ 0 ≤ P (A) ≤ 1; P (S) = 1; P (∅) = 0

Addition Law of Probability

Key Formulae
General:
P (A ∪ B) = P (A) + P (B) − P (A ∩ B)
Mutually Exclusive:
P (A ∪ B) = P (A) + P (B)
For three events:

P (A ∪ B ∪ C) = P (A) + P (B) + P (C) − P (A ∩ B) − P (B ∩ C) − P (A ∩ C) + P (A ∩ B ∩ C)

5
15 Questions — Exam Ready Statistics & Probability — Quick Notes

Conditional Probability & Multiplication Law

Key Formulae
Conditional Probability:

P (A ∩ B)
P (A|B) = , P (B) > 0
P (B)

Multiplication Law:

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

Independent Events: P (A ∩ B) = P (A) · P (B) ⇔ P (A|B) = P (A)

Baye’s Theorem

Baye’s Theorem
If B1 , B2 , . . . , Bn are mutually exclusive and exhaustive events, and A is any event:

P (Bi ) · P (A|Bi )
P (Bi |A) = n
X
P (Bj ) · P (A|Bj )
j=1

Prior probabilities P (Bi ) are updated to posterior probabilities P (Bi |A) after observing A.

Random Variable
ˆ Discrete RV: Takes countable values (e.g., number of defects). Described by PMF.
ˆ Continuous RV: Takes uncountable values in an interval. Described by PDF.
ˆ CDF: F (x) = P (X ≤ x)

Probability Mass Function (PMF) — Discrete

Key Formulae
P
Conditions: p(x) ≥ 0 for all x; x p(x) = 1
P
Expectation: E(X) = µ = x · p(x)
Variance: Var(X) = E(X ) − [E(X)]2 =
2
P 2
x p(x) − µ2

6
15 Questions — Exam Ready Statistics & Probability — Quick Notes

Probability Density Function (PDF) — Continuous

Key Formulae
Z ∞
Conditions: f (x) ≥ 0; f (x) dx = 1
−∞
Z b
P (a ≤ X ≤ b) = f (x) dx
a
Z ∞
Expectation: E(X) = xf (x) dx
Z −∞

Variance: Var(X) = x2 f (x) dx − µ2

Mathematical Expectation
E(aX + b) = aE(X) + b, Var(aX + b) = a2 Var(X)
E(X + Y ) = E(X) + E(Y ); If independent: E(XY ) = E(X) · E(Y )

Binomial Distribution

Key Formulae
X ∼ B(n, p): n independent trials, probability of success p, failure q = 1 − p.
 
n r n−r
P (X = r) = p q , r = 0, 1, 2, . . . , n
r

Mean Variance SD
√
µ = np σ 2 = npq σ= npq
Use when: Trials are fixed, independent, binary outcomes, constant p.

Poisson Distribution

Key Formulae
X ∼ Poisson(λ): Models rare events in a fixed interval.

e−λ λr
P (X = r) = , r = 0, 1, 2, . . .
r!

Mean Variance Note

µ=λ σ2 = λ Mean = Variance
Poisson as limit of Binomial: when n → ∞, p → 0, np = λ (constant).

7
15 Questions — Exam Ready Statistics & Probability — Quick Notes

Tests of Hypothesis

Hypothesis Testing Framework

1. State H0 (null) and H1 (alternative) hypothesis.
2. Choose significance level α (usually 5% or 1%).
3. Compute test statistic.
4. Compare with critical value (or use p-value).
5. Reject H0 if test statistic > critical value.
Type I Error: Rejecting true H0 (probability = α).
Type II Error: Accepting false H0 (probability = β).

Large Sample Tests (n ≥ 30) — Z-test

Key Formulae
Test for Single Mean:
x̄ − µ
Z= √
σ/ n
Test for Difference of Means:
x̄1 − x̄2
Z=r 2
σ1 σ22
+
n1 n2
Test for Proportion:
p̂ − p0
Z=p
p0 q0 /n
Critical values: |Z| > 1.96 (5%), |Z| > 2.58 (1%) ⇒ Reject H0 .

Small Sample Tests (n < 30) — t-test

Key Formulae
One sample t-test:
rP
x̄ − µ (x − x̄)2
t= √ , s= , df = n − 1
s/ n n−1

Two sample (independent) t-test:

s
x̄1 − x̄2 (n1 − 1)s21 + (n2 − 1)s22
t= q , Sp =
Sp n11 + n12 n1 + n2 − 2

Paired t-test: sP
d¯ d2i − nd¯2
P
di
t= √ , d¯ = , sd =
sd / n n n−1
Compare computed |t| with table value at α and df.

8
15 Questions — Exam Ready Statistics & Probability — Quick Notes

Chi-Square Test (χ2 )

Key Formulae
X (O − E)2
χ2 =
E
(Row Totali ) × (Col Totalj )
Expected frequency: Eij =
Grand Total
Degrees of freedom: df = (r − 1)(c − 1) for r rows, c columns.
Uses: Test of independence (contingency table), goodness of fit.
Condition: Each expected frequency ≥ 5.

Analysis of Variance (ANOVA) — One Way

Key Formulae
Tests whether means of k groups are equal.
H0 : µ1 = µ2 = · · · = µk H1 : At least one mean differs.

MST (Between groups)

F =
MSE (Within groups)

Source SS df MS F
SST MST
Between (Treatment) SST k − 1 MST = k−1 MSE
SSE
Within (Error) SSE N − k MSE = N −k
Total TSS N − 1
TSS = SST + SSE
If computed F > table Fα (k − 1, N − k), reject H0 .
T2 PP
Correction Factor: CF = , where T = xij
N

All topics covered. Good luck in your exam!