foresight school

Revision Guide CFA L1

Quantitative Methods

foresight school
[Link]

The Time Value of Money

© 2017 FinTree Education Pvt. Ltd.

LOS a
Interest rate can be interpreted as - Required rate of return, Discount rate
or Opportunity cost

LOS b 1 International Fischer Relationship (approx.)

@ 10% p.a.
100 110

Consumption cost - True saving -

107 3

Inflation Real rate of

return

Nominal risk-free rate = Real risk-free rate + Expected inflation

2 Treasury bonds = Real RFR + Expected inflation

Corporate bonds = Real RFR + Expected inflation + Risk premium
Return on Non-investment grade bond > Return on Investment grade bond, because
risk of Non-investment grade bond > risk of Investment grade bond

3 Types of risks

Default Liquidity Maturity

risk risk risk

Risk that borrower Risk of receiving Risk of volatility

will not make less than FV for an of price of a bond
promised payments investment if it because of its
in a timely manner must be sold for longer maturity
cash quickly

Default risk Liquidity risk Maturity risk

premium premium premium

Low default rate Less liquidity Shorter maturity

= = =
Low DRP High LRP Low MRP

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS c Calculation and interpretation of effective annual rate
The rate of interest that an investor actually earns as a result of
compounding is known as EAR
Effective Annual Rate = 1 + (Int. rate/m)m - 1
m = compounding frequencies per year
EAR on TI BA II Plus Professional - 2nd 2

LOS d TVM with different compounding frequencies

Annual Semiannual Quarterly Monthly

N=1 N=1X2=2 N=1X4=4 N = 1 X 12 = 12

I/Y = 13.25 I/Y = 13.25/2 = 6.625 I/Y = 13.25/4 = 3.3125 I/Y = 13.25/12 = 1.104
PV = -100 PV = -100 PV = -100 PV = -100
FV = 113.25 FV = 113.68 FV = 113.92 FV = 114.08

LOS e 1 Annuity
It is a stream of equal cash flows occurring at equal intervals.

Ordinary annuity Annuity due

End mode Beginning mode

0 1 2 3 0 1 2 3

100 100 100 100 100 100 100

PV of perpetuity = CF
2
Disc. rate

LOS f 1 Amortization schedule

Loan - 100,000 Int. rate - 10% Tenure - 4 yrs.

Year Opening Instalment Interest Principal Closing loan

loan (Op. loan X repayment (Op. loan -
rate of int.) (Inst - Int.) Princ.
repayment)

1 100,000 31,547 10,000 21,547 78,452

2 78,452 31,547 7,845 23,701 54,751

3 54,751 31,547 5,475 26,071 28,679

4 28,679 31,547 2,868 28,679 -

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.

2 Using amort function in TI BA II plus professional

Ÿ 2nd CLR TVM (FV)
Ÿ PV = -100,000 N = 4 I/Y = 10 CPT PMT = 31,547
Ÿ 2nd AMORT (PV)
Ÿ 2nd CLR WORK (CE|C)
Ÿ P1 = 1
Ÿ P2 = 1
Ÿ BAL = 78,452
Ÿ PRN = 21,547
Ÿ INT = 10,000
Ÿ P1 = 2
Ÿ P2 = 2

3 PV and FV of uneven CFs

Eg. Discount rate = 10% N = 6 years

CFs: Year 1 = −1,000 Year 2 = −500 Year 3 = 0 Year 4 = 4,000 Year 5 = 3,500 Year 6 = 2,000

Using CF function in TI BA II plus professional

Ÿ CF
Ÿ 2nd CLR WORK (CE|C)
Ÿ CF0 = 0
Ÿ CF1 = −1,000
Ÿ CF2 = −500
Ÿ CF3 = 0
Ÿ CF4 = 4,000
Ÿ CF5 = 3,500
Ÿ CF6 = 2,000
Ÿ I = 10 → Enter → ↓ (down key)
Ÿ CPT NPV = 4711.91

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.

Discounted Cash Flow Applications

LOS a
NPV PV of inflows − PV of outflows

IRR Rate at which PV of inflows = PV of outflows

At IRR, NPV = 0

LOS b Decision rule

NPV IRR

+ve = Accept
If IRR > WACC = Accept
−ve = Reject
If IRR < WACC = Reject
Mutually exclusive projects -
Accept project with highest NPV

For a single project NPV and IRR rules lead to same accept/reject decision
If IRR > WACC, NPV =+ve
If IRR < WACC, NPV =−ve

LOS c Holding period return (HPR)

Ending value − beginning value Or Ending value − 1
Beginning value Beginning value

Total return
Ending value - beginning value + CF received Or Ending value + CF received − 1
Beginning value Beginning value

LOS d Money-weighted rate of return Time-weighted rate of return

(MWRR) (TWRR)

IRR Geometric mean of HPR

Appropriate if manager has complete Provides better measure of manager’s

control over inflows and outflows ability to select investments

! TWRR is not affected by timing of the cash flows, therefore it is more preferred method of
performance measurement
! If funds are contributed to a portfolio just prior to a period of relatively poor performance,
MWRR < TWRR
! If funds are contributed to a portfolio just prior to a period of relatively high returns,

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS e & f 3 mistakes analogy to
remember the formulas
Effective earning yield (indicated in red) Bond equivalent yield

(1+3.09%)365/90- 1 j Compounding j No Compounding

90 3.09%
= 13.13% k 365 days k 365 days
l Investment 365 12.53% l Investment
Or value as base value as base

N = 90/365 Holding period yield

PV = -970
FV = 1000
I/Y = 13.14 % T - bill
970 1000
3.09
90 days
1 + 3.09% 30
= 3.09%
970

Bank discount yield Money market yield

j No Compounding j No Compounding
90 30/1000 =3% k 360 days 90 3.09% k 360 days
l Face value as 360 12.36% l Investment value
360 12%
base 3 as base
1 − 3%

foresight school
 Statistical Concepts and Market Returns
[Link] © 2017 FinTree Education Pvt. Ltd.

LOS a
1 Descriptive statistics Inferential statistics

Used to summarize important Used to make forecasts of

characteristics of large data large data

Eg. Average of weekly tests Eg. Forecast on pass or not

2 Population Sample

Set of all possible members of a stated group Subset of population

CFA level 1 candidates globally CFA level 1 candidates in class

3 Types of measurement scales

Nominal Ordinal Interval Ratio

Provides relative
ranking and Most refined level
Higher level of assurance that of measurement
measurement than differences
Contains least nominal scales between scale Provides ranking
information values are equal and equal
Observation is differences
Classification assigned to a Weakness - Zero between scale
has no category doesn’t mean total values
particular order absence
Eg. MF’s star Has a true zero
rating Eg. Temperature point as origin
measurement

LOS b Parameter Sample statistic Frequency distribution -

Tabular presentation of statistical data

Measure used to It is used to measure
describe a a characteristic of a Data employed with a frequency
characteristic of a sample distribution may be measured using any
population type of measurement scale

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS c Relative frequency and cumulative relative frequency

Interval / Frequency Cumulative Relative frequency Cumulative

class frequency relative frequency

10 - 15 7 7 (7/50) 14% 14%

15 - 20 12 19 (12/50) 24% 38%

20 - 25 21 40 (21/50) 42% 80%

25 - 30 10 50 (10/50) 20% 100%

Total 50 100%

LOS d Histogram and Frequency polygon

Frequency Frequency

Interval
Interval
midpoints
Histogram Frequency polygon

LOS e Measures of central tendency

1 Mean

Arithmetic mean Weighted mean Geometric mean Harmonic mean

AM = 10 + 14 + 4 + 8 WM = 10(20%) + GM = HM =
4 14(20%) + 4
4
4(35%) + √1.1 X 1.14 X 1.04 X 1.08 − 1 1/10 + 1/14 + 1/4 +1/8
8(25%)

AM = 9 WM = 8.2 GM = 8.94 HM = 7.32

ª Sum of deviations from arithmetic mean is always zero

ª To calculate portfolio return, weighted mean is used

ª Geometric mean is used for calculating investment returns over multiple periods

ª Harmonic mean is used to calculate average of ratios

ª Arithmetic mean > Geometric mean > Harmonic mean

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.

2 Median It is the midpoint of a data set

Median = [(n+1) X 50%]th observation
Data needs to be arranged in ascending order
to calculate median using above formula

1 2 3 4 5 6 7 8 9
Median = [(9+1) X 50%] = 5th observation

3 Mode Value that occurs most frequently in a data set

A data set can have more than one mode or
even no mode
If a data set has one/two/three modes it is
said to be unimodal/bimodal/trimodal

LOS f Quartiles, quintiles, deciles, percentiles

Quartiles Quintiles Deciles Percentiles

Distribution is Distribution is Distribution is Distribution is

divided into divided into divided into divided into
quarters fifths tenths hundreds

[(n + 1) × 25%]th [(n + 1) × 20%]th [(n + 1) × 10%]th [(n + 1) × 1%]th

LOS g Measures of dispersion

Range Mean absolute Variance Standard deviation

deviation
(MAD)

Maximum value ∑|(x − x)| Population variance - Population SD -

− minimum n
value ∑ (x − μ)2
n
Sample SD -
Sample variance -

∑ (x − x)2
n−1

Variance = σ2

SD can be calculated directly on TI BA II plus professional.

Ÿ Use DATA (2nd 7) to enter data then,
Ÿ Use STAT (2nd 8) to see SD

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS h Chebyshev’s inequality

Applies to sample or population data, normal or skewed distribution

Calculated as,

1 − 1/k2 where k > 1

60 70 80
Eg. SD = 5 Chebyshev’s inequality = 1 − 1/22

= 1 − 1/4
K = 10/SD
K=2 = 75%
Interpretation: 75% observations lie within ±2 SD of mean

LOS i Coefficient of variation (CV) Sharpe ratio

It is used to measure excess

It is used to measure return per unit of risk
the risk per unit of
expected return aka reward-to-variability
ratio
CV = SDx
X SR = Portfolio return − RFR
SD of portfolio
Lower the better
Higher the better

Sharpe ratio

Motorcycle takes 3 ltrs of petrol Motorcycle takes 2.2 ltrs of petrol

to cover the entire distance to cover the entire distance
10

10
km

km

RFR 15 km RFR 10 km
Which is more economical ?

Wrong interpretation - 25/3 = 8.33 ✘ 20/2.2 = 9.09 ✓

Correct interpretation - 15/3 = 5 ✓ 10/2.2 = 4.54 ✘

25 − 10 Rp − RFR 20 − 10
Sharpe ratio = 3 2.2
SDp

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS j Skewness

Positively skewed/ Negatively skewed/

No skew
right skew left skew

Normal distribution Skewed distribution

Symmetrical Asymmetrical
distribution distribution

Skewness: Extent to which data is not symmetrical

Negative skew in returns distributions indicates increased risk

LOS k Locations of mean, median and mode

Mean, Mode Mean Mean Mode

median,
mode Median Median

Mean = Median = Mode Mean > Median > Mode Mean < Median < Mode

LOS l Kurtosis

Mesokurtic Leptokurtic Platykurtic

distribution distribution distribution

Kurtosis = 3 Kurtosis > 3 Kurtosis < 3

Excess kurtosis = 0 Excess kurtosis = +ve Excess kurtosis = -ve

Kurtosis: Measures the peakedness of a distribution

Positive kurtosis in returns distributions indicates increased risk
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
2 3
Sample skewness - Sk = ∑(X-X) x 1 Sk > 0.5 indicates significant level of skewness
SD3 n
4
Sample kurtosis - Sk = ∑(X-X)
4
x 1 Excess kurtosis > 1 is considered a large value
SD n

LOS m Use of arithmetic mean and geometric

mean when analyzing investment returns

Arithmetic mean return is appropriate for forecasting single period returns in future periods

Geometric mean return is appropriate for forecasting future compound returns over multiple periods

foresight school
[Link]

Probability Concepts
© 2017 FinTree Education Pvt. Ltd.

LOS a
Mutually Exhaustive
Random variable Outcome Event exclusive events
events

Uncertain quantity/ Observed value An outcome or Events that can All possible events
number of a random a set of not happen
variable outcomes together

LOS b 1 Two defining properties of probability

è Probability is always between 0 & 1

è If we have mutually exclusive and exhaustive events then sum of

probabilities of those events will always be 1

2 Probabilities

Objective Subjective

Empirical Priori
Least formal
method of
Established by Determined using developing
analyzing historical formal reasoning probabilities
data
Involves personal
Eg. Historical pass Eg. Throwing a die judgement
rates = 1/6

LOS c Probability of an event in terms of odds

If probability of an Then for 10 experiments,
event is 20% success = 2 failure = 8
2/10

Odds for Odds against

2/8 8/2

Two-to-eight Eight-to-two

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS d Unconditional and conditional probabilities

Unconditional Conditional

µ Refers to probability of an event µ Occurrence of one event affects the

regardless of occurrence of other probability of occurrence of other
events event

µ Also known as marginal probability µ A conditional probability of an

occurrence is also called its likelihood
Eg.
Eg.
µ P(heads) = 50%
µ P(head/rains) =50% µ P(pass/study) = 80%
µ P(pass/studyc) = 50%

LOS e Multiplication, addition and total probability rules

Multiplication rule Addition rule

Used to determine the joint Used to determine the probability

probability of two events that at least one of two events
will occur

Apply this rule when a Apply this rule when a

question says ‘and’ question says ‘or’
P(AB) = P(A|B) × P(B) P(A or B) = P(A) + P(B) − P(AB)

P(A|B) = P(AB)
P(B)

Total probability rule - Used to determine unconditional probability of an event,

given conditional probabilities
P(A) = P(A|B1) × P(B1) + P (A|B2) x P(B2) +....... P(A|Bn) × P(Bn)

Joint probability - Probability that all the events will occur at the same time
LOS f
For mutually exclusive events the joint probability is zero

For events that are not mutually exclusive, joint probability must be subtracted
from the total of unconditional probabilities to avoid double counting

Eg. P(A) = 60% P(B) = 30%

P(Both) = 60% × 30% = 18%

P(A) P(B)
P(At least one) = 60% + 30% − 18% = 72%

P(None) = 1 − 72% = 28% Or

P(AB)
(1 − 60%) × (1 − 30%) = 28%
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS g Dependent and independent events

Independent events Dependent events

Ÿ Occurrence of one event has no Ÿ Occurrence of one event is dependent

influence on occurrence of other events on occurrence of other events

Ÿ P (A|B) = P(A)

Ÿ Getting 5 on a second roll of die is

independent of getting 5 on the first roll
of die

LOS h, i, j 1 Unconditional probability using total probability rule

Interest rates Prob of good

increase economy and
rate increase
Prob = 75%
0.4 x 0.75 = 30%
Good economy next year

Prob = 40% Prob of good

Interest rates
economy and
decrease
rate decrease
Prob = 25% 0.4 x 0.25 = 10%

Unconditional Conditional Joint

probability probability probability

2 Expected value

60% EPS = 10 0.4 × 0.6 × 10 = 2.4

%
= 40
od)
P(Go 40%
Economy EPS = 8 0.4 × 0.4 × 8 = 1.28
P(
Go
od c EPS = 6 0.6 × 0.7 × 6 = 2.52
)= 70%
60
%

30
%
EPS = 3 0.6 x 0.3 x 3 = 0.54

Expected value of EPS = 2.4 + 1.28 + 2.52 + 0.54

= 6.74

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS k Covariance and correlation
Covariance Correlation
µ It is a measure of how two assets move µ Standardized measure of covariance
together
µ Measures strength of linear relationship
µ Covariance of return with itself is its between two random variables
variance
µ Does not have a unit
µ Expressed in terms of square units
µ r = Cov(x,y)
µ Population Cov(x,y) = ∑(X − X) (Y − Y) σx × σ y
n
µ Does not exhibit causal relationship
µ Sample Cov(x,y) = ∑(X − X) (Y − Y)
n−1 µ Range = −1 to +1

µ Cov(x,y) = r × σx × σy µ r = 1 means perfectly +ve relation

µ r = 0 means no correlation
µ Range = −∞ to +∞ µ r = −1 means perfectly −ve relation

µ Only +ve and −ve sign matters for

determining relationship b/w the variables

LOS l Expected value, variance and standard deviation of portfolio

1 Expected value = W1E(R1) + W2E(R2) +W3E(R3) +.......+WnE(Rn)

Variance = (W1σ1)2 + (W2σ2)2 + 2W1σ1W2σ2 × r
(W1σ1)2 + (W2σ2)2 + 2W1W2 × Cov(x,y)

Standard deviation = √Variance

2 When r = −1, When r = 0, When r = 1,

Sdp = (W1σ1) − (W2σ2) Sdp= √(W1σ1) + (W2σ2)
2 2
Sdp = (W1σ1) + (W2σ2)
Sdp = Lowest Sdp = Highest

LOS n Bayes’ formula

It is used to calculate updates probability

Eg. P(A|B) = 30% P(Ac|Bc) = 40% P(B) = 30% P(B|Ac) = ?

% B A = 9%
=30
P(A)
%
= 30
P(B) P(A c
) =7
0% B Ac = 21% B Ac =
21
P( 21+28
B )c
=60% Bc A = 42%
=
70 P(A) = 42.86%
%

P(
A )c
=
40
%
Bc Ac = 28%
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS o Counting problems

Labeling Permutation Combination

n! n
n
Pr Cr = nCn-r
n1! x n2! x .... x nk!

Eg. A person has 8 cars. He uses 3 cars Eg. How many different Eg. How many different
for work, 3 other for long distance trips ways are there to select ways are there to select
and 2 other for commute other than 3 players from 5, if the 3 players from 5, if the
work. Calculate the no. of different order of selection is order of selection is not
ways to label them. important ? important ?

8! 5 → 2nd(−) nPr → 3 5 → 2nd(+) nCr → 3 or 2 (5−3)

3! x 3! x 2!
= 60 = 10
= 560

Permutation is used when order of selection is important

Combination is used when order of selection is not important

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.

Common Probability Distribution

LOS a
Probability distribution - Describes the probabilities of all possible outcomes
of a random variable. Probabilities of all outcomes
should equal to 1

Discrete random variable - There is a finite number of possible outcomes. Eg.

number of stocks in portfolio

Continuous random variable - There is an infinite number of possible outcomes. Eg.

Return earned in portfolio

No. of stocks % return

100 200 300 0.001 %

Discrete random variable Continuous random variable

LOS b Probability Function - P(X) = P(X=x)

Eg. X =1,2,3,4 P(X) = x else P(X) = 0

10
The above function satisfies both the conditions of probability which are ;
a) 0 ≤ P(x) ≤ 1
b) ∑ P(x) = 1

LOS c
1 Discrete non - uniform Discrete uniform
variable variable

0 110 1 2 3 4 5 6

Number of people present in class Probability distribution of a roll of a die

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.

2 Continuous non - uniform Continuous uniform

variable variable

Eg. Puncture on tyre. There are infinite

P(X) = 0, P(x) is between x1 and x2 number of points on a tyre where
puncture can happen. Each point has
equal probability of occurance

LOS c & d Cumulative distribution function (cdf)

Eg. X =1,2,3,4 P(X) = x
10
Probability Cumulative
density function distribution function
P(1) = 1/10 = 10% F(1) = 10%
P(2) = 2/10 = 20% F(2) = 30%
P(3) = 3/10 = 30% F(3) = 60%
P(4) = 4/10 = 40% F(4) = 100%

F(-1) = 0.1587

0.1587

−1 0 +1 −1 0

LOS e & f
Binomial random Eg.
variable
P(win) = 70%, 4 matches, exactly 2 wins

Outcome can be either Px x (1-P)n-x x nCr

‘success’ or ‘failure’
(70%)2 x (1−0.7)4-2 x 4C2

When number of trials is 1, = 26.46%

it is called Bernoulli
random variable
Ÿ Mean of binomial distribution = np

Px x (1−P)n-x x nCr Ÿ Variance of binomial distribution = npq

Ÿ q=1−P
P = Probability of success
n = No. of trials
X = No. of successes
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS g Binomial tree
Stock price (S) = 850 Uptick (u) = 1.2 Downtick (d) = 1/U = 1/1.2

Su 1020 x 1.2
= 1224
Suu
850 x 1.2
= 1020
1020 x 1/1.2
Sud
= 850
S 850
708 x 1.2 =
850
Sdu
850 x 1/1.2
= 708
708 x 1/1.2
Sd = 590
Sdd

LOS h Tracking error = Return on portfolio − Return on benchmark

LOS i Continuous uniform distribution

Properties of continuous uniform distribution

ª For all a < x1 < x2 < b
(i.e. for all x1 and x2 between the boundaries a and b)

e
ª P(X < a or X > b) = 0
(i.e. probability of X outside the boundaries is zero)
re
ª P(x1 < X < x2) = (x2 - x1)/b - a
(This defines the probability of outcomes between x1 and x2)

ª Continuous uniform distribution will always have lower and upper bound (a,b)
ª Probability of X taking any value below ‘a’ or above ‘b’ will be zero

Eg. X is uniformly distributed between 2 & 20. Calculate the probability that X will be between 6 & 15.
nT

6 15

2 20

P(4<8) = 15 − 6
20 − 2

P(x) = 0 Because it is a continuous distribution

Fi

LOS j Normal distribution

è It is a continuous non-uniform distribution

è Mean and variance needs to be defined
è Skewness = 0
è Kurtosis = 3
è Mean = median = mode
è A linear combination of a normally distributed random variable is also normally distributed
è The probabilities of outcome further above and below mean get smaller and smaller but do
not go to zero (i.e. the tails get very thin but extend infinitely
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS k Univeriate Multiveriate
distribution distribution

Single variable More than one variables

A multivariate distribution with 10 variables has - 10 means 10 Variances 45 Correlations

n x (n-1) = 10 x 9 = 45
2 2

LOS l Confidence interval

Eg. X = 700 σ = 200

Calculate 90%, 95%, 99%
confidence interval

34% 34%
X ± (z-value)σ
13% 13%
90% - 700 ± (1.65)200
= 370-1030
3% 3%
95% - 700 ± (1.96)200
-3σ -2σ -1σ 1σ 2σ 3σ = 308-1092
68%
95% 99% - 700 ± (2.58)200
= 184-1216
99%
Interpretation: We are 99% of the time
confident that the expected outcome will
lie between 184 and 1216

LOS m Standard normal distribution

Standard normal It is a normal distribution that is standardized so that its

distribution - mean = 0 and standard deviation = 1

Z-value = Observation − Population mean

standard deviation

Eg. X = 400 σ = 200 μ = 700

400 − 700 = −1.5 At 1.5 Z-value, Probability = 93.32%

Z score =
200
Therefore probability of value less
than 400 = 1 − 0.9332 = 6.68%

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS n Shortfall risk and Safety first ratio

Shortfall risk Safety first ratio

Probability that portfolio value Excess return per unit of risk

or return will fall below a over minimum acceptable
particular value or target over a return/threshold level.
given period of time
SF ratio = Rp − Threshold return
Sdp

Lower the better Higher the better

Eg. Average return = 20% SD = 3% Threshold level = 15%

Z-value = 15 − 20 = −1.66
Shortfall 3
SF ratio = 20 - 15
risk 3
Probability at −1.66 z-value = 95.15%
Therefore shortfall risk ; = 1.66
1 − 0.9515 = 4.85%

15% 20%

LOS o Normal and lognormal distribution

Normal distribution Lognormal distribution

Ÿ No skew Ÿ Skewed to the right

Ÿ Not bounded by zero Ÿ Bounded by zero
Ÿ Useful for modeling asset prices,
because they can not take
negative values

The logarithms of lognormally distributed random variables are normally distributred

LOS p Discrete and continuous compounding

Discrete compounding - Annual, semi-annual, quarterly, monthly etc.

Continuous compounding - No. of compounding periods within a given time period

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
Continuous compounding calculations

100 20% 117.35 100 20% 117.35 100 20% 117.35

0 0.8 0 0.8 0 0.8

100 x e0.2 x 0.8 = 117.35 ln 117.35 0.8 → 16% 117.35 x e-0.2 x 0.8 = 100
100
1 → 20%

LOS q Monte Carlo simulation

Technique based on repeated generation of one or more risk factors that

affect security values, to generate a distribution

It is used to Its limitations are

Ÿ Value complex securities Ÿ It is complex

Ÿ Simulate profits/losses from a trading strategy Ÿ It is subject to model risk and input risk
Ÿ Calculate estimates of VaR to determine the Ÿ Simulation is not an analytic method, but a
riskiness of a portfolio statistic one.
Ÿ Simulate pension fund assets and liabilities to Ÿ Increased complexity does not necessarily
examine the variability of the differences ensure accuracy
between the two
Ÿ Value portfolios of assets that do not have
normal returns distribution

LOS r Historical simulation

It is based on actual change in value or actual change in risk factor for some prior period

Each iteration of simulation involves randomly selecting one of these past changes for
each risk factor and calculating the value of the asset or portfolio in question, based
on those changes in risk factor

Its advantage is that it uses actual distribution of risk factors, which need not be estimated.

Its limitations are :

Past changes in risk factor may not be a good indication of future changes
It can not address the sort of ‘whatif’ questions that Monte Carlo simulation can

foresight school
 Sampling and Estimation
[Link] © 2017 FinTree Education Pvt. Ltd.

LOS a Simple random sampling and sampling distribution

1 Simple random sampling Systematic sampling

Method of selecting a sample in Another way to form an

such a way that each item in the approximately random sample.
population has same likelihood
of being included in the sample.

Eg. Drawing a sample of 5 Eg. Selecting every nth item

apples from 50 to calculate from the population
average weight.

2 Sampling distribution - It is a probability distribution of all possible sample statistic

computed from samples drawn from the population

Sampling distribution does not have to be normal distribution

LOS b Sampling error = Sample statistic − Population parameter

Mean, Variance, Mean, Variance,

Standard Deviation of Standard Deviation of
sample population

LOS c Stratified random sampling - Uses a classification system to separate the

population into small groups, based on one
or more distinguishing characteristics. Each
subgroup is called as stratum.

Eg. Avg. calorie intake of a nation Sample4 Results of these samples

are then pooled to form a
Sample5 combined sample
N

Sample3 W C E Sample1
S

Sample2

It is often used in bond indexing because of the difficulty and cost of replicating entire
population of bonds.

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS d Time-series and Cross-sectional data

Time-series data Cross-sectional data

It consists of observations taken It consists of observations taken

over a period of time at a single point in time

Time-series and cross-sectional data can be pooled in the same data set.

Longitudinal data - Observations over time of multiple characteristics of the

same entity. Eg. Unemployment, GDP growth rates,
inflation of a country over 10 years.

Panel data - Observations over time of same characteristic of the

multiple entities. Eg. analysis of D/E ratio of 20
companies over 8 quarters.

Panel and longitudinal data are typically presented in table or spreadsheat form.

LOS e Central limit theorem

ª Sample mean(x) approaches population mean(μ) as sample size becomes large

ª Variance equals ‘σ2/n’ as sample size becomes large

ª If sample size n, is sufficiently large (n ≥ 30), the sampling distribution of the sample
means will be approximately normal

ª If central limit theorem works, population mean(μ) = mean of sampling distribution

ª Standard deviation of sampling distribution = σ/√n (standard error)

LOS f Standard error of sample mean

Population variance known Population variance unknown

σ s
√n √n

LOS g Describe properties of an estimator

ª Unbiasedness - It is one for which the expected value of the estimator is equal to the
parameter you are trying to estimate

ª Efficiency - Unbiased estimator is also efficient if the variance of its sampling distribution
is smaller than other unbiased estimators of parameter you are trying to estimate

ª Consistency - An estimator for which the accuracy of the parameter estimate increases as
the sample size increases

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS h Point estimate and confidence interval estimate

Point estimate - It is a single sample value used to estimate population

parameter
Confidence interval It is a range of values in which population parameter is
estimate - expected to lie.

LOS i
Student’s t-distribution Properties of t-distribution

è It is a bell-shaped probability distribution ª It is defined by degrees of freedom(DoF) (n − 1)

ª It has more probability in the tails (fat tails)
è It is symmetrical about its mean ª As DoF increase, t-distribution approaches
standard normal distribution (z-distribution)
è It is appropriate to use when n < 30, ª t-distribution is flatter and has fatter tails than
population variance is unknown and normal distribution
distribution is normal ª As number of observations increase, distribution
becomes more peaked and tails become thin i.e.
it converges to z-distribution

LOS j Computation and interpretation of confidence interval

1 Significance level (α) = 1 − Confidence interval

90% confidence level = 10% significance level = 5% in each tail

2 Construction of confidence interval

Point estimate ± (Reliability factor × Standard error)

3 Selection of test for

reliability factor

Population Population
variance is variance is
known unknown

Normal Non -normal Normal Non -normal

distribution distribution distribution distribution

Z - distribution t - distribution
n ≥ 30 n < 30 n ≥ 30 n < 30

Z - distribution No t/z distribution No

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS k Appropriate sample size and different biases
Sample size

ª Generally large sample size is better (consistency property of an estimator)

ª 2 issues with larger sample size;

ΠLarger samples may contain observations from a different population, which

may not improve precision of our population parameter estimates.

 Larger sample size may not be cost effective

Different types of biases

Data mining · Data mining occurs when analysts use same database to search for patterns until one
that works for them is discovered.
· Data mining bias refers to overestimated results because it involves data mining.
· Lack of an economic theory that is consistent with the empirical results.
· To avoid data mining, test the potentially profitable trading value on a data set
different from the one used to develop the rule
Sample selection · Some data is systematically excluded from the analysis, usually because of lack of
bias availability.

Survivorship bias · Most common form of bias.

· Most mutual fund databases only include the existing funds(survivors).

Look-ahead bias · Occurs when a study tests a relationship using sample data that was not available on
the test date.
MPS Available at the end of accounting period
BVPS Not available at the end of accounting period
Therefore it is estimated

Time-period bias · Can result if the time period over which the data is gathered is either too short or too
long.

foresight school
 Hypothesis Testing
[Link] © 2017 FinTree Education Pvt. Ltd.

LOS a
1 Hypothesis - It is a statement about the value of a population parameter developed for the
purpose of testing a theory or belief.

2 Steps of hypothesis testing 3 Null and Alternative hypothesis

State the hypothesis
(Defining null and alternative)

Null Hypothesis Alternative Hypothesis

Calculate the appropriate test
statistic Hypothesis that the Hoped for outcome
researcher wants to
reject

Specify level of significance Designated as H0 Designated as Ha

Has ‘=’ ‘≥’ or ‘≤’sign Has ‘>’ ‘<’ or ‘≠’ sign

State the decision rule Eg. μ = 25 Eg. μ < 25

Calculate sample statistic

Make decision regarding

hypothesis

LOS b

One-tailed test Two-tailed test

Look for ‘<’ or ‘>’

To determine whether left tail or right

tail, look for the sign of Ha
Look for ‘=’
Critical value;
H0: μ = Hypothesized value
Left tail = ‘-’
Ha: μ ≠ Hypothesized value
Right tail = ‘+’
In practice, most hypothesis tests are
H0: μ ≥ Hypothesized value
constructed as two-tailed tests
Ha: μ < Hypothesized value

For above, Left tail test should be used

since the sign of Ha is towards left

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS c
1 Test statistic - Sample statistic − Hypothesized value = X − HV
Standard error σ/√n

2 Type I and Type II errors

Type I error - Rejecting the null hypothesis when it is actually true

Type II error - Failure to reject the null hypothesis when it is actually false

ª Any reduction in the probability of Type I error comes at the

cost of increase in probability of Type II error (vice versa)

ª Probabilities of both Type I and Type II errors can be

reduced simultaneously by increasing the sample size

3 Significance level - It is the probability of making Type I error

P(Type I error) = Significance level (α)

1 − P(Type I error) = Confidence level

LOS d 1 Power of test = 1 − P(Type II error)

2 Relation between confidence interval and hypothesis testing

Eg. Sample Mean = 100 Sample standard deviation = 40 Sample size= 36

Calculate 95% confidence interval & decide whether hypothesis μ = 75 should be rejected

Calculation - Null hypothesis H0: μ = 75 Alternate hypothesis Ha: μ ≠ 75

Test statistic = Sample mean − Hypothesized value = 100 − 75 = 3.75

Standard error 40/√36

Confidence interval = 100 ± )√36 )

40 × 2.03 = 86.45 to 113.54

Since population variance is unknown and distribution is normal, t-test should be used

-2.03 2.03 DoF = 35, α = 0.05

3.75 is outside the range

Therefore reject the null hypothesis

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS e Statistical results and economically meaningful results
Statistical significance does not necessarily imply economic significance.

Reasons - Transaction cost, taxes, risk

LOS f P-value
Reject

Accept Accept Reject

Accept

5 ft. 3.8 ft. 4 ft. 6 ft. 4.5 ft. 5 ft.

Significance level

P-value is the lowest level of significance at which null hypothesis is rejected.

LOS g Hypothesis testing

Population Population
variance is variance is
known unknown

Normal Non - normal Normal Non - normal

distribution distribution distribution distribution

Z - distribution t - distribution
n ≥ 30 n < 30 n ≥ 30 n < 30

Z - distribution No t/z distribution No

Parametric Non parametric

LOS h, i, j test test

t-test
Chi-square Tests concerning a Wilcoxon signed
Variance(sample) single mean rank test
test = X (n-1) z-test
Variance(population)
DoF = n − 1 Two tests t-test
concerning Mann - Whitney
differences Approximate t- U test
Variance(larger)
F test = between mean test
Variance(smaller)
Test concerning Wilcoxon signed
DoF = n1 − 1 and n2 − 1 mean difference rank test
t-test
(Paired comparison
test) Sign test
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
Test of difference between mean

Independent Dependent
sample sample

Variance assumed Variance assumed

to be equal to be unequal

ª Normally distributed ª Normally distributed

population population
Paired comparison test
ª DoF - n1 + n2 − 2 ª Large DoF formula

LOS k
Parametric test - It is concerned with parameter and its validity

t-test, F-test and chi-square tests are parametric tests

Non parametric test - It is used when assumptions of parametric test can’t be supported

It is also used when data is given in ranks or hypothesis is not

with respect to parameter

Spearman’s rank correlation can be used when data are not normally distributed

foresight school
 Technical Analysis
[Link] © 2017 FinTree Education Pvt. Ltd.

LOS a Principles of technical analysis

Technical analysis It is the study of collective market sentiment
Prices are determined by interaction of supply &
demand
Key assumption of technical analysis is that efficient
markets hypothesis (EMH) does not hold
Usefulness is limited in illiquid markets as well as
markets subject to large outside manipulation

Technical Analysis Fundamental Analysis

Uses share price & trading
volume to project a target price Attempts to determine intrinsic
value
Advantages:
Uses financial statements &
ΠData is observable other information
 Can be applied on assets
without CFs

LOS b Types of price charts

Candlestick Point & figure

Line charts Bar charts
charts charts

Same data as bar

Opening & closing & charts
high & low prices
Use boxes shaped as ∆ in the direction of
Use cross-hatches & candle’s body for price
vertical lines as opening & closing
Show closing prices as symbols price Horizontal axis
data points on a reflects no. of ∆ in
continuous line Op. price - Initiated as Box is clear if closing price not time
a point on the right price > opening price
side of the line Price change
Box is filled if closing represents the height
Cl. price - Initiated as price < opening price which is ‘box size’
a point on the left side
of the line Price moves are much
more visible

ª Relative strength analysis - Asset closing price/Benchmark value

ª Ç in trend - Asset is outperforming
ª È in trend - Asset is underperforming

ª Volume chart - Usually included at the bottom of many charts. Volume is on vertical axis

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS c Trend in prices

Uptrend Downtrend

Reaching higher highs & Lower lows & retracing lower

retracing higher lows highs

Shows demand is increasing Shows supply is increasing

relative to supply relative to demand

Trend line connects Trend line connects

increasing lows increasing highs

Breakdown below uptrend Breakdown above uptrend

line (significant price change) line (significant price change)

Support Level - Price range in which buying activity is sufficient to stop decline in price

Change in polarity - Breached resistance levels become support levels & vice versa

Resistance Level - Price range in which selling activity is sufficient to stop rise in price

LOS d Common chart patterns

Reversal Double top Continuation

patterns & triple top patterns
Signals the end of a trend
Indicate weakening buying
pressure (Similar to H&S)
Head & shoulder (H&S)
patterns must be preceded by
Selling pressure appears after Used to predict the resumption
uptrend & inverse H&S must be
resistance level of a market trend
preceded by downtrends
Double bottom & triple bottom
Analyst use size of H&S pattern
for downtrends
to project price target

Triangles Rectangles

Form when trading temporarily

Form when the range between range b/w support &
high and low prices narrows resistance level

Can be symmetrical, ascending Is a form of continuation

or descending pattern with one formed by
connecting the high prices and
Suggest buying & selling the other by connecting the
pressure roughly equal lows

Measuring implication: height Flags & pennants - Form over a

of triangle at formation short period of time, on a daily
price chart of triangle and
rectangle at formation
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS e Common technical analysis indicators
Price based indicators

Moving Bollinger
Oscillators
average lines bands

Mean of closing prices over a

Based on standard deviation
specified number of periods
of closing prices over last n
periods Tool to identify overbought
The longer the time-frame or oversold market
used to create moving
Analyst draw high & low
average (MA) “n”, smoother Based on market prices but
bands above & below n-
the avg. line scaled so that they ‘oscillate’
period MA
around a value or between
In uptrend - Price is higher two values.
Long-term investors may
than moving avg. & vice
buy (sell) when price
versa Charts used to identify
significantly exceeds (falls
below) the upper (lower) convergence or divergence
MA for different periods can of oscillator & market prices
bound
be used together
Contrarian strategy - Buy Convergence - Oscillator
Golden cross - Short-term shows same pattern as
(sell) when price reaches
M.A. crosses from prices
the upper (lower) band
underneath long-term MA,
indicates buy signal Divergence - Oscillator
Prices above Bollinger bonds
indicates overbought market shows different pattern than
Dead cross - Short-term MA prices
crosses from above long-
Prices below Bollinger bonds
term moving average,
indicates oversold market
indicates sell signal

Examples of oscillators

ROC
RSI MACD Stochastic
(Momentum)

Calculated from latest

MACD - Moving
ROC - Rate of change RSI - Relative closing price &
average convergence
Strength Index highest & lowest
/divergence
Calculated as - prices
100 × Diff. b/w last Based on ratio - Uses exponentially
closing price & closing Total price increase Use two lines
smoothed market
price ‘x’ days ago Total price decrease bounded 0 & 100
values

Buy when the Oscillate b/w 0 & 100 %K = Diff. b/w latest
Oscillate around 0 but
oscillator changes price & recent low as
not bounded
from −ve to +ve in Value > 70 = % of diff. b/w recent
uptrend (vice versa) Overbought market high & lows
MACD line crossing
above the smoother
Can be around 0 or Value < 30 = %D = Avg. of last
signal line - Buy (vice
around 100 Oversold market three %K values
versa)
calculated daily
foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
Non-price based indicators

Sentiment Flow of funds

indicators indicators

CBOE Volatility Margin Debt Short interest

Put/Call ratio
Index (VIX) (MD) ratio

Put volume Short interest is no.

Call volume of shares borrowed
Measure volatility of Ç in MD, Ç buying, & sold short
Ç in ratio = options on S&P 500 when reach their
Market sentiment is stock index limit, buying È, Short interest ratio
extremely negative. prices È, investor = Short interest /
likely increase in High VIX = fear sell securities to Avg. daily trading
price declines in stock meet margin calls volume
market
Viewed as Ç MD coincides with Some analysts may
contrarian indicator Technical analysts Ç prices & È MD believe that if ratio
interpret VIX in with È prices increases, market
Extremely high ratio contrarian way should express
= Bearish outlook decrease in price
(vice versa) and vice versa

Margin Debt Mutual fund New equity

TRIN
(MD) cash position issuance

Short term trading index

aka Arms index

Measure of funds flowing

into or out of advancing &
declining stocks
Ratio =
TRIN = Fund’s cash IPO add to supply of stocks
No. of advancing issues/No. Total assets
Ç in MD investor wants to Secondary issues do not
of declining issues
buy more stocks Uptrend = Ratio È increase the supply of stock
Volume of advancing
issues/Volume of declining but rather increase shares
È in MD investor wants to Downtrend = Ratio Ç available for trading
issues
buy more stocks
Technical analysts view Issuer tends to issue when
Index value close to 1 =
mutual fund cash as market peaks, so issuance
Flowing evenly to advancing
contrarian indicator coincide with high price
& declining stocks

Value > 1 = Majority in

declining stocks

Value < 1 = Majority in

advancing stocks

Sentiment indicators - Used to gain insight into trends

Flow of funds indicators - Useful for observing changes in demand & supply of securities

foresight school
[Link] © 2017 FinTree Education Pvt. Ltd.
LOS f Cycle Periods

Presidential Cycles Decennial Patterns Kondratieff wave

Broken down on the

Tied to US
basis of the last digit
Presidential 18-year cycles or
in the year; years
elections cycle with 54-year cycles
ending with a 0(5)
the third year being
have had the worst
prior to election year
(best) performance

LOS g Elliott wave theory

ª It is based on a belief that financial market prices can be described by an
interconnected set of cycles

ª Waves refer to chart patterns

ª Uptrend - 5 upward waves & 3 downward waves

ª Downtrend - 5 downward waves & 3 upward waves

ª Fibonacci ratios - Ratios of the size of the subsequent wave

ª Fibonacci numbers are found by starting with 0 and 1 and the each
subsequent number in the sequence is the sum of the two previous numbers.
Eg. 0, 1, 1, 2, 3, 5, 8 and so on

ª Ratio of 0.618 and 1.618 used to project price targets

LOS h Intermarket analysis

It refers to analysis of interrelationship among MV of asset classes (e.g. stocks, bonds)

Relative strength ratio - To identify outperforming asset class, then assets within class

foresight school