Financial Modelling

Dr. Aishwarya Krishna

 Zoom Video Communications prices plot
Adj Close
600 120.00%

500 100.00%

400 80.00%
Axis Title

300 60.00%

200 40.00%

100 20.00%

0 0.00%

Axis Title
 Zoom Video Communications return plot

1.2 50.00%

40.00%
1
30.00%

0.8 20.00%
Axis Title

10.00%
0.6
0.00%

0.4 -10.00%

-20.00%
0.2
-30.00%

0 -40.00%

Axis Title
Course outline
 Linear and Multivariate
Regression
Panel regression Stationarity
Dynamic dependence Models AR
Autocorrelation MA
Theories ARMA
Time series analysis Modeling
𝑟𝑡 Forecasting ARIMA

Modelling volatility 𝜎𝑡2 ARCH

GARCH

Non-synchronous trading
High Frequency Data and Algorithmic Trading Bid Ask Spread
Empirical characteristics of data
Modelling

Investments Portfolio optimization

Simple Monte Carlo
Monte Carlo Methods for Option Pricing Antithetic
 1 3-Jan-22 Estimation theory Class Notes
Assignment 1
5-Jan-22
Session Plan 2 Linear and Multivariate regression
Class Notes

10-Jan-22 Class Notes

3 Panel Data regression
4 12-Jan-22 Quiz 1

5 17-Jan-22
Tsay Ch 2
6 19-Jan-22 Time Series Analysis
7 24-Jan-22
Conditional Heteroscedastic
8 31-Jan-22 Models

9 2-Feb-22 Tsay Ch 3
High Frequency Data and Tsay Chapter 5
14-Feb-22
10 Algorithmic Trading Mid Term Exam
11 16-Feb-22

12 21-Feb-22 Portfolio Models and BA Ch 8,11,13

13 23-Feb-22 Optimization Quiz 2
BA Ch 16 & 19
28-Feb-22
14 Modelling Option Prices
15 2-Mar-22 Monte Carlo Methods for Option BA Ch 23
16 7-Mar-22 Pricing Assignment 2
17 9-Mar-22 Project Presentation
18 14-Mar-22 Project Presentation
19 16-Mar-22 Project Presentation
End Term Exam
 Specific Assessment Weightage
Method
End term Exam 35%
Evaluation Quizzes/ Tests 25%
Project/ presentations 20%

Assignments 20%
 Qualitative data or
Nominal /ordinal categorical variable

and Numeric data Quantitative data or

numerical variable

Same point in time

Types of Data or varies across
Cross-sectional data
Time-series data
time

Based on the source Primary data

of data Secondary data

1/16/2022 9
 Cou
Database Available data ntry Details
Company Data,
CMIE Prowess dx India
M&A etc. 34,000 Indian companies
Databases CMIE Prowess IQ
Company Data,
M&A, etc.
India
37780 Indian companies
Capitaline Plus Company Data India 35,000+ Indian listed and unlisted companies
Live Capital Worl
Bloomberg Market Data d Data on 52,000 companies worldwide
ACE Knowledge & Research Company 30,000 companies listed & unlisted in Indian
Portal Information India markets
It has information on over 1.9 million
Company &
companies, 2.8 million signatories (directors),
Industry
CMIE First Source India ,0.5 million annual reports, etc.
current trends and the long-term performance
India
CRISIL Database Industry Analysis outlook on 72 industries in India.
The CRSP US Index History Files – Intraday
Stock Data (NYSE, include second-by second levels of CRSP’s
CRSP AMEX, Nasdaq) US investible Equity Indexes and End of Day level
Company, Worl 37,000 active quoted companies and over 24,000
Datastream (WorldScope) Industry, Equity d inactive ones across 79 markets.
Provides annual compensation data on top five
Company & Worl executive officers within North American
Execucomp Industry data d company from 1992 to 2009.
 Database Available data Country Details

Databases
ISS provides data on Historical Governance, Historical Directors data,
Institutional Shareholder Company and Financial Voting Results data, Incentive Lab data, Voting Analytics with N-PX
Services Data US data, and Shareholder Proposal data
This database offers a comprehensive and unique collection of
company, industry and country analysis and report, extending across
MarketLine Advantage Company-Industry Data World 200+ countries and every major industry sector
NASSCOM Member
Directory Company Data India Indian IT and BPO industries with the short profile.
Company and Financial financial and non-financial information based on user-specified search
SeekEdgar Data US strings

UCLA-LoPucki Bankruptcy Company & Industry The BRD contains data on all of the more than 1000 large public
Research Database financial data US companies that have filed bankruptcy cases since October 1, 1979.
Company and Financial Provides fundamental and market data for 24,328 companies (11,640
Compustat Data US Active and 12,688 Inactive).

Info on over 88,000 companies (58,000+ active and 30,000+ inactive)

Company and Financial traded in more than 164 exchanges in over 120 countries,
Thomson Reuters Eikon Data World representing more than 99% of the world’s market cap.
NSE Data CM & FAO Trading data India Historical trade data of Indian listed firms
Data on 1,083,315 firms through the various forms submitted to the
MCA Company financials India Ministry
FIG Market Intelligence Company financials World This platform covers 50,000+ listed, non-listed

It provides over 15,000 business, legal and news, access thousands of

newspapers, periodicals, television and radio broadcasts, newswires,
LexisNexis News World blogs, corporate directories and financial information.
An aggregator database delivers information on Asian emerging
EMIS Intelligence Company financials Asia markets
Estimation
 Sample number Sample mean (𝑥ҧ )
Motivation

• Consider that the NSE has about 1 51814

2400 companies listed.
• A random sample of size 30 from 2 52670
a population of 2400 is used.
Suppose the full population data
is not available. . .
• Suppose we repeat the process
of selecting a simple random . .
sample of 30 companies over
and over again, each time . .
computing the values of 𝑥ҧ .
500 51752
Introduction to sampling distribution of 𝑥ҧ
Mean market cap Frequency Relative frequency
 Sampling
Distribution
of 𝑥ҧ
(Histogram)
Estimation
Point estimator

• Desirable properties of an estimator

• Unbiasedness: 𝐸 𝑥ҧ = 𝜇, the sample mean is an unbiased estimator of
population mean.
• Efficiency: The variance its sampling distribution is smaller than all the
other unbiased estimators of the parameter you are trying to estimate
• Consistency: Accuracy of the parameter increases as the sample size
increases.
• Linear: A point estimate can be used as a function of the sample data
If the estimator is the best(least variance), exhibits linearity and is unbiased,
it is said to be best linear unbiased estimator( BLUE).
Practical value of the
sampling distribution of 𝑥ҧ
What is the probability that
the sample mean computed
using a simple random sample
of 30 companies will be within
500 of the population mean
μ=51,800 𝑎𝑛𝑑 𝜎𝑝=4000?
What will
happen if
n=100?
Form of the
sampling
distribution
 • Statement: Distribution of means of random
samples taken from a population having mean
𝜇 and finite variance 𝜎 2 approaches the
normal distribution with mean 𝜇 and variance
𝜎2
as n goes to infinity.
Central limit 𝑛
• Significance: It permits us to use sample
theorem statistics to make inferences about population
parameters without knowing anything about
the shape of the frequency distribution of that
population other than what we can get from
the sample.
Judging
criteria at
various
significance
levels
2. S.E: Reliability and precision of a sample
The standard error gives an idea about the reliability and precision of a sample.
The smaller the S.E., the greater the uniformity of sampling distribution and
hence, greater is the reliability of sample. Conversely, the greater the S.E., the
greater the difference between observed and expected frequencies.

3. Confidence interval
 • Why do we use the t-distribution?
T-Values • Critical value for a certain level of
significance is far more to the right
in the case of t distribution. This
makes us to be more cautious. The
smaller the number of observations,
smaller the degrees of freedom, the
larger the critical values would be.
This requires more evidence against
the null hypothesis before we reject
it.

• The t-distribution is a bell-shaped

with fatter tails.

• It is symmetric about zero similar

to Normal distribution.
• 𝑡𝑇−2 − 𝑁 0,1 𝑎𝑠 𝑇 − 2 −→ ∞