STOCK RISK

Dr Aneesh MR
Introduction
• A stock’s risk reflects the uncertainty about future return.
• The actual return may be less than expected.
• The return from investing in stock over a particular period is measured
as
Measures of Risk
The risk of a stock can be measured by using
• Its price volatility,
• Its beta, and
• The value-at-risk method.
Volatility of a Stock
• A stock’s volatility serves as a measure of risk because it may indicate
the degree of uncertainty surrounding the stock’s future returns.
• The volatility is often referred to as total risk because it reflects
movements in stock prices for any reason, not just movements
attributable to stock market movements.
Standard Deviation to Forecast Stock Price
Volatility
• One way to forecast stock price volatility is by using a historical period
to derive a stock’s standard deviation of returns, and then using that
estimate as the forecast over the future.
• Although a stock price’s level of volatility may change over time, this
method can be useful when there is no obvious trend in volatility.
Volatility Patterns to Forecast Stock Price
Volatility
• A second method for forecasting stock price volatility is to use
volatility patterns in previous periods.
• Various economic and political factors can cause stock price volatility
to change abruptly, so even a sophisticated time-series model does not
necessarily generate accurate forecasts of stock price volatility.
Example
• The standard deviation of daily stock returns is determined for each of the last
three months.
• Then, a time-series trend of these standard deviation levels is used to form an
estimate for the standard deviation of daily stock returns over the next month.
• This method uses information beyond that contained in the previous month.
• For example, the forecast for September might be based on the following
weighting scheme:
• 50 percent of the standard deviation in the most recent month (August) plus,
• 30 percent of the standard deviation in the month before that (July) plus,
• 20 percent of the standard deviation in the month before that (June).
Implied Volatility to Forecast Stock Price
Volatility
• A third method for forecasting stock price volatility is to derive the
stock’s implied standard deviation (ISD) from the stock option pricing
model.
Implied Volatility
• A metric derived from option prices
• Indication of perceived future volatility of an underlying
• Expressed annually as 1SD %
Stock Option
• A stock option gives an investor the right—but not the obligation—to
buy or sell a stock at an agreed-upon price and date.
• There are two types of options:
• Put option, which is a bet that a stock will fall, and
• Call option, which is a bet that a stock will rise.
• It has shares of stock (or a stock index) as its underlying asset, stock
options are a form of equity derivative and may be called equity
options.
Forecasting Stock Price Volatility of the Stock
Market
• To forecast volatility of the stock market in general can monitor the
volatility index (VIX) derived from stock options on the S&P 500
stock index.
• At a given point in time, this index measures investors’ expectation of
the stock market volatility over the next 30 days.
• Some investors refer to VIX as an indicator of stock market fear.
• If investors expect more uncertainty surrounding stock prices over the
next 30 days, this means that investors would require a higher
premium in order to sell call options.
• As the option premium on options representing the S&P 500 index
increase, the VIX index will increase.
Forecasting Stock Price Volatility of the
Stock Market
• If conditions in the stock market are expected to be more stable over
the next 30 days, the premium on call options would decline and the
VIX index would decline.
• It would be impossible to survey all market participants to obtain their
view of stock market volatility, the movements in the premium of
options on the stock market index are commonly used to derive an
estimate of expected stock market volatility.
Volatility of a Stock Portfolio
• Participants in the stock market tend to invest in a portfolio of stocks
rather than a single stock
• They are more concerned with the risk of a portfolio than with the risk
of an individual stock.
• A portfolio’s volatility depends on the volatility of the individual
stocks in the portfolio, on the correlations between returns of the
stocks in the portfolio, and on the proportion of total funds invested in
each stock.
Volatility of a Stock Portfolio
• The portfolio’s volatility can be measured by the standard deviation:

• The portfolio standard deviation (or risk), denoted as σp

Volatility of a Stock Portfolio
• wi​, wj These represent the weights (proportion of total investment) of
asset i and asset j in the portfolio.
• σi​, σj​: The standard deviations (individual risks) of asset i and asset j,
respectively.
• CORRij​: The correlation coefficient between the returns of asset i and
asset j. Correlation values range from -1 to 1, where:
• 1 means the assets move perfectly in sync.
• −1 means they move perfectly in opposite directions.
• 0 means no correlation.
Volatility of a Stock Portfolio
• The first two terms represent the risk contribution of each individual
asset i and j, taking into account their respective weights and standard
deviations.
• The third term accounts for the combined risk from the correlation
between the assets. The double summation sums over all the pairs of
assets i and j in the portfolio.

• This formula essentially considers both the individual risks of each

asset and how the assets interact with each other through their
correlations to give the total portfolio risk.
Beta of a Stock portfolio
• A stock’s beta measures the sensitivity of its returns to market
returns.

• the portfolio beta is a weighted average of the betas of stocks that

make up the portfolio, where the weights reflect the proportion of
funds invested in each stock.
Beta of a stock
• The beta of a stock and its volatility are typically related. High-beta
stocks are expected to be relatively volatile because they are more
sensitive to market returns over time.
• Likewise, low-beta stocks are expected to be less volatile because they
are less responsive to market returns.
Value at Risk
• Value at risk is a measurement that estimates the largest expected loss
to a particular investment position for a specified confidence level.
• It is intended to warn investors about the potential maximum loss that
could occur.
• If the investors are uncomfortable with the potential loss that could
occur in a day or a week, they can revise their investment portfolio to
make it less risky.
Value at Risk (VaR)
• Value at Risk (VaR) is a financial risk management tool used
to estimate the potential loss in value of an asset, portfolio, or
investment over a specific time period, given normal market
conditions, and at a certain confidence level.
• VaR provides a threshold for the maximum expected loss
with a predefined probability, helping institutions understand
and manage the risks of their investment portfolios.
For example:
• If a portfolio has a 1-day VaR of ₹1 million at a 95% confidence level, there is a
95% chance that the portfolio will not lose more than ₹1 million in one day.
Value at Risk (VaR)
VaR is typically calculated for three components:
1. Time Horizon (e.g., one day, one week).
2. Confidence Level (e.g., 95%, 99%).
3. Loss Amount (in monetary terms).
• VaR is commonly used by banks, asset managers, and regulators to
assess risk exposure, though it has limitations, such as not accounting
for extreme events or "tail risks."
Methods of Calculation
1. Historical Method
• The historical method calculates VaR by using historical data of asset or portfolio
returns. It assumes that past performance is indicative of future risks.
Steps:
• Collect historical returns data of the asset or portfolio over a given time horizon.
• Rank the returns from worst to best.
• Choose the confidence level (e.g., 95% or 99%).
• Identify the return corresponding to the confidence level. For a 95% confidence
level, the VaR is the return at the 5th percentile of the worst returns.
• For example, if you have 100 historical daily returns and want to calculate the
95% VaR, you would take the 5th-worst return. Multiply this return by the
portfolio value to get the monetary loss.
Advantages:
• No assumptions about distribution: The historical method does not
assume that returns follow a normal distribution. It relies on actual
past data, making it useful for assets with non-normal distributions.
• Simplicity: It’s easy to implement, as it directly uses historical data
without requiring complex models.
• Captures real market events: It accounts for real past events,
including extreme events or outliers, which are important for
understanding tail risks.
Limitations:
• Past is not always a good predictor of the future: This method
assumes that future market behaviour will resemble the past, which
may not always be true, especially in volatile or changing markets.
• Limited by historical data: If historical data is limited, the method
may fail to capture all potential risks. If the market is going through
structural changes, relying on past data may be misleading.
• Ignores forward-looking factors: It does not account for current
market dynamics, such as shifts in volatility or other predictive factors.
Variance-Covariance (Parametric) Method
• This method assumes that returns follow a normal distribution, making it
relatively simple to calculate VaR based on the mean and standard deviation
of returns.
• Steps:
• Calculate the mean (expected) return and the standard deviation of the asset
or portfolio’s returns.
• Choose the confidence level (e.g., 95% or 99%).
• Use the z-score associated with the confidence level (for 95%, the z-score is
1.65; for 99%, it's 2.33).
• Formula: VaR=μ+Z×σ
• Where: μ is the mean return, is the z-score corresponding to the confidence
level, σ is the standard deviation of returns.
Example
• For a portfolio worth $1,000,000 with a mean return of 0 and standard
deviation of 2%, and a confidence level of 95% (z-score = 1.65),
• VaR=0+(1.65×2%)×1,000,000=$33,000

• This means that with a 95% confidence level, the portfolio could lose
up to $33,000 in a given period.
Advantages:
• Simplicity and speed: Since it only requires calculating the mean and
standard deviation, this method is computationally efficient, especially
for large portfolios.
• Useful for normally distributed returns: For assets or portfolios
with returns that are close to normal distribution, this method provides
reasonable estimates.
• Easily scalable: It's easy to implement for large portfolios by
aggregating individual asset volatilities and correlations.
Limitations:
• Assumption of normality: it assumes returns follow a normal
distribution, which is often not the case in financial markets, especially
during extreme events (e.g., market crashes, fat tails).
• Underestimates tail risks: By assuming normal distribution, the
parametric method may underestimate extreme events and risks that
occur beyond the normal range of outcomes.
• Ignores skewness and kurtosis: The method does not account for
skewed returns (where losses are more frequent than gains, or vice
versa) or "fat tails" (where extreme losses are more common than
predicted by a normal distribution).
Monte Carlo Simulation
• This method simulates a large number of potential future price movements
based on assumed probability distributions.
Steps:
• Simulate multiple scenarios of future asset or portfolio returns using random
sampling.
• Calculate the portfolio value for each scenario.
• Rank the outcomes and determine the loss at the desired confidence level
(similar to the historical method).
• This approach is more flexible, as it can incorporate non-normal
distributions, volatility clustering, or other complex factors affecting asset
prices.
Example:
• If you simulate 10,000 possible returns for a portfolio and 5% of those
simulations result in losses greater than $50,000, then the 95% VaR is
$50,000.
Advantages:
• Flexibility: Monte Carlo can model complex portfolios and
accommodate any distribution of returns, including those with
skewness, kurtosis, or non-normal features.
• Customizable for different assumptions: It allows for the
incorporation of forward-looking factors such as changing volatility,
time-varying correlations, or specific market scenarios.
• Captures extreme events: By simulating numerous potential
outcomes, it can provide better insight into tail risks and extreme
market movements.
Limitations:
• Computationally intensive: Monte Carlo simulations require a lot of
processing power, especially for large portfolios or long time horizons,
making it slower and more expensive to run compared to simpler
methods.
• Complexity in modelling: Designing the simulation requires making
assumptions about the distribution of returns, correlations, and
volatility, which can be difficult to estimate accurately.
• Garbage in, garbage out: The accuracy of the results depends
heavily on the quality of the input assumptions. Poor assumptions lead
to unreliable VaR estimates.
Summary
Tail risk
• Tail risk refers to the risk of extreme price movements or events
occurring that lie in the far ends (or "tails") of a probability
distribution.
• These are not captured well by standard risk models assuming normal
distribution.
• These extreme events have a low probability of occurring but can lead
to very large losses.
Tail Risk
• In a normal distribution (bell curve), most outcomes cluster around the
mean, and the probability of extreme deviations (events in the "tails")
is very small.
• However, in reality, financial markets often experience "fat tails,"
meaning extreme events (large gains or losses) occur more frequently
than predicted by normal distribution models.
• Tail risk focuses on these unexpected, large negative outcomes.
Examples
• Imagine you are managing a portfolio of stocks and bonds,
and your risk models are based on normal distribution
assumptions. The models predict that under "normal"
conditions, your maximum loss in a day might be 2%, which
would happen once every hundred days.
• Black Monday (1987): On October 19, 1987, global stock
markets crashed, and the U.S. Dow Jones Industrial Average
fell by more than 22% in a single day. This kind of market
crash was an extreme tail event — a rare, catastrophic event
that normal risk models would have deemed almost
impossible.
Examples
• The Sensex surged 17. 34 % in May 18, 2009 after the UPA
government led by the Congress party won the general elections,
creating political stability and boosting investor sentiment.