Portfolio Management Flashcards
portfolio perspective (Markowitz framework)
evaluating how individual investments relate to the wider portfolio
Three steps to portfolio management process
- Planning step
- Execution step
- Feedback step
diversification ratio
s.d. portfolio returns / average s.d of returns of the individual securities in the portfolio
pretax nominal return
returns before tax
after-tax nominal return
return after tax
leveraged return
returns that are a multiple of the return of the
efficient frontier
portfolios with the greatest level of return for each level of risk
The efficient frontier outlines the set of portfolios that gives investors the highest return for a given level of risk or the lowest risk for a given level of return. Therefore, if a portfolio is not on the efficient frontier, there must be a portfolio that has lower risk for the same return. Equivalently, there must be a portfolio that produces a higher return for the same risk.
global minimum-variance portfolio
the best return profile with minimal level of risk
utility function
investor preference with regards to risk/return
The capital allocation line
a straight line from the risk-free asset through the optimal risky portfolio.
According to Markowitz, an investor’s optimal portfolio is determined where the
investor’s highest utility curve is tangent to the efficient frontier.
utility/indifference curve
a curve across all points in which the investor is indifferent (happy to invest across) - spread across different risk and returns
The capital market line (CML)
plots return against total risk, which is measured by standard deviation of returns.
A portfolio to the right of the market portfolio on the CML is:
an inefficient portfolio.
A portfolio to the right of a portfolio on the CML has more risk than the market portfolio. Investors seeking to take on more risk will borrow at the risk-free rate to purchase more of the market portfolio
Assumptions of CAPM:
- mean-variance framework
- unlimited lending/borrowing at Rf
- homogenous expectations
- one-period time horizon
- divisible assets
- frictionless markets
- no inflation and interest rate changes
- capital markets equilibrium
- investors are price takers
Cognitive dissonance
where an individual has conflicting beliefs e.g. a new piece of evidence challenges their assumption
Conservatism bias
not changing your opinion of something when new information is released
Confirmation bias
ignoring information that disagrees with your established views
Representative bias
assuming that all members of a population/sample share the same characteristics
e.g. base-rate neglect, sample-size neglect
Illusion of control bias
thinking you can control something but you cannot
e.g. illusion of knowledge, self-attribution, overconfidence,
Hindsight bias (‘i knew is all along phenomenon’)
being selective in your memory of past events, resulting in a tendency to see events being more predictable than they really are
Anchoring/adjustment bias
a cognitive bias that causes us to rely too heavily on the first piece of information we are given about a topic.
may lead to overtrading, underestimation of risk, and lack of diversification
Mental accounting bias
viewing money in different accounts or from different sources differently when making investment decisions e.g. treating a bonus differently to your regular income
Framing bias
Occurs when decisions are affected by the way in which the question is framed
Availability bias
putting undue emphasis on on information that is readily available/easy to recall e.g. picking a manager you know
loss-aversion bias
feeling more pain from losses than joy with equal gains
overconfidence bias
overestimating own abilities to make decisions - can also lead to illusion of knowledge bias and self-attribution bias
self-control bias
occurs when individuals lack self-discipline and favour short-term sacrifices to meet long-term goals.
status quo bias
occurs when comfort with an existing situation causes a resistance to change
Endowment bias
occurs when an asset is felt to be special and more valuable because it is already owned
Regret-aversion bias
occurs when investors do not take action, due to fear of being wrong
e.g. herding behaviour
Delta
sensitivity of derivative values to the price of underlying asset
Gamma
sensitivity of delta to the price of the underlying asset
Vega
sensitivity of derivative values to the volatility of the price of the underlying asset
Rho
sensitivity of derivative values to changes in the risk-free rate
Tail risk
the uncertainty about the probability of extreme (negative outcomes) e.g. downside risk, VaR
Value-at-Risk (VaR)
minimum loss over period with a specific probability
e.g. 1-month VAR of $1m with 5% probability = an expected loss of at least $1m in 5% of months
Conditional VaR
expected value of loss, given that loss exceeds a specific amount.
It is calculated as the probability-weighted average loss for all losses over a certain amount
self-insurance
where a company bear the losses of a particular risk factor
risk transfer
when another party takes on a specific risk
surety bond
where an insurance company agrees to make a payment if a third-party fails to perform its duty to an organisation.
fidelity bonds
where an insurance company agrees to make a payment in the event of employee theft/misconduct.
risk shifting
distributing risks via the use of derivative contracts
Risk management process / framework
The risk management process should identify an organization’s risk tolerance, identify the risks it faces, and monitor or address these risks. The goal is not to minimize or eliminate risks.
This includes the procedures, analytical tools, and infrastructure to conduct the risk governance process
Risk governance should most appropriately be addressed within an organization at:
the enterprise level.
Risk governance should be approached from an enterprise view, with senior management determining risk tolerance and a risk management strategy on an organization-wide level
Risk budgeting
Selecting assets or securities by their risk characteristics up to the maximum allowable amount of risk. The maximum amount of risk to be taken is established through risk governance.
Sources of financial risk
- market risk
- credit risk
- liquidity risk
Technical analysis
Driven only by share price and volume of trading data to project price.
3 key principles:
1. Market prices reflect all known info
2. Market prices exhibit trends and countertrends that persist.
3. Patterns and and cycles repeat themselves in predictable ways.
A market that is ‘uptrend’ in prices
demand is increasing relative to supply (prices consistently rising) (+1 gradient)
A market that is ‘downtrend’ in prices
supply is increasing relative to demand (prices consistently declining) (-1 gradient)
A market that is in ‘consolidation’
there is neither an uptrend or downtrend apparent.
support level
Price where buying pressure limits a downtrend (the lowest/bottom price of a stock)
resistance level
price where selling pressure limits and uptrend (the highest/top price of a stock)`
‘change in polarity’
breached resistance levels –> support levels and breached support levels –> resistance levels
Technical indicators
- Price-based indicators e.g. moving averages, Bollinger bands
- Momentum oscillators e.g. ROC, RSI, MACD
- Sentiment (non-price) indicators e.g. put/call, VIX, margin debt
momentum oscillator
indicators based on market prices but scaled so they ‘oscillate around a given value.
Convergence
When the oscillator shows the same pattern as prices
Divergence
When the oscillator shows a different pattern as prices
Rate of Change (ROC) oscillator
100x the difference between the latest closing price and the closing price n periods earlier. Oscillates around 0
Relative Strength Index (RSI)
based on the ratio of total price increases to total price decreases over n number of periods. The ratio is then scaled to oscillate between 0-100, with high values indicating an overbought market and visa versa.
Moving Average Convergence/Divergence (MACD)
MACD oscillators are drawn using smoothed moving averages - placing greater weight on recent observations. The MACD line is the difference between two exponentially smoothed moving averages of the price.
- Used to identify convergence/divergence with the price trend.
Stochastic oscillator
Calculated from the latest closing price and highest/lowest prices reached in a recent period.
- the “%K” line is the difference between the latest price and the recent low as a percentage of the difference between the recent high and low. The “%D” line is a 3-period average of the %K line.
put / call ratio
put volume / call volume
p/c ↑ negative outlook for price of asset
Volatility Index (VIX) (calculated by the Chicago Board Options Exchange)
- measures the volatility of options on the S&P 500 stock index.
VIX ↑ investors fear a decline in the stock market
Margin debt
total margin debt ↑, aggressive buying by bullish margin investors.
intermarket analysis
an analysis of the relationships between market values of major asset classes, such as stocks, bonds, commodities and currencies.
Relative strength charts
used to determine which asset classes are outperforming
the ‘Buy signal’ - when using moving average
‘golden cross’: a shorter-term average above a longer-term average
the ‘Sell signal’ - when using moving average
‘dead cross’: a shorter-term average below a longer-term average
‘Big Data’
- all potentially useful data (traditional + alternative data).
- volume, velocity and variety
Data science and processing methods
How we extract information
- capture
- curation
- storage
- search
- transfer
supervised learning
a machine learning technique in which a machine is given labelled input and output data and then models the output data based on the input data
unsupervised learning
a machine is given input data in which to identify patterns and relationships, but no output data to model
Deep learning
a technique to identify patterns of increasing complexity, and may use supervised or unsupervised learning.
Overfitting
A model that is overfit (too complex) will tend to identify spurious relationships in the data. Labelling of input data is related to the use of supervised or unsupervised machine learning techniques.
Underfitting
Underfitting describes a machine learning model that is not complex enough to describe the data it is meant to analyse. An underfit model treats true parameters as noise and fails to identify the actual patterns and relationships.
Tokenization
maintaining ownership records for physical assets on a distributed ledger.
