Concepts Description
Capital Market Expectation
Role of capital market expectation Capital market expectation include :
‐ Macro expectation ‐ expectations of asset classes : for top‐down approach
‐ Micro expectation ‐ expectations of individual asset : for bottom‐up approach
Capital market expectation Beta research : formulating capital market expectations ‐ related to systematic risk
framework Alpha research : calculate excess return within specific asset group
7 steps to formulate capital market expectation :
‐ Step 1 : Determine the specific capital market expectation needed based on investor's tax status, allowed asset classes, time horizon
‐ Step 2 : Determine performance driver based on historical data + Establish range of possible future performance
‐ Step 3 : Identify valuation model used + requirements
‐ Step 4 : Collect data. Considerations are :
+ Calculation method
+ Data collection techniques
+ Data definitions
+ Error rate
+ Investability + Correction for free float
+ Turnover in index components
+ Potential biases
‐ Step 5 : Assess current conditions + Assign value to required inputs, based on experience and judgment
‐ Step 6 : Formulate capital market expectation
‐ Step 7 : Monitor performance + refine
Problems when developing capital 1. Limitations to using economic data :
market forecasts ‐ Time lage between data distribution and collection
‐ Revisions of data are not made at the same time as publication
‐ Change of data definitions and methodology over time
‐ Change of the base upon which indices are calculated over time
2. Data measurement error + bias :
‐ Transcription errors : incorrect reporting/recording of data
‐ Survivorship bias : only record data of survivors
‐ Appraisal data of illiquid assets → smoothen the return
3. Limitations of historical estimates :
‐ Advantages of long‐period :
+ Statistically required : Number of data points must > number of covariances to calculate historical covariance
+ More precise estimates + smaller variances
+ Less sensitive to starting and ending point
‐ Disadvantages of long‐period :
+ Regime changes : change in underlying fundamentals → new subperiod with different characteris cs
+ Nonstationary : invalidate statistics before and after each subperiod
4. Use of ex post data (after the fact) to determine ex ante risk and return :
5. Non‐repeating data patterns :
‐ Data mining : data appears to have relationship by random chance
‐ Time period bias : relationship only appears in a certain time period
6. Failing for account for conditioning information : relationship between return and variables are not constant over time → account fot current condi ons in the forecast
7. Misintepretation of correlations
8 . Psychological traps :
‐ Anchoring trap : overweight 1st information received
‐ Status quo trap : predictions are highly influenced by recent past
‐ Confirming evidence trap : only consider information supporting existing belief
‐ Overconfidence trap : ignore past mistakes, overestimate accuracy of forecasts, and take the lack of comments from others as agreement
‐ Prudence trap : overly conservative forecasts → avoid making extreme forecasts
‐ Recallability trap : overweight easiest to remember events
9. Model and input uncertainty :
‐ Model uncertainty : do not know which model to use
‐ Input uncertainty : do not know the correct input values
Statistic tools 1. Descriptive statistics : to summarise data
2. Inferential statistics : use data to forecast
3. Stationary past data : parameter driving past and future are unchanged → could use historical data to es mate future data
4. Return can be calculated using arithmetic (for single period) or geometric (for multiple periods) average of past returns
5. Shrinkage estimate : weighted average based on history and target projection
6. Time series models : use past value of a variables to calculate its future value (e.g.: Volatility clustering)
7. Multifactor models : Forecast returns based on sensitivities (β) and risk factor (F)
‐ Advantages of multifactors :
+ Returns, covariances, variances : calculated from same set of risk factors
+ A set of well‐chosen, consistent factors → reduce chance for random varia on of es mates
+ Allow for tesing the consistency of covariance matrix
Calculate Covariance of 2 markets (A,B) based on Covariance of 2 Global insutrments (F1, F2) and the Sensitivities factors of each market with each Global instruments
𝐶𝑜𝑣 𝐴, 𝐵 𝛽 , 𝛽 , 𝜎 𝛽 , 𝛽 , 𝜎 𝛽 , 𝛽 , 𝛽 , 𝛽 , 𝐶𝑜𝑣 𝐹 , 𝐹
Calculate variance of each market
𝜎 𝛽 , 𝜎 𝛽, 𝜎 2 𝛽 , 𝛽 , 𝐶𝑜𝑣 𝐹 , 𝐹
, Discounted Cash flow models Discounted cash flow model : to calculate intrinsic value of asset based on PV of future CF
Advantages :
‐ Emphasis on future CF
‐ Ability to back out a required return
Disadvantages :
‐ Do not take into account current market conditions (supply/demand)
𝐷𝑖𝑣 𝐷𝑖𝑣
𝑃 →𝑟 𝑔
𝑟 𝑔 𝑃
If we take into account inflation , change in share outstanding and change in P/E ratio
𝐷𝑖𝑣 𝐷𝑖𝑣
𝑟 𝑔 ∆𝑆 ∆ 𝑃 ⁄𝐸 ∆𝑆 𝑔 𝑖 ∆ 𝑃 ⁄𝐸
𝑃 𝑃
In which :
𝐷𝑖𝑣
𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑖𝑛𝑐𝑜𝑚𝑒 𝑟𝑒𝑡𝑢𝑟𝑛 ∆𝑆
𝑃
𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑒𝑎𝑟𝑛𝑖𝑛𝑔 𝑔𝑟𝑜𝑤𝑡ℎ 𝑔 𝑖
𝑅𝑒𝑝𝑟𝑖𝑐𝑖𝑛𝑔 𝑟𝑒𝑡𝑢𝑟𝑛 ∆ 𝑃/𝐸
Build‐up model 𝑟 𝑟 𝐼𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛 𝑅𝑃 𝐷𝑒𝑓𝑎𝑢𝑙𝑡 𝑅𝑃 𝐼𝑙𝑙𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦 𝑅𝑃 𝑀𝑎𝑡𝑢𝑟𝑖𝑡𝑦 𝑅𝑃 𝑇𝑎𝑥 𝑝𝑟𝑒𝑚𝑖𝑢𝑚
Financial Equilibrium Models Financial Equilibrium model : asume balance supply and demand in global asset markets. One model is International Capital Asset Pricing Model (ICAPM)
𝑟 𝑟 𝛽 𝑟 𝑟
𝐶𝑜𝑣 𝑖, 𝑚 𝜌, 𝜎
𝛽 →𝑟 𝑟 𝑟 𝑟
𝜎 𝜌, 𝜎 𝜎
𝛽
𝐶𝑜𝑣 𝑖, 𝑚 𝜎
𝜌, → 𝐶𝑜𝑣 𝑖, 𝑚 𝜌, 𝜎 𝜎
𝜎 𝜎
Liquidity risk premium = Sharpe ratio ‐ market Sharpe ratio
Segmented market : capital does not flow freely across borders
Integrated market : capital flows freely across borders
To calculate risk premium of a not fully segmented/integrated market :
𝑅𝑃 𝑤 𝑅𝑃 𝑤 𝑅𝑃
𝑓𝑢𝑙𝑙𝑦 𝑠𝑒𝑔𝑚𝑒𝑛𝑡𝑒𝑑 𝑚𝑎𝑟𝑘𝑒𝑡 →𝜌 , 1
Covariance between 2 markets, given 1 factor driving returns
𝐶𝑜𝑣 𝑖, 𝑗 𝛽 𝛽 𝜎
Surveys / Judgment Survey : polls of expectations from market participants (economists, analysts) about economy and capital market
‐ Panel method : group polled constant over time
Judgment : adjust the expectations based on experience and insight to improve the forecast
Components and measures of 2 Components of economic growth :
economic activity 1. Cyclical growth : short‐term, including :
‐ Inventory cycle :
+ last 2 to 4 years
+ Measured using inventory / sales ratio
+ Cycle is as follows : Businesses gain confidence
‐ Decrease inventory/sales ratio ‐ Increase inventory/sales ratio
‐ Decrease employment Decrease inventory for Increase inventory for ‐ Increase employment
‐ Subsequent slow down economic ‐ Subsequent increase in economic
Adverse factors (e.g.: tightening
‐ Business cycle :
+ last 9 to 11 years
+ have 5 phases : (1) Initial recovery, (2) Early upswing, (3) Late upswing, (4) slowdown, (5) recession
2. Trend‐growth : long‐term
Measures of economic activity :
1. GDP : measured in real terms
2. Output gap = LT trend line GDP ‐ current GDP
3. Recession : decrease in GDP over 2 consecutive quarters
Capital Market Expectation
Role of capital market expectation Capital market expectation include :
‐ Macro expectation ‐ expectations of asset classes : for top‐down approach
‐ Micro expectation ‐ expectations of individual asset : for bottom‐up approach
Capital market expectation Beta research : formulating capital market expectations ‐ related to systematic risk
framework Alpha research : calculate excess return within specific asset group
7 steps to formulate capital market expectation :
‐ Step 1 : Determine the specific capital market expectation needed based on investor's tax status, allowed asset classes, time horizon
‐ Step 2 : Determine performance driver based on historical data + Establish range of possible future performance
‐ Step 3 : Identify valuation model used + requirements
‐ Step 4 : Collect data. Considerations are :
+ Calculation method
+ Data collection techniques
+ Data definitions
+ Error rate
+ Investability + Correction for free float
+ Turnover in index components
+ Potential biases
‐ Step 5 : Assess current conditions + Assign value to required inputs, based on experience and judgment
‐ Step 6 : Formulate capital market expectation
‐ Step 7 : Monitor performance + refine
Problems when developing capital 1. Limitations to using economic data :
market forecasts ‐ Time lage between data distribution and collection
‐ Revisions of data are not made at the same time as publication
‐ Change of data definitions and methodology over time
‐ Change of the base upon which indices are calculated over time
2. Data measurement error + bias :
‐ Transcription errors : incorrect reporting/recording of data
‐ Survivorship bias : only record data of survivors
‐ Appraisal data of illiquid assets → smoothen the return
3. Limitations of historical estimates :
‐ Advantages of long‐period :
+ Statistically required : Number of data points must > number of covariances to calculate historical covariance
+ More precise estimates + smaller variances
+ Less sensitive to starting and ending point
‐ Disadvantages of long‐period :
+ Regime changes : change in underlying fundamentals → new subperiod with different characteris cs
+ Nonstationary : invalidate statistics before and after each subperiod
4. Use of ex post data (after the fact) to determine ex ante risk and return :
5. Non‐repeating data patterns :
‐ Data mining : data appears to have relationship by random chance
‐ Time period bias : relationship only appears in a certain time period
6. Failing for account for conditioning information : relationship between return and variables are not constant over time → account fot current condi ons in the forecast
7. Misintepretation of correlations
8 . Psychological traps :
‐ Anchoring trap : overweight 1st information received
‐ Status quo trap : predictions are highly influenced by recent past
‐ Confirming evidence trap : only consider information supporting existing belief
‐ Overconfidence trap : ignore past mistakes, overestimate accuracy of forecasts, and take the lack of comments from others as agreement
‐ Prudence trap : overly conservative forecasts → avoid making extreme forecasts
‐ Recallability trap : overweight easiest to remember events
9. Model and input uncertainty :
‐ Model uncertainty : do not know which model to use
‐ Input uncertainty : do not know the correct input values
Statistic tools 1. Descriptive statistics : to summarise data
2. Inferential statistics : use data to forecast
3. Stationary past data : parameter driving past and future are unchanged → could use historical data to es mate future data
4. Return can be calculated using arithmetic (for single period) or geometric (for multiple periods) average of past returns
5. Shrinkage estimate : weighted average based on history and target projection
6. Time series models : use past value of a variables to calculate its future value (e.g.: Volatility clustering)
7. Multifactor models : Forecast returns based on sensitivities (β) and risk factor (F)
‐ Advantages of multifactors :
+ Returns, covariances, variances : calculated from same set of risk factors
+ A set of well‐chosen, consistent factors → reduce chance for random varia on of es mates
+ Allow for tesing the consistency of covariance matrix
Calculate Covariance of 2 markets (A,B) based on Covariance of 2 Global insutrments (F1, F2) and the Sensitivities factors of each market with each Global instruments
𝐶𝑜𝑣 𝐴, 𝐵 𝛽 , 𝛽 , 𝜎 𝛽 , 𝛽 , 𝜎 𝛽 , 𝛽 , 𝛽 , 𝛽 , 𝐶𝑜𝑣 𝐹 , 𝐹
Calculate variance of each market
𝜎 𝛽 , 𝜎 𝛽, 𝜎 2 𝛽 , 𝛽 , 𝐶𝑜𝑣 𝐹 , 𝐹
, Discounted Cash flow models Discounted cash flow model : to calculate intrinsic value of asset based on PV of future CF
Advantages :
‐ Emphasis on future CF
‐ Ability to back out a required return
Disadvantages :
‐ Do not take into account current market conditions (supply/demand)
𝐷𝑖𝑣 𝐷𝑖𝑣
𝑃 →𝑟 𝑔
𝑟 𝑔 𝑃
If we take into account inflation , change in share outstanding and change in P/E ratio
𝐷𝑖𝑣 𝐷𝑖𝑣
𝑟 𝑔 ∆𝑆 ∆ 𝑃 ⁄𝐸 ∆𝑆 𝑔 𝑖 ∆ 𝑃 ⁄𝐸
𝑃 𝑃
In which :
𝐷𝑖𝑣
𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑖𝑛𝑐𝑜𝑚𝑒 𝑟𝑒𝑡𝑢𝑟𝑛 ∆𝑆
𝑃
𝐸𝑥𝑝𝑒𝑐𝑡𝑒𝑑 𝑒𝑎𝑟𝑛𝑖𝑛𝑔 𝑔𝑟𝑜𝑤𝑡ℎ 𝑔 𝑖
𝑅𝑒𝑝𝑟𝑖𝑐𝑖𝑛𝑔 𝑟𝑒𝑡𝑢𝑟𝑛 ∆ 𝑃/𝐸
Build‐up model 𝑟 𝑟 𝐼𝑛𝑓𝑙𝑎𝑡𝑖𝑜𝑛 𝑅𝑃 𝐷𝑒𝑓𝑎𝑢𝑙𝑡 𝑅𝑃 𝐼𝑙𝑙𝑖𝑞𝑢𝑖𝑑𝑖𝑡𝑦 𝑅𝑃 𝑀𝑎𝑡𝑢𝑟𝑖𝑡𝑦 𝑅𝑃 𝑇𝑎𝑥 𝑝𝑟𝑒𝑚𝑖𝑢𝑚
Financial Equilibrium Models Financial Equilibrium model : asume balance supply and demand in global asset markets. One model is International Capital Asset Pricing Model (ICAPM)
𝑟 𝑟 𝛽 𝑟 𝑟
𝐶𝑜𝑣 𝑖, 𝑚 𝜌, 𝜎
𝛽 →𝑟 𝑟 𝑟 𝑟
𝜎 𝜌, 𝜎 𝜎
𝛽
𝐶𝑜𝑣 𝑖, 𝑚 𝜎
𝜌, → 𝐶𝑜𝑣 𝑖, 𝑚 𝜌, 𝜎 𝜎
𝜎 𝜎
Liquidity risk premium = Sharpe ratio ‐ market Sharpe ratio
Segmented market : capital does not flow freely across borders
Integrated market : capital flows freely across borders
To calculate risk premium of a not fully segmented/integrated market :
𝑅𝑃 𝑤 𝑅𝑃 𝑤 𝑅𝑃
𝑓𝑢𝑙𝑙𝑦 𝑠𝑒𝑔𝑚𝑒𝑛𝑡𝑒𝑑 𝑚𝑎𝑟𝑘𝑒𝑡 →𝜌 , 1
Covariance between 2 markets, given 1 factor driving returns
𝐶𝑜𝑣 𝑖, 𝑗 𝛽 𝛽 𝜎
Surveys / Judgment Survey : polls of expectations from market participants (economists, analysts) about economy and capital market
‐ Panel method : group polled constant over time
Judgment : adjust the expectations based on experience and insight to improve the forecast
Components and measures of 2 Components of economic growth :
economic activity 1. Cyclical growth : short‐term, including :
‐ Inventory cycle :
+ last 2 to 4 years
+ Measured using inventory / sales ratio
+ Cycle is as follows : Businesses gain confidence
‐ Decrease inventory/sales ratio ‐ Increase inventory/sales ratio
‐ Decrease employment Decrease inventory for Increase inventory for ‐ Increase employment
‐ Subsequent slow down economic ‐ Subsequent increase in economic
Adverse factors (e.g.: tightening
‐ Business cycle :
+ last 9 to 11 years
+ have 5 phases : (1) Initial recovery, (2) Early upswing, (3) Late upswing, (4) slowdown, (5) recession
2. Trend‐growth : long‐term
Measures of economic activity :
1. GDP : measured in real terms
2. Output gap = LT trend line GDP ‐ current GDP
3. Recession : decrease in GDP over 2 consecutive quarters
