Class Notes (836,127)
Canada (509,645)
MIS 4500 (31)

Chapter 23.docx

6 Pages
Unlock Document

Management Info. Systems
MIS 4500
Pourang Irani

Chapter 23: Capital Market Expectations Frameworks for Developing Capital Market Expectations: 1. Specify the final set of expectations that are needed, including time horizon to which they apply 2. Research historical record i. collect macroeconomic and market info on 1. geographic area; or 2. broad asset class 3. Specify the method(s) and/or model(s) that will be used and their information 4. Determine the best sources for information needs 5. Interpret the current investment environment using the selected data and methods, applying experience and judgment 6. Provide the set of expectations that are needed, documenting conclusions 7. Monitor actual outcomes and compare them to expectations, providing feedback to improve the expectations-setting process Beta research: related to systematic risk and returns to systematic risk; development of capital market expectations Alpha research: related to capturing excess risk-adjusted returns by a particular strategy Good forecasts are:  unbiased, objective and well researched  efficient (reducing magnitude of forecast errors to a minimum)  internally consistent Challenges in Forecasting:  Limitations of economic data o time lag of collection, processing and dissemination o changes in definitions and calculation methods (say CPI-U)  re-basing indices  Data measurement errors and biases o Transcription errors: errors in gathering and recording o Survivorship bias: data series reflect only survivors o Appraisal (smoothed) data (say real estate or alternative investments): results in 1. calculated correlations w/ other assets tend to be smaller in absolute value than true correlations; 2. true standard deviation of asset is biased downward  Limitations of historical estimates: analysis should include discussion of what may be different from past o changes in technological, political, legal, and regulatory environments; disruptions such as wars o change of regime: change of governing set of relationships creates nonstationarity (different parts of data series reflect different underlying statistical properties) o long data series:  risk that data cover multiple regimes  time series of required length may not be available  in order to get data series of required length, temptation is to use high- frequency data (weekly or daily): more sensitive to asynchronism (discrepancy in dating of observations that occurs b/c stale (out-of-date) data may be used in absence of current date) across variables, producing lower correlation estimates  Ex Post Risk Can be a Biased Measure of Ex Ante Risk o ex post returns may reflect that didn’t materialize resulting in overstated estimates of ex ante returns  Biases in Analysts’ Methods: o Data-mining bias: repeatedly ―drilling‖ or searching dataset to find statistically significant pattern o Time-period bias: research findings that are sensitive to selection of starting and/or ending dates, may bias out-of-time period analysis  Failure to account for conditioning info: analyst should condition forecasts on the state of economy to formulate most accurate expectations (say different betas in expansion economies and recession economies)  Misinterpretation of correlations: o distinguish b/w exogenous and endogenous variables; o correlation may be spurious w/ no predictive relationship  test w/ multiple regression variable significance  test using time series analysis w/ independent variables including lagged value of dependent variable, lagged value of tested variable and lagged value of control variables  Psychological traps: o anchoring trap: tendency to give disproportionate weight to first info received on topic o status quo trap: tendency to perpetuate recent forecasts—to predict no change o confirming evidence trap: bias that leads individuals to give greater weight to info that supports existing or preferred point of view  examine all evidence w/ rigor  enlist an independent-minded person to argue against  be honest about motives o overconfidence trap: tendency to overestimate accuracy of forecasts o prudence trap: tendency to temper forecasts so that they don’t appear extreme; to be overly cautious in forecasting o recallability trap: tendency of forecasts to be overly influenced by events that have left strong impression on person’s memory o model uncertainty: uncertainty whether selected model is correct  input uncertainty: uncertainty whether inputs are correct Tools for Formulating Capital Market Expectations:  Formal tools: established research methods amenable to precise definition and independent replication of results o Statistical methods: descriptive statistics; inferential statistics  historical statistical approach: sample estimators (assuming stationarity)  sample arithmetic mean total return or sample geometric mean total return as estimate of expected return  sample variance as estimate of variance  sample correlations as estimate of correlations  Shrinkage estimation: taking weighted average of historical estimate of parameter and some other parameter estimate based on analyst’s belief of weights  target covariance matrix: selecting an alternative estimator of covariance matrix  Time-Series Estimators: forecasting a variable based on lagged variables  good for short-term forecasts  volatility clustering: tendency for large (small) swings in prices to be followed by large (small) swings of random direction 2 2 2 o  t   t1 1   t ; beta is the rate of decay of the influence of the value of volatility in one period on future volatility; epsilon is random noise  Multifactor Models:  useful for estimating covariances: o estimates of covariances b/w asset returns can be derived using assets’ factor sensitivities o may filter out noise o make it relatively easy to verify consistency of covariance matrix  factor covariance matrix: cross tabulations showing covariances / variances of factors o Discounted Cash Flow Models:  equity markets:  Gordon (constant) growth model o E(R )  D 1  g
More Less

Related notes for MIS 4500

Log In


Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.